Chemical engineering education ( Journal Site )

Material Information

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


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


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

Record Information

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

Full Text

Ipg. 14}


I arma

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

Tim Anderson

Phillip C. Wankat

Carole Yocum

James 0. Wilkes, U. Michigan

William J. Koros, University of Texas, Austin


E. Dendy Sloan, Jr.
Colorado School of Mines

Pablo Debenedetti
Princeton University
Dianne Dorland
University of Minnesota, Duluth
Thomas F. Edgar
University of Texas at Austin
Richard M. Felder
North Carolina State University
Bruce A. Finlayson
University of Washington
H. Scott Fogler
University of Michigan
William J. Koros
University of Texas at Austin
David F. Ollis
North Carolina State University
Ronald W. Rousseau
Georgia Institute of Technology
Stanley L Sandler
University of Delaware
Richard C. Seagrave
Iowa State University
M. Sami Selim
Colorado School of Mines
Stewart Slater
Rowan University
James E. Stice
University of Texas at Austin
Donald R. Woods
McMaster University

Winter 2001

Chemical Engineering Education

Volume 35 Number 1 Winter 2001

2 Dianne Dorland, D.A. Barsotti

8 University of Melbourne, David Shallcross

14 Combustion Synthesis of Advanced Materials, Arvind Varma

22 Use of the Residue Theorem to Invert Laplace Transforms, N.W. Loney

25 Truth in Advertising, Richard M. Felder

26 Analysis and Simulation of a Solar-Powered Refrigeration Cycle,
Jude T. Sommerfeld
46 Dynamics of a Stirred-Tank Heater: Intuition and Analysis,
J.A. Romagnoli, A. Palazoglu, S. Whitaker
62 Student Motivation, Attitude, and Approach to Learning: Notes from a Novice
Teacher, Eduardo Vivaldo-Lima
68 The Pitzer-Lee-Kesler-Teja (PLKT) Strategy and Its Implementation by Meta-
Computing Software. William R. Smith, Martin Lisal, Ronald W. Missen
80 A Choose/Focus/Analyze Exercise in ChE Undergraduate Courses
G.K. Sureshkumar

32 The Effect of Publication Rate Profile on Citation Statistics,
Mordechai Shacham, Neima Brauner

36 Experiments to Demonstrate Chemical Process Safety Principles,
Brian D. Dorathy, Jamisue A. Mooers, Matthew M. Warren, Jennifer L. Mich,
David W. Murhammer

50 An Analysis of Enrollment Cycling in ChE, R. Russell Rhinehart

58 Cooperative Education: A Key Link Between Industry and Engineers in the
Making, Tanya Bradburn

74 Electrochemical Engineering in the Process Laboratory Course, Jan B. Talbot

31 Note from Octave Levenspiel
45 In Memoriam: Sami Selim
73 Positions Available

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 2001 by the Chemical Engineering Division, American
Society for Engineering Education. The statements and opinions expressed in this periodical are those of the writers and
not necessarily those of the ChE Division, ASEE, which body assumes no responsibility for them. Defective copies replaced
if notified within 120 days of publication. Write for information on subscription costs and for back copy costs and
availability. POSTMASTER: Send address changes to CEE, Chemical Engineering Department., University of Florida,
Gainesville, FL 32611-6005. Periodicals Postage Paid at Gainesville, Florida.

a l=1educator



Firmly Ancuhored For Now

Three phases of Dianne land hair
styles): as a young girl in South Dakota:
as a graduate student at liest I irginia
D.A. BARSOTTI Universit: and. todai. as Dean of
Rfon an's College of Engineering.
n her undergraduate days at South -
Dakota School of Mines and Tech-
nology, Dianne Dorland worked for
the Institute of Atmospheric Sciences,
first as the office "go-for" then as an
undergraduate technician. She learned
plenty about clouds, but even as she
soared above, beyond, and through the
nebulous matter-seeding clouds, col-
lecting data, or studying nucleation in
cloud systems-no one would quip that
Dianne had her "head in the clouds."
Today Dianne Dorland is Dean of
Engineering at Rowan University,
Glassboro, New Jersey. She is still soar-
ing, still seeding, still seeking-but ex-
periences have sharpened her vision and
heightened her dreams. Today, as dean
of Rowan's innovative new engineer-
ing program, she is excited about the opportunln\ to develop and maintain Isrong
links between engineering education program and Indutr\ in a w a that \ Ill benefit
both the students and the region where she has jus recently landed.
From drawing weather maps to charting ne trerrnori in chemical engineering
education, Dianne has accumulated an enormous \\ejlth of experience. both in her
professional endeavors and in her personal undertakings The umn ot those experi-
ences is obvious from the very first time you meet her

When you first shake hands with Dianne Dorland. \ ou know right a% aj that she is
a substantial person, says Gary Finley, a member of the Arrrow\ head Chapter of the
Copyright ChE Division of ASEE 2001

Chemical Engineering Education

National Society of Professional Engineers in Duluth, Min-
nesota, where Dianne was an active participant. 'She's got a
good grip." Finley, who is technically retired from the engi-
neering profession but consults and keeps active in his pro-
fession, had a chance to interact with Dianne when she
headed the Department of Chemical
Engineering at the University of Min- Fronr
nesota, Duluth (UMD). "Dianne is a
professional engineer and a profes- territory i
sional educator," he says. has accumul

But it is that first impression that
seems to best capsulize what others
find in Dianne.


Alan Nelson is a PhD candidate at Michigan Technologi-
cal University. He had the opportunity to do his undergradu-
ate studies under Professor Dorland's watch. "When I ar-
rived for the first day of class," he recalls, I quietly selected
a seat in the back of the room to avoid unnecessary attention.
I pulled my books out of my pack and adjusted my hat is
such a way as to prevent eye contact with the instructor."
Instead of being an unknown in the back of the room, Nelson
found he had captured the attention of the instructor. "Little
did I know that she was not particularly fond of hats worn
indoors, let alone baseball caps," he says. "She politely
asked me to remove my hat," recalling that he did so with
reservation. "I sat through the class with my hair protruding
out in every direction," he says, but he quickly realized that
it didn't matter because the instructor had a genuine interest
in her students. "Dr. Dorland involved the students in a
discussion-to learn about them and about their expecta-
tions for the course,' he remembers. "By the end of the class,
the instructor had earned my respect; a respect that continues
to grow even today."
"Dianne puts her students first," says Ron Visness. Be-
fore, and even after, he retired from his position as manager
of a research program for the State of Minnesota Minerals
Division, Visness frequently worked with Dianne in her
capacity as head of UMD's Chemical Engineering Depart-
ment. Even though Dianne's calendar was full of adminis-
trative tasks, he says, she made sure she had time for the
students. "She is also industry oriented and has a good feel
for what industry wants and needs," he comments, adding
that students find that perspective valuable. "Dianne has
been where they are going. She knows what happens when
things go wrong in a chemical plant-and she knows the
working end of a pipe wrench."

Business incubator. Technology center. Industrial outreach.
Student interaction. As dean of Rowan's College of Engi-
neering, Dianne is melding the experiences she's had in her
career to write the next exciting chapter on chemical engi-
neering education. From her office in a state-of-the-art engi-
Winter 2001

neering building, she brings a unique vantage point to the
chore. Her experiences will allow her to assess the needs of
each of the partners-students, industry, and the commu-
nity-that will be affected by Rowan's innovative new engi-
neering endeavor. "I can't promise to deliver everything that

Drawing weather maps to charting new
chemical engineering education, Dianne
lated an enormous wealth of experience,
oth in her professional endeavors and in
her personal undertakings.

everybody wants," Dianne says, "but I have to find out what
each of them wants, figure out what we can provide, and
then determine what's optimal for us and what's optimal
for our relationship."
Several factors attracted Dianne to Rowan University. The
engineering program, whose first class of engineers gradu-
ated in May 2000, was established with strong industrial
interaction. This interaction is the basis for the clinic projects,
the "hands-on, minds-on" clinic experiences that offer pro-
gressively more involved engineering situations for all four
years of the program. In recruiting the faculty, the college
sought forward-thinking candidates who had strong back-
grounds and the potential to develop relationships within
industry. When Founding Dean James Tracey set the course
for Rowan's College of Engineering, he was laying a foun-
dation for the kind of engineers who would meet the chal-
lenges of the new millennium.
"This is a new educational experience," Dianne admits. "It
incorporates interaction between students in the different
disciplines of engineering, between students in different
class levels, between students in different industrial
projects, between students and faculty. At Rowan, the
first thing that freshman come to understand is that engi-
neering is a state of mind."
And that seems to be the state of mind where Dianne has
taken up permanent residence. Though she has been settled
in the dean's office for just a short time, her presence
fills the room. .her hard hat, a congratulatory vase of
long-stemmed red roses from a dear friend, and a calen-
dar full of opportunity.

Opportunities were not as apparent when Dianne was grow-
ing up in her South Dakota hometown of Belle Fourche.
Nestled on the northern side of the Black Hills, the rural
community of 4000 offered a solid education for its youth,
but not many options. "There were few job prospects," Dianne
explains. "I could become a rancher's wife or go to college."
Because of her aptitude in science and math, and her expo-

sure to engineering through her participation in the Junior
Engineering Technology Society during those high school
years, it was natural that Dianne chose to enroll in the
engineering program at the South Dakota School of Mines
and Technology. A strong recruiting effort for chemical
engineers set the direction as Dianne began her under-
graduate studies.
Dianne's youngest sister, Thais M'Annette Dorland
Armstrong, recalls the beginning of Dianne's educational
experiences. "I remember
that when she went to col-
lege," she says, "there were
so few women in attendance
that she was forced to live
off campus because there
was no housing for female
students." 1
"Engineering offered me
a challenge," Dianne says.
"I wasn't sure of the oppor-
tunity at the time, yet once
on board, I knew it was a
good fit for me." Dianne is
a licensed professional en- In her work with AS, Di
gineer in the states of West Here she is picturedfillii
Virginia and Minnesota, Generator foi
and will soon be licensed
in New Jersey.
Dianne soon realized that rigors of the chemical engineer-
ing curriculum provided additional benefits. "I knew I had to
work during my college years," she says, adding that since
she knew there would be no financial support from her
family, she would be paying her own way. Because of the
demands of her studies, she quickly learned valuable skills-
time management, communication skills, perseverance, and
flexibility. Those same skills helped Dianne find and bal-
ance part-time jobs while she carried out her course load.
She had a number of interesting part-time ventures, rang-
ing from bookkeeping to taking in ironing to house painting.
"But the best job I had during that time was at the Institute
of Atmospheric Sciences," she says. That's where she
had a chance to soar into a fair-weather adventure that
has endured, trying her hand at everything from drawing
weather maps to operating an instrument data-collection
package at 10,000 feet.
Because of her position at the Institute while an under-
graduate, Dianne was able to take some graduate courses in
meteorology. "It was the flexibility of the chemical engi-
neering department that allowed me to use funds from the
Institute to complete my graduate studies," she admits, add-
ing that the Department was willing to combine resources
and curriculum with the Institute to make it possible for her

me s
ng tt

to earn her Masters Degree in chemical engineering. Her
fascination with clouds prompted her to minor in meteorol-
ogy. "I am appreciative of the Department and the faculty,"
Dianne says. "What I owe them goes beyond words." The
program at South Dakota is the model for what Dianne sees
as a successful engineering program.

Much like the collaboration that takes place in a successful
engineering program, so is
there collaboration in the
molding of an individual. And
at this point in her life, Dianne
--- can look upon inherited at-
tributes, life lessons, and
character-building moments
to gain some perspective.
There have been many men-
tors, friends, and family mem-
bers who have made an im-
pact on her, professionally
th and personally.
"On my mother's side,
grandfather was a farmer who
pent time in the clouds. ws fascinated with airplanes,"
e plane's AgI-Acetone Dianne recalls, thinking of
ud seeding, her own interest in the
weather and her passion for flying. "My grandmother Dorland
lived with us, and when I was ten years old, she taught me to
touch type using an instruction book with an 1898 copyright
date," Dianne remembers, perhaps explaining her tendency
to set goals, look ahead, and be prepared.
Dianne's father was an optometrist, and her mother stayed
at home to raise the family. It was her parents' hope that all
of their children would do more with their lives education-
ally. "My mom was the driving force in those efforts,"
Dianne says, relating one instance when her mom went into
school to argue that Dianne be admitted into a drafting class.
Thais recalls Dianne's determination to set out in new
directions. "One of my earliest memories in regards to career
aspirations was when some friends of my parents were visit-
ing. Being all of four years old, I was privy to their conversa-
tion, which turned to the subject of careers. I think the
conversation sticks out in my mind because of the distinct
shock on their faces when I piped in that I planned to get my
masters degree in chemical engineering when I grew up. Of
course, my aspirations and awareness were completely formed
because of my older sister, Dianne."
Thais' relationship with Dianne has bloomed from sister-
hood to best friend. "I admire Dianne for her strength, her
ability to focus, her sense of clarity, and her desire to have
fun in the process." Thais, who is a full-time artist,
birdwatcher, and world traveler, says she is constantly aware

Chemical Engineering Education

of the full emotional support of her oldest sister.
Those moments of family love weren't always as
obvious as the Dorland siblings were growing up. The
second eldest of six children, Dianne found plenty
of opportunity to practice conflict resolution, learn-
ing when it was wise to let go and when it was time
to give in, when it was good to win and when it was
okay to lose.
Dianne's mother had quite an influence on her. Her
mother read to the family and provided games for them
to play. "We didn't have a TV in our house," Dianne
says, adding that they did live next door to the li-
brary-"my second home."
"My mother had an adventuresome spirit," Dianne
says. "I remember one time when she took a cigar box
full of silver dollars, loaded us into the car, put the
cigar box under the front seat, and headed out to have
an adventure. When the box was half empty, we started
the journey back home." Her mother arranged many
such journeys... to visit cousins, to go exploring, or to
drive toward a sunrise. "She planted that spirit of ad-
venture in me," Dianne professes. "I learned to be to
be spontaneous and flexible."

It wasn't long before Dianne had a chance to test her
flexibility. Following an on-campus interview with
Union Carbide in 1969, Dianne left South Dakota to
Winter 2001

4 Dianne, an
enthusiastic scuba
diver, shown here
on a dive in the

SDianne is shown 1
here just before
firing a blast at a
taconite mine in

join the company's R&D department in South Charleston, West
Virginia. Unfortunately, her job evaporated before she even ar-
rived, due to a shift in the company's economic situation. "I
floated between several different departments," Dianne remem-
bers. "It was fascinating. I was exposed to a myriad of projects,
and I began to see how flexible an engineer must be to survive on
the job." For a while, Dianne worked on a process for insulating
the hulls of ships used to transport liquefied gases. She learned
about problems with certain insulating materials and became aware
of the health and safety issues that exist in the chemical industry.
After several other projects at Union Carbide, Dianne had a
chance to move from textbook learning into the real world. "I
worked on assessment of a butane oxidation process facility," she
explains. "This was the nasty core application of everything you
learn in undergraduate school. It was no longer a simple problem,
but a multi-component one-with all the by-products that occur in
the commercial world."
Another dose of reality hit when a company lay-off left Dianne
out of work. Part of the lay-off targeted female spouses. Dianne,
who, by then, had married a co-worker, was affected. Not wanting

to dwell on this, she summoned her energies and worked to
get another job. She found success at a DuPont facility in
Belle, West Virginia. She was hired as a process engineer,
and provided technical support for operations in the plant. "I
worked with Para Amino Cyclohexyl Methane (PACM), the
chemical precursor to the fiber Qiana Nylon," she says. "It
was an exciting time to be working at DuPont. The business
was expanding and I got to work on the design and expan-
sion of the process and to offer technical support during the
start-up of the new equipment." Dianne was also responsible
for providing quality control and dealing with the technical
issues that were involved in the process.
Fortunately, there weren't too many issues with the fact
that Dianne was the first full-time female chemical engineer
at the site. "There was a period of adjustment for me and the
plant," Dianne recalls, "but my supervisor, Dick Sherman,
was more interested in my engineering talents than in my
gender." A genuine bond developed between Dianne and her
colleagues. A few years later, when Dianne took maternity
leave for the birth of her first child, the plant operators found
a special baby shower present for her. "They gave me a
gift-prophylactics, with the admonition that I learn how
to use them," Dianne says with a grin. "It was quite a
compliment to know that they wanted me to come back
to work-and not leave again."

In 1975, Dianne resigned from DuPont and gave birth to
her son Brad. Two years later, her daughter Decker was
born. During their early years, Dianne stayed at home with
Brad, who is now a PhD candidate in cell biology at Oregon
State University, and Decker, who recently landed a full-
time position as Program Assistant for the Greater Minne-
apolis Metro Housing Corporation.
While she was at home with her children, Dianne took an
active interest in the community. She also found a part-time
position in chemical equipment sales. In 1981, a teaching
opportunity at West Virginia Institute of Technology be-
came available and Dianne was hired to teach evening classes.
She was eventually offered a full-time position as assistant
professor. At some point during that time frame, Dianne
realized that she wanted to pursue her PhD, and in the spring
of '83, she headed to West Virginia University.
A grant for non-West Virginia undergraduates that was
available through the Department of Energy took Dianne
down new paths. She worked with Al Stiller on a novel
method for processing coal. "I took note of Al's interactions
with government agencies and of the political involvement
in funding," Dianne says. Not only did she find the process
of coal extraction quite interesting, but she also became
aware of the intersection between academics and politics.
"I attended graduate school with Dianne from 1983-1985,"
says David Bernemann (now Engineering and Math Instruc-

tor at North Iowa Area Community College) "and have
maintained my friendship with her since that time. I think of
Dianne as one of the smartest people I know." Bernemann
acknowledges that Dianne has reached a high mark aca-
demically, but he is more impressed that she knows what is
really happening in the world.
Phillip Kneisl is another of Dianne's friends from graduate
school. He is currently a Senior Chemical Engineer at the
Schlumberger Reservoir Completions Center in Rosharon,
Texas. From his experiences, he feels that a strong academic
background is only part of the formula for a good engineer.
"Dianne knows her technical limits," Kneisl says. "She is on
good terms with everyone. She practices good politics. And
she is always positive." Bernemann continues, "Dianne seems
to be extremely effective at facilitating other people to do
their best. This means that she is able to provide those
around her with the resources to accomplish their goals. I
think she gains personal satisfaction in assisting others with
their accomplishments."

As friends, Bernemann and Kneisl shared in some of
Dianne's non-academic pursuits. Kneisl remembers that
Dianne's house was always the center of non-academic ac-
tivities for chemical engineering graduate students in
Morgantown. "There were many memorable parties, week-
end brunches, lunches and dinners-and general bull ses-
sions," he says.
Bernemann reveals another side of Dianne. "When we
were in grad school, Dianne had a private pilot's license and
her own airplane-a Cessna 172," he says. Bernemann was
one of the grad students that went flying with her, and he
recalls one instance that showed Dianne's ingenuity. "The
carburetor heat cable on Dianne's plane broke. (She did
many of her own repairs on the plane.) She was going to buy
a new cable but found out it cost something like $120,"
Beremann says. To Dianne, the cable looked like a lawn
mower throttle cable. "So she decided to replace it with a
lawn mower throttle cable," he continues. "She asked me to
help her with the repair. First we went to a local auto parts
store and bought the lawn mower cable, then we went to the
airport to make the repair." Bernemann admits that a Cessna
172 is a pretty simple machine, so the repair was pretty
straightforward. He knew that the lawn mower cable was the
same as the carburetor heat cable-without the large markup.
What surprised him was Dianne's insistence that since he
helped with the repair, he should take the first flight with
her. Bernemann was not surprised at Dianne's love of ad-
venture, but he found it impressive that she wanted to en-
courage that sense of adventure in others.
Dianne completed her PhD in chemical engineering in the
fall of '85. It had been a fairly tough haul, including an
emphasis in environmental engineering. She felt she needed
Chemical Engineering Education

an escape. Leaving her Cessna behind, she turned to her
scuba diving and discovered a passion for the underwater
world. She spent some time diving in the Red Sea and then
toured the exotic lands of the Near East.

When Dianne returned, she worked for the Department of
Energy while she looked for an academic position. In 1986,
she heard of a new program being initiated at the University
of Minnesota, Duluth (UMD). She remembers thinking that

her varied background could be an asset to
this new program. At UMD, Dianne was able
to both teach and initially work as an envi-
ronmental engineer for Sea Grant, an exten-
sion program at the university.
In the following years, Dianne worked her
way through the ranks at UMD, becoming a full
professor and head of the chemical engineering
program. She had an immediate impact on the
new department. Linda Deneen (currently Di-
rector of Information technology at UMD) first
met Dianne when she became the head of the
Department of Chemical Engineering. She adds
her observations about one of her closest friends:
"Dianne is a professional collaborator and team
builder. She's the one who brings groups of
researchers together and paves the way for them
to be successful as a team. She is also an excel-
lent teacher."
Dianne left UMD with a very strong Chemi-
cal Engineering Department, one that was ranked
second in US News and World Report this year.



paves he
for diem

Dianne is proud of the program's success. "I credit the won-
derful people who gravitated to our program," she says.
Richard Davis, Associate Professor in the UMD Depart-
ment of Chemical Engineering, comments that much of that
credit reflects Dianne's goals for engineering education. "She
was able to see the big picture," he says, noting that Dianne
wanted to move the department forward as a group of fac-
ulty-instead of as individuals. "She constantly looked for
ways to share her contacts, research projects, funding oppor-
tunities, and professional development experiences."
Gary Finley cites more evidence of Dianne's positive im-
pact on the students and the region: "In the past five years,
95% of UMD's chemical engineering students took and
passed the fundamentals test. When the national average is
70%, it speaks volumes about Dianne's effectiveness in
challenging those students."
"Dianne brings out the professionalism in her students,"
Finley observes. "Industries in the region are tickled with
her students. They come in ready to go to work. Twenty-five
percent never leave the Arrowhead, and twenty-five percent
Winter 2001

never leave Minnesota. The local folks appreciate that."
That local impact, and that professional quality in Dianne
and in her students, made her an impressive candidate for
Rowan's Dean of Engineering. According to Ralph Dusseau
(Professor and Chair of Civil Engineering), Dianne was
highly regarded by the Search Committee, by Rowan Uni-
versity President Donald Farish, and by Henry Rowan, the
man who made the school of engineering a reality with the
establishment of a $100 million endowment. "Dianne defi-
nitely had the best industry background and the greatest
potential for establishing effective industry contacts for the
College of Engineering," continues Dusseau. "To

use a sports analogy, Dianne hit a grand slam."
fThat professional quality hasn't gone unno-
ticed by Ted Schoen, Dean of Rowan's Busi-
0- 1 3 ness School. "Dianne is really willing to reach
M... out, to make sure that our ideas for a proposed
business incubator are right for our university,"
he says. He has been traveling around the re-
Sgion with Dianne, visiting other incubators to
i d learn how they operate. "Dianne has a way to
isfy get us thinking about the issues," he says. "She
.tO is willing to share her expertise, she has a lot of
--- ~energy, and she has a great sense of humor."
S-: Dianne brings a wide range of expertise to her
-)- new position. An array of published works and
presentations illustrates her commitment to the
-environment (pollution prevention and indus-
trial waste management), her dedication to the
education of twentieth-century engineers (cur-
riculum topics and student outreach propos-
als), and her knowledge of the role of engineers in indus-
try. Her affiliation with professional societies like the
AIChE, ASEE, and SWE is underscored by her promi-
nent and active leadership positions.
"This is the fastest growing region in the state," says
Stewart Slater (Professor and Chair of Chemical Engineer-
ing at Rowan). He surmises that as Dean, Dianne will be
able to have an important impact on the advancement of
industry in the area. But it is Dianne's mission in chemical
engineering education that really interests Slater and the
other members of the engineering faculty. They feel that her
focus on students, on the links with industry, and on the
intersection of the classroom and the real world will advance
chemical engineering education at Rowan University.
Dianne Dorland has come to the Garden State. She zips
around her eight-acre rural tract on her little green garden
tractor, wearing a wide-brimmed garden hat and earplugs.
She tends to the blueberries and studies nature at the edge of
her pond. She plants the seeds and waits for the harvest. The
life is bountiful and rewarding. 0


ChE at the




University of Melbourne Melbourne 3010, Australia
he University of Melbourne has a reputation as being
Australia's leading research university, attracting more com-
petitive research funds than any other university in the country.
Likewise, the Chemical Engineering Department at Melbourne is recog-
nized as one of Australia's leading departments in terms of both its
research and its teaching.
The city of Melbourne, with a population of some three million people
and located in the southeastern corner of Australia, was founded in 1835
by settlers who moved there from the first settlement in Sydney. It
prospered when gold was discovered some 110 km to the west in the
1850s, and by 1861 it had become the largest city in the country as a
result of the influx of immigrants eager to try their luck in the goldfields.
In 1901, Melbourne became the temporary capital of the newly indepen-
dent country of Australia and it soon became the manufacturing base of
Australia. Following World War II, European immigrants brought di-
verse cultures to the city, and in 1956, it became known around the world
as the home of the Olympics. Decades later, a second wave of immi-
grants, mainly from neighboring Asian countries, again enhanced the
culture of the city. Home to major sporting events and superb restaurants,
Melbourne has repeatedly won accolades as being the "World's Most
Livable City."
The University of Melbourne became Australia's second university
when it was founded in 1853. The University's motto, "Postera crescam
laude" is often rendered as "Growing in the esteem of future genera-
tions." Its mission is to make the University of Melbourne one of the
finest universities in the world. It is located just one kilometer north of
the downtown area and is a broad-based university with faculties of
architecture and planning, arts, economics and commerce, education,

rrojessor Uavia uoger s work wizt
non-Newtonian fluids has defined a
new class of fluid behavior.

engineering, law, medicine, music, science, and
veterinary science. The university has over 34,000
enrolled students, including over 9,000 graduate
students, and over 2,100 academic staff members.
It is a founding member of the Universitas 21
international consortium of universities.
The first engineering degree was awarded in
1883. Since then, the University has played a lead-
ing role in the education of generations of profes-
sional engineers in Australia. In more recent times,
the University has established a reputation for ex-
cellence in the Asia-Pacific region, attracting in-
creasing numbers of international students to its
various engineering disciplines. The university of-
fers degree programs in chemical engineering, civil
engineering, computer engineering, computer sci-
ence, electrical engineering, electronic engineer-
Copyright ChE Division of ASEE 2001
Chemical Engineering Education

ing, environmental engineering, geomatics, manufacturing
engineering, mechanical engineering, mechatronics, and soft-
ware engineering.
The Chemical Engineering Department is one of the larg-
est in Australia, with more than 600 undergraduate students
currently enrolled in its courses. In addition, more than 80
students are presently enrolled in its postgraduate programs.
Women represent nearly half of the undergraduate popula-
tion, one of the highest proportions of any engineering de-
partment in Australia.

Flexibility is the key word that describes the undergradu-
ate chemical engineering program at Melbourne. It is this
flexibility that allows the majority of the students in the
Department to enroll in not one, but two, undergraduate
degree programs.
Until the late 1980s, the four-year structure
of the chemical engineering degree was very reputa
rigid. The subjects and the sequence in which
they were to be studied were prescribed. Stu- reg
dents studied the basic sciences of mathemat-
ics, chemistry, and physics. In their first year
they also received general engineering educa-
tion across the other major engineering disci-
plines, and over the last three years of the course they under-
took studies in chemical engineering science, practice, and
design. They also studied process economics, management,
and engineering law.
Occasionally, permission was given for students to take
one or two non-engineering subjects, allowing them to pur-
sue some other interest. These subjects were taken either in
addition to their normal study loads (i.e., as an overload) or
in lieu of certain specified subjects. If the student wished to
complete two degrees, then there was no alternative but to
pursue them consecutively. Usually some credit was given
toward the second degree for work performed in the first
degree. For example, a student who had completed a three-
year Bachelor of Science (BSc) degree might be given one
year of credit toward a four-year Engineering (BE) degree.
This would reduce the total time taken to obtain both under-
graduate degrees from seven years to six years.
Beginning in the late 1980s, students were permitted to
pursue two degrees simultaneously. By taking extra subjects
in each semester and by spreading the content of both de-
grees over the entire time, it became possible to achieve a
reduction in the time required to complete the requirements
of both degrees. For example, instead of taking seven years
to complete studies toward undergraduate degrees in engi-
neering and commerce, it became possible to gain both
degrees in only five years. The workload required to accom-
plish this feat, however, was very heavy, with overloads
Winter 2001

required in every semester. Only students with demonstrated
abilities were permitted to enroll in concurrent degrees.
In 1990, the first combined-degree program involving
chemical engineering was introduced when the combined
Bachelor of Engineering/Bachelor of Science (BE/BSc) pro-
gram was offered. Its structure made it possible for a student
to meet the requirements of both degrees in just five years.
The reduction in the time required to complete two degrees
is achieved by the existence of material common to both
degrees, e.g., chemistry and mathematics. It should be noted,
however, that it is not possible to obtain a BSc degree by
simply studying for an additional year after completing the
BE degree. Study for the two degrees must be integrated
from the moment the student enters the University.
The Bachelor of Engineering/Bachelor of Arts (BE/BA)
combined degree program was introduced in 1992, allowing
students to pursue an interest in a language, history, or other

... the University has established a
tion for excellence in the Asia-Pacific
lion, attracting increasing numbers of
international students to its various
engineering disciplines.

arts major while undertaking the engineering degree. Like
the Bachelor of Engineering/Bachelor of Commerce (BE/
BCom) degree introduced later, the BE/BA degree requires
five years of study. The combined Engineering and Law
(BE/LLB) degree requires six years of full-time study.
Since their introduction, the popularity of the combined-
degree programs has increased to the point where students
enrolled in them now make up the majority of chemical
engineering students. Prior to 1990, about three students
each year began studying concurrent degrees, and the num-
ber of students entering the chemical engineering under-
graduate program was steady at between 60 and 70 students.
Following the introduction of the first combined-degree pro-
grams, enrollments began to increase, initially at the expense
of the single-degree program. By 1994, the number enrolling
in the chemical engineering programs had almost doubled,
peaking at over 140.
While such combined-degree programs are available at
most Australian universities, the University of Melbourne
has the highest proportion of combined-degree students in
the country. Currently, 70% of the students in the Depart-
ment are enrolled in combined degrees. Since its introduc-
tion, the BE/BSc program has been the most popular of all
combined-degree programs. Not surprisingly, chemistry has
proven to be the most popular science major among the
students, as many see it as the natural complement to their
chemical engineering studies. After chemistry, the next most

popular science majors are the biological-based sciences,
biochemistry, and microbiology. In more recent years,
the study of pharmacology and genetics has become popu-
lar with the BE/BSc students. Other science disciplines
that combined-degree students have enrolled in include
biology, botany, computer science, geology, psychology,
and zoology.
Of the BE/BA students enrolling over the same period,
more than 80% have undertaken a language as their Arts
major. Other majors include criminology, fine arts, history,
linguistics, politics, psychology, and women's studies. The
diverse nature of studies chosen by the students is indicative
of the flexibility of these programs.
BE/BCom students are able to enroll in commerce sub-
jects ranging from microeconomics to accounting and from
personnel management to international trade. BE/LLB stu-
dents often choose to complement their chemical engineer-
ing studies with subjects such as corporate law, corporate
governance, and international law.
In a recent survey of the future directions for chemical
engineering education, James Wei suggested that it was time
for chemical engineering educators to seek a new paradigm.
He suggested several possibilities that would empower engi-
neering graduates to meet the challenges of the new century.
Should chemical engineering become more oriented toward
perceived societal needs such as environmental protection,
manufacturing efficiency, and sustainability? Should it move
to embrace developing disciplines as exemplified by infor-
mation technology, nanomaterials, and tissue engineering?
Should the educational processes focus more on people,
teamwork, leadership, and communication skills? Should it
broaden to more hybrid degrees of financial engineering?
Or should chemical engineering education focus not so
much on the design of new processes, but more on the
development of new products?
As in any industry, chemical engineering educators must
also consider the demands of their clients, namely their
students, the processing industries, and the chemical engi-
neering profession. In the last two decades of the 20th cen-
tury, the petrochemical industries, traditional employers of
chemical engineering graduates, cut back their graduate re-
cruitment programs. In order to find employment, graduates
began to consider the opportunities in non-traditional indus-
tries such as food, finance, and pharmaceuticals. Can the
search for Wei's new paradigm and the changing require-
ments of the clients be answered in a single development?
At the University of Melbourne, we believe that our exten-
sive combined-degree program is possibly the new para-
digm. While no single engineering program can produce
chemical engineering graduates at home in all the emerging
areas from tissue engineering to intelligent processes, the
chemical engineering graduates take with them into the work

place understanding and expertise in a range of disciplines.
This is the Melbourne paradigm.
The Department also has an excellent record in teaching.
Every year the Graduate Council of Australia surveys gradu-
ates from all universities across all technical and non-techni-
cal disciplines. Using the responses to twenty-four ques-
tions, each teaching department in the country is given a
score for several different categories. The survey of 1998
graduates showed that Melbourne's Chemical Engineering
Department was one of the top departments in its discipline.
It ranked second among the ten Australian chemical engi-
neering departments on the Good Teaching and Clear Goal
scales and in terms of overall satisfaction with the course.
International exchange is also a feature of the undergradu-
ate program. As a member of the Universitas 21 consortium,
the Department has a number of undergraduate exchange
programs in place with leading universities around the world.
These programs generally permit our students to spend up to
twelve months at a foreign institution, usually in their third
year. Each year the Department also has a number of study-
abroad students from around the world who come to the
university for a single semester. These students are always

The Department is one of Australia's leading chemical
engineering departments in terms of research. In 1998 it had
the highest competitive-grant research income of any Aus-
tralian chemical engineering department (with an income of
more than $217,000 per full-time equivalent staff member)
and published more papers per government-funded staff mem-
ber than any other Australian chemical engineering depart-
ment. The research activities of the Department are diverse
and are focused principally around the three research centers
The G.K. Williams Cooperative Research Centre for Extrac-
tive Metallurgy
The Particulate Fluid Processing Centre, a Special Research
Centre of the Australian Research Council
The Cooperative Research Centre for Bio Products
The centers are at the forefront of international research
and are the outward face of the Department's research ex-
pertise. Supporting the centres is a fundamental research
infrastructure comprising: non-Newtonian fluid mechanics,
high-temperature thermodynamics, separation processes, and
computational fluid dynamics. Other research areas of sig-
nificance are biochemical engineering, environmental engi-
neering, fluoride process engineering, development tech-
nologies, pulp and paper, and, more recently, polymer sci-
ence and mineral processing.
The G.K. Williams CRC for Extractive Metallurgy creates
new, technically competitive advantages for the Australian
smelting industry based on enhanced smelting proficiency.

Chemical Engineering Education

van Deventer
van Jaarsveld
with a

Geoff Stevens and
recent graduate
Dr. Brenda Hutton
setting up an
with supercritical
carbon dioxide.

It was established in 1991 as a joint venture between the
Commonwealth Scientific Industrial Research Organization
(CSIRO) and the University of Melbourne. Research is fo-
cused on high-temperature processes involving both funda-
mental and applied research, with world-class research capa-
bility in measurement and modeling of thermodynamic and
transport properties of solids and melts at high temperature,
physical modeling of high-temperature operations, flow vi-
sualization, and laser flow diagnostics and computational
fluid dynamics of multi-phase complex flows. Many re-
search achievements have resulted in improved technical
and economic performance for industry, including a licens-
ing agreement with major international companies to com-
mercialize a new patented concept and design of composite
refractory cooling systems.
The Particulate Fluids Processing Centre (PFPC) develops
key science for the processing of particulate fluids of all
kinds, concentrating on systems involving solid and liquid
particles where the dispersed phase is colloidal in nature.
Through the coordination of proven international strengths
in surface chemistry, continuum mechanics, and non-
Newtonian fluid mechanics, the PFPC effectively solves
particulate fluid processing problems experienced by the agri-
cultural, chemical, food, inkjet printing, mineral, water treat-
ment, waste management, ceramic, and pigment industries.

Winter 2001

The Cooperative Research Centre (CRC) for Bio Products
is a collaborative venture with three participants: The Uni-
versity of Melbourne (Botany School and Chemical Engi-
neering Department), CSIRO, and industry. Professor David
Boger leads the Department's participation and heads the
Fundamental Testing and Hydrocolloids node of the CRC.
The main goal is to "establish the science and technology
underpinning the manufacture of plant biopolymers for the
food and other industries." Research of the group is focused
toward understanding the fundamental structure-function re-
lationships of biological molecules directed at applications.
The key fundamental research areas are understanding the
adsorption to and stabilization of the oil-water interface us-
ing biopolymers and the rheology and gel behavior of these
systems. Fundamental programs include theological charac-
terization of novel gelling biopolymers for new material

)r. Andrea O'Connor's work with
mesoporous silica has potential
applications in the Australian
wine industry.

Chemical Engineering Department, University of Melbourne
Melbourne, Victoria 3010, Australia
Phone: +61 3 8344 6631 Fax: +61 3 8344 4153 Web:

David Boger is Laureate Professor of Chemical Engineering and Director of the Particulate Fluids Processing Centre. He is also a Program Leader in the
Cooperative Research Centre for Bio Products. His research is primarily in non-Newtonian fluid mechanics, with interests ranging from basic polymer and
particulate fluid mechanics to applications in the minerals, coal, oil, food, and polymer industries. The winner of many international awards, he has published
nearly 300 papers and consults widely around the world.
Geoff Stevens is Professor and Head of the Department. He has an international reputation in solvent extraction, interfacial phenomena, and emulsion stability and
is presently Secretary General of the International Solvent Extraction Committee. His research is primarily in the hydrometallurgical field, but also covers aspects
of food, pharmaceutical processing, and environmental or waste-water processing.
Jannie van Deventer is Professor of Mineral and Process Engineering. He is the leader of the Mineral Processing Group. His research interests include diagnostic
leaching of gold ores, extraction of gold by activated carbon, modeling of simultaneous leaching and adsorption processes, image analysis of flotation froth,
transport processes in froth flotation, geopolymerization of waste materials, immobilization of toxic waste, and simulation of ill-defined processes using artificial
David Wood is Dean of the Faculty of Engineering and Professor of Engineering. He is the immediate Past Chair of the Institution of Chemical Engineers in
Australia and a former Vice-President of the Institution of Chemical Engineers (UK). Over the last thirty years his research has ranged widely, but in more recent
years it has focused on fluoride processes. He is presently Chair of the Sixth World Congress of Chemical Engineering, which will be held in Melbourne in
September 2001.

Professorial Fellow
David Solomon is a Professorial Fellow within the Department. He leads the Polymer Science Group and is best known in Australia as the leader of the research
team that developed the plastic bank note, currently in circulation throughout Australia.

0 Readers and Associate Professors
Malcolm Davidson is a Reader within the Department. Trained as an applied mathematician, his research interests include computational fluid dynamics in process
engineering. Much of his current work is associated with the G.K. Williams Cooperative Research Centre for Extractive Metallurgy.
Neil Gray is a Reader within the Department. After spending several years with BHP Central Research at Newcastle, New South Wales, he has continued his main
research interest in the area of physical and mathematical modeling of rate phenomena in metallurgical processes. He is currently Program Leader in the G.K.
Williams Cooperative Research Centre for Extractive Metallurgy. In conjunction with WMC Resources Ltd, a licensing agreement has recently been signed with
major international companies to commercialize a new patented concept and design of composite refractory cooling systems.
Peter Scales is a Reader in the Department. His research activities include understanding the measurement and application of compressional dewatering of solid
materials and the theological and electrokinetic characterization of concentrated suspensions of particles. Unit operations of interest include clarification,
thickening, and filtration and an aim is to be able to predict and optimize these operations from first principles on the basis of laboratory measurements. Interests in
the area of characterization of suspensions include the use of electroacoustics and the failure of concentrated flocculated suspensions of particles in shear.
Neville Pamment is an Associate Professor of Chemical Engineering and leads the Biochemical Engineering activities in the Department. He is a Commissioner of
the International Yeast Commission. His research interests range from ethanol production from lignocellulose using recombinant bacteria to the kinetics and
physiology of product inhibition in microbial fermentations.
David Shallcross is an Associate Professor and Universitas Fellow in the Department. His research interests include ion exchange and enhanced oil recovery. The
author of two books, he is active in the secondary school community, developing teaching material aimed at raising the profile of the engineering profession among
school students. He is also Chair of the Program Committee for the Sixth World Congress of Chemical Engineering and is Associate Dean (International) for the
Faculty of Engineering.

Mike Conner is Senior Lecturer and Deputy Head of the Department, responsible for all undergraduate matters. He is also a Deputy Director of the University's
Office of Environmental Programs. His research interests lie mainly in the areas of environmental engineering and policy and thermochemical biomass conversion.
Geoff Covey is Senior Lecturer and is responsible for supervising the final-year design project. After a career in the pulp and paper industry spanning twenty years,
he continues his research interest in this area. His other areas of interests include process development and economics and pan pelletizers. He is also a member of
the fluoride research group.
David Dunstan is Senior Lecturer and is a physical chemist by training who leads a team in the Cooperative Research Centre for Industrial Plant Biopolymers. His
key research areas are understanding the adsorption to and stabilization of the oil-water interface using biopolymers and the rheology and gel behavior of these
systems. Fundamental programs include theological characterization of novel gelling biopolymers for new material design.
Leong Yeow is a Senior Lecturer and his research interests include non-Newtonian fluid mechanics, inverse problems in rheology, and hydrodynamic stability.
Andrea O'Conner is a Lecturer whose research focuses on surfactant behavior and separation processes, particularly those for the food and pharmaceutical
industries and for waste-water treatment. Mesoporous molecular sieves are synthesized and tailored for selective separations by adsorption via size exclusion and
targeted surface chemistry. The current focus of this work is on the development of these materials for purification of high-value biological molecules.
Sandra Kentish is a Lecturer and joined the Department in early 2000. She has a strong industrial background, with experience in the petrochemical, photographic,
and paper industries, as well as in chemical-hazard management. She is developing research interests in a number of areas, including biopolymer processing,
biofouling, solvent and supercritical extraction, carbon dioxide absorption from flue gases, and the use of ultrasonics in industry. Molecular dynamics simulation is
another area of more fundamental interest.

2 Chemical Engineering Education

design. Rheooptic, time-resolved fluorescence, and light-
scattering measurements aimed at developing understanding
of solution flow behavior are also studied.
Other research groups are also very active within the De-
partment. The Separation Processes Group investigates
rate phenomena involved in separation processes with par-
ticular reference to hydrometallurgy, waste treatment, and
biochemical separations. A major focus is on gaining a bet-
ter understanding of the interplay between diffusion, hydro-
dynamics, and the interfacial reactions occurring in solvent-
extraction systems. In particular, novel techniques have been
developed based on attenuated total internal reflectance spec-
trophotometry and, more recently, atomic force microscopy
to study interfacial phenomena and reactions occurring at
the interface. Mechanistic models for axial dispersion in
pulsed sieve plate columns have been developed. In addi-
tion, the group has been active in liquid membrane pro-
cesses, coalescence processes, diffusion in liquid systems,
and in ion exchange. Biochemical separation processes be-
ing studied involve a range of operating and potential tech-
nologies for the food, pharmaceutical, and water-treatment
industries. These include adsorption using specifically tai-
lored adsorbent materials, electrophoresis, supercritical ex-
traction, and ultrafiltration. Surfactant aggregation phenom-
ena and their applications are also under investigation. Novel
ion exchange processes presently being studied include ion
exchange in radial flow and ion exchange equilibria in dual
exchanger systems. Much of the work is in collaboration
with others in the University, with the CSIRO Division of
Chemical and Polymers, Tsinghua University Beijing, Mas-
sachusetts Institute of Technology, and with industry.
The Mineral Processing Group focuses on reactions at
the fluid-solid interface. Electrochemical and mineralogical
aspects of the extraction of metals from ores are studied at
both laboratory and plant levels. Significant advances have
been made in understanding the kinetics and equilibrium of
the interaction between reacting and dissolving solids and
the presence of adsorbents such as activated carbon and ion
exchange resins. Various features of artificial intelligence
have been integrated with systems of differential equations
to describe the dynamics of operating plants, especially for
cases where fundamental models are lacking. The transfor-
mation of reactive alumino-silicate wastes into useful con-
struction materials is being investigated extensively at both
laboratory and pilot-plant scale. Emphasis is placed on inter-
facial phenomena and the evolution of microstructure during
the formation of geopolymers. This group succeeds in inte-
grating fundamental research with the needs of operating plants.
The Polymer Science Group has interests in a diverse
range of macromolecular related projects that combine the
disciplines of polymer science and chemical engineering.
The span of research includes phenolic resins and their com-
posites; minimal shrinking monomers for specialty applica-
Winter 2001

tions; novel multifunctional monomers for controlled net-
work formation; grafting studies of polyolefins; develop-
ment of living radical polymerization for the generation of
predetermined macromolecular properties; and the determi-
nation of kinetics of propagation.
The Fluoride Process Engineering Group has special
expertise in fluoride chemistry. Its work involves the treat-
ment of minerals with fluoridating agents to produce pure
products. It is also developing processes for the treatment of
carbonaceous materials such as coal and spent pot-lining
from aluminium smelting to produce commercial products.
While some government support has been obtained, the group
almost totally interacts with industry, both large and small,
Australian and overseas. Currently it has three major projects:
a new process for production of titanium oxide pigments; a
process for the production of ultra-clean coal; and a process
for the recovery of spent pot-lining.
The Computational Fluid Dynamics Group investigates
fluid dynamics and transport phenomena in single and
multiphase flows in process engineering. A major theme is
the fundamental study of dispersed multiphase flows and
two-fluid flows with deforming interfaces. Current topics
include the dispersion of solids in metallurgical melts, mol-
ten slag foaming, droplet breakup, heat and mass transfer in
soils, drop impact on solid surfaces and liquid films, slump-
ing of yield stress materials, mass transfer at deforming
interfaces, and self-sustained oscillations of confined jets.
The Biochemical Engineering and Fermentation Tech-
nology Group works on the application of microorganisms
and enzymes to chemical processing. A major focus is on the
use of recombinant bacteria to produce fuel ethanol from
lignocellulosic materials such as wood and straw. The group
also has an international reputation for its fundamental re-
search on the factors affecting product inhibition in fermen-
tations, the focus being most recently on the key role of
acetaldehyde as an inhibitor or stimulant of yeast and bacte-
rial alcohol fermentations, depending on the concentration
of this metabolite.

The Department is one of the strongest research depart-
ments in the University. As its activities and student num-
bers have grown over the last two decades, it has gradually
spread out from its original building built in the 1960s. In
2001, construction will begin on extensions to the existing
building that will allow the Department much needed re-
search space for expansion.
It is the success of the Department's graduates that contin-
ues to illustrate its success in chemical engineering educa-
tion and research. In its ranks of graduates, the Department
boasts three Rhodes Scholars and the current Australian
Chief Scientist. Its future is assured by the continued success
of its graduates. 0



The Union Carbide ChE Divi-
ion Leceship arbid is bes ed University of Notre Dame Notre Dame, IN 46556
sion Lectureship Award is bestowed
annually on an exceptional engi-
neering educator and is designed to since prehistoric time, mankind has used exothermic
recognize and encourage outstand- reactions for its survival-for example, burning of
ing achievements in important fields wood for warmth and preparation of food. The energy
of fundamental chemical engineer- from exothermic reactions has also been used to modify
ing theory or practice. properties of materials. Thus it was discovered more than ten
Arvind Varma is the Arthur J. thousand years ago that heating a piece of clay in fire con-
Schmitt Professor of Chemical En- verts it into a ceramic with very different and useful proper-
gineering at the University of Notre ties. Modern technologists sinter net-shape bodies consoli-
Dame. A native of India, he received his PhD degree from the dated from powders in furnaces to produce, for example,
University of Minnesota in 1972 and remained there for one
University of Minnesota in 1972 and remained there for one ceramic shields to protect spacecraft. In both cases, the basic
year as an assistant professor. He was a senior research engi-
neer with Union Carbide Corporation for two years before principle is the same: application of external heat to rear-
joining the Notre Dame faculty in 1975, where he achieved range chemical bonds and shift the material properties in the
the rank of Professor in 1980 and his current Schmitt Chair desired direction. Since rearrangement of chemical bonds
position in 1988. may release significant energy itself, however, it is attractive
Dr. Varma's research interests are in chemical and catalytic to use this energy directly to produce valuable materials.
reaction engineering, and synthesis of advanced materials. He Such a method was discovered some thirty years ago and is
has published over 200 research papers in these areas and co- called combustion synthesis (CS). Comprehensive reviews
authored three books: Mathematical Methods in Chemical of this field are available in the literature,3" with recent
Engineering (Oxford University Press, 1997), Parametric Sen- achievements summarized in a popular account.[4'
sitivity in Chemical Systems (Cambridge University Press,
1999), and Catalyst Design: Optimal Distribution of Catalyst NATURE OF COMBUSTION SYNTHESIS
in Pellets, Reactors, and Membranes (Cambridge University
in Pellets, 2001)Reactors, and Membranes (Cambridge University Let us introduce this process by considering fine (less than
Press, 2001). 100 micron, about the thickness of a human hair) powders to
He is also the founding Editor (1996-present) of the Cam- two metals: nickel (Ni) and aluminum (A. Taking these
two metals: nickel (Ni) and aluminum (Al). Taking these
bridge Series in Chemical Engineering, a series of textbooks
and monographs published by Cambridge University Press. powders in the appropriate ratio (e.g., Ni/Al about 2 by
weight), we can mix them thoroughly and create an article
Varma served as Department Chair at Notre Dame during with tailored form and shape by using a pressing technique.
the penod from 1982 to 1988. He has served as Visiting
the period from 1982 to 1988. He has served as Visiting Next, one spot on the article surface is heated for a few
Professor at a number of institutions, including the University
of Wisconsin, Caltech, and Princeton. He has received several seconds by, for example, a heated tungsten coil or laser
awards for his teaching and research activities, including the beam. The reaction between Ni and Al starts at the hot point
R.H. Wilhelm Award (1993) of the AIChE. and propagates rapidly (about 10 cm/sec) along the volume
of the article in the form of a bright glowing (combustion)
Copyright ChE Division of ASEE 2001
14 Chemical Engineering Education


- .- _
.. ..I- ",,.a .- = = _-- "-'. -

--: _- .:_ _- .~. -. .
: :: ... = _

wave. In the wave, nickel and aluminum melt and react with each other to form (synthesis)
an intermetallic compound, nickel aluminide (NiAl). Such intermetallics have a number of
attractive characteristics, including low density, high corrosion resistance, high strength
even at high temperature, and relatively low cost. For these reasons, NiAl-based materials
are good candidates for demanding high-temperature applications, such as aircraft tur-
bines and other engine parts.
Some advantages of the CS method over conventional powder metallurgy techniques of
advanced materials production include low energy requirements, short synthesis times (on
the order of a few seconds), high temperatures (2000-4000 K), and high heating rates (up
to 106 K/s), which allow one to produce unique (e.g., metastable) compositions. Also,
owing to intensive volatilization of impurities at the very high temperatures in the reaction
wave, the products of CS are frequently purer than the starting reaction mixture. In
addition, there is essentially no limitation on size of the synthesized item, since the heat is
generated not by an external heating device but by a chemical reaction that proceeds at
every point inside the sample.
These characteristic features make CS different from conventional technologies, which
usually take tens of minutes or hours at temperatures that can be achieved in common
furnaces (usually less than 2000 K), and possess a non-uniformity of temperature (and
hence properties) distribution along large-scale samples. Owing to the fact that CS
technology has the potential to prepare advanced materials and net-shape articles in one
step, and that it has extremely low external energy requirements, it is also well-suited for
use on space platforms.
Efficiency of the method is not the only point of interest regarding this process,
however. Even more interesting is the fact that under the unique conditions of CS, reaction
kinetics, mechanism of reaction, and product microstructure formation become different
from those realized under conventional isothermal or low-heating-rate conditions.
From the viewpoint of chemical nature, three main types of CS processes can be
distinguished. First, gasless combustion synthesis from elements, where all initial reac-
tants, intermediate, and final products remain in condensed (solid or liquid) state. For this
reason, such reactions are sometimes called solid flame. The most popular examples are
reactions of transition metals with carbon and boron, e.g.,
Ta's' + C" = TaC"' + 146 kJ/mol
Ti"' + C'` = TiC"`' + 230 kJ/mol
Nb1' + 2B') = NbB,~' + 175 kJ/mol
The second type, called gas-solid combustion synthesis, involves at least one gaseous
reagent in the main combustion reaction. Nitridation of titanium and silicon are common
Ti'` + 0.5N,'' = TiN"' + 335 kJ/mol
3Si'` + 2N,19' = Si3N4P' + 750 kJ/mol
The third main type of CS is reduction (thermite) combustion synthesis, where metal or
nonmetal oxides (e.g., Fe,03, B203, TiO,) react with a reducing metal (e.g., Al, Mg, Zr,
Ti), resulting in the appearance of another, more stable oxide, and reduced metal. This
reaction may be followed by the interaction of the formed reduced metal with other
elemental reactants to produce desired products. An example of this type of CS is
B203'" + 2A'ls + Ti"(s = A1203'" + TiB,"'' + 700 kJ/mol
where TiB2 is the desired product and A2O,3 can be removed (e.g., by centrifugal separa-
tion) and used separately, or a ceramic composite material (ALO3 + TiB,) can be pro-
In addition to the three main types of CS processes, there are two more recently
developed types where the initial reactants are all in either gas1561 or liquid phase,'71 while
Winter 2001


advantages of

the CS method





techniques of




include low



short synthesis

times..., high

(2000-4000 K),

and high

heating rates

(up to 106 K/s),
which allow

one to produce

unique (e.g.,



the final products are in solid state and are formed by a fast
propagating reaction wave.
In physical terms, there are two modes by which combus-
tion synthesis can occur: self-propagating high-temperature
synthesis (SHS) and volume combustion synthesis (VCS). In
both cases, reactant powders are pressed into a pellet, typi-
cally cylindrical or parallelepiped in shape. The samples are
then heated by an external source (e.g., tungsten coil, laser)
either locally (SHS) or uniformly (VCS) to initiate an exo-
thermic reaction. The characteristic feature of the more preva-
lent SHS mode is, after initiation locally, the self-sustained
propagation of a high-temperature reaction wave through the
heterogeneous mixture of reactants. Thus, the SHS mode of
reaction can be considered as a well-organized wave-like
propagation of the exothermic chemical reaction through a
heterogeneous medium, followed by the synthesis of desired
condensed products. During VCS, the second mode of CS,
the entire sample is heated uniformly in a
controlled manner until the reaction occurs
essentially simultaneously throughout the
volume. This mode of synthesis is more
appropriate for weakly exothermic reactions
that require preheating prior to ignition. A
sequence of video frames of reaction wave
propagation during CS of nickel aluminide
by the SHS mode is shown in Figure 1.

Since the initial discovery of the process,
the number of products synthesized by CS
has increased rapidly and currently exceeds
500 different compounds. Specifically, these
include advanced materials such as carbides,
borides, silicides, nitrides, oxides, interme-
tallics, and their composites. Examples of

these, along with their applications, are given in Table 1.
Most of these compounds possess high heats of formation-
this is the main reason why they can be produced by CS
without external heating.
In general, methods for the large-scale production of ad-
vanced materials by combustion synthesis consist of three
main steps:
Preparation of the green mixture
High-temperature synthesis
Post-synthesis treatment
A schematic diagram of these steps is presented in Figure 2.
The first step is similar to the procedures commonly em-
ployed in powder metallurgy, where the reactant powders
are dried (e.g., under vacuum at 80-1000C), weighed into the
appropriate amounts, and mixed (e.g., by ball mixing). For
some applications, cold pressing of the green mixture is

Figure 1. Sequence of video frames of reaction wave propagation during
combustion synthesis of nickel aluminide by the SHS mode: (a) ignition,
t=O; (b); (c) 0.2s; (d) 0.4s; (e) 0.6s; (f) complete reaction, 0.8s.

Types of Materials Synthesized

Chemical Formula
TiC, ZrC, HfC, TaC, NbC, SiC, TiC-CrC,3
ZrB, TiB,, HfB,, MoB, TaB,, LaB6, NbB,
TiSi,, TiSi, MoSi Zr Si, ZrSi
Aluminides of Ni, Zr, Ti, Cr, Co, Mo, Cu, etc.
Titanites: Ti-Ni, Ti-Co; Ti-Fe
TiN, ZrN, NbN, HfN, TaN, VN, AIN, Si ,N, BN

Examples ofApplications
Abrasives, cutting tools, ceramic reinforcements
Abrasives, cutting tools, cathode
Heating elements, electrical connectors, Schottky barriers for electronics
Aerospace and turbine materials, shape memory alloys

Ceramic engine parts, ball bearings, nuclear safety shields

TiH; ZrH; ZrNiH,; TiCoH3Zr,,; Nbo C,N, C xHos Hydrogen storage, catalytic materials
YBa,Cu 07; Bi4V On; LaSrCrO,; Na.sBi4,TiO 15; BaBi,Ni,O, High-temperature superconductors, gas sensors, fuel cells

Chalcogenides Phosphides Sulfides of Mg, Ti, Zr, Mo, W; GaAs;
Phosphides of Al, Ga, and In

High-temperature lubricants, semiconductors

Chemical Engineering Education



necessary, especially for the production of low-porosity or
poreless materials. The final step in sample preparation de-
termines the type of product synthesized; a powder product
results from uncompacted powder reactants, while sintered
products from cold-pressed compacts. Pressing the green

Green Mixture Preparation


Drying Weighing Mixing Cold press
of reactant powders of the p
........ .-.....-..-...-.:;. .- .. ....
Synthesis Technologies
Synthesis of powders Densification
and sintering

A '
Hydraulic Shock-wave Extrusion Hot isostatc Hot
hot pressing pressing pressing rolling
----- ....- -

'i. 'i Post Synthesis Treatment

S.Ball killing Polish


Figure 2. The main steps of combustion synthesis.

Figure 3. Examples of articles produced by combustion synth
graph courtesy of Academician Alexander G. Merzhanov, Chernogolc
Winter 2001

mixture into special molds or machining pressed initial com-
pacts yields complex-shaped articles.
The main production technologies of combustion synthe-
sis are presented in the second block of Figure 2. They may
be classified into several major types: powder production
and sintering, and densification, by tech-
niques including hydraulic, isostatic, or
shock-wave pressing, extrusion, and hot-roll-
ping. For highly exothermic reactions, where
the products are in molten state, centrifugal
casting is used to produce, for example, ce-
ramic-lined pipes.
sing The third main step of CS technologies is
lt treatment. Powder milling and
S sieving are used to yield powders with a
Casting and desired particle size distribution. The syn-
coatig thesized materials and articles may also be
|< machined into specified shapes and surface
finish. Examples of combustion-synthesized
articles, including ceramic engine parts, elec-
tric heating elements, high-temperature di-
---- electrics, and cutting tools, are shown in Fig-
ure 3.
Owing to the characteristic features of CS,
including high temperatures and short reac-
tion times, unique materials can be made
"g........ that cannot be synthesized by alternative tech-
niques. Thus, CS combined with pressing
has been used to obtain ceramic and inter-
metallic matrix-diamond composites with up
to 20 wt% synthetic diamond, for advanced
cutting tools.181 The graphitization during
high-temperature treatment of diamond
makes it impossible to produce such com-
posites using conventional furnace techniques,
while in the rapid combustion wave, diamond
particles retain their shape, surface quality,
and mechanical properties. The characteristic
microstructure of the diamond-containing por-
tion of a NiAl/diamond composite synthe-
sized by CS is shown in Figure 4a (next page).
Another unique example is synthesis of sili-
con nitride whiskers with aspect ratio (ratio
of length to diameter) of more than 104 and
length up to several centimeters191 (see Figure
4b). Such whiskers are widely used to rein-
force different types of brittle ceramics to
enhance their mechanical properties.
Finally, CS has been applied successfully
in the production of so-called functionally
graded materials (FGM). The concept of
esis (photo- FGMs is to tailor nonuniform spatial distri-
ivka, Russia). bution of components and phases in materi-

als, and hence combine mechanical, thermal, electrical,
chemical, and other properties that cannot be realized in
uniform materials. For example, the material structure may
have a smooth transition from a metal phase with good
mechanical strength on one side to a ceramic phase with
thermal resistance on the other side (see Figure 5a). With a
gradual variation in composition, FGMs do not have
intermaterial boundaries found in multilayer materials, and
hence they exhibit better resistance to thermal stress.1101 The
microstructure of a Cr3C2/Ni FGM, consolidated from a
green compact consisting of powder layers with different
Cr-C-Ni compositions by the CS+hot pressing method"'" is
shown in Figure 5b.

Figure 4. Examples of unique materials
produced by combustion synthesis:
(a) NiAl matrix/diamond composite,
(b) silicon nitride whiskers.

Figure 5. Functionally graded materials (FGMs):
(a) the concept of FGM; (b) CrC2/Ni FGM

Combustion Synthesis Basic Processes

SFinal product

Post-combustion and
structure formation

K Combustion zone

melt crystallization
during cooling

| rain gj|wt l

ard initial product
lii*:;: formation I
impurity gasificarion

|llh|reatirn|fer tooi ll
* t~~i~


Figure 6. Characteristic structure of the reaction wave
during combustion synthesis.

Chemical Engineering Education

The features of high heating rates, high temperatures, and
short times of reaction completion discussed above, although
attractive for the synthesis of unique compounds, also make
it difficult to study the mechanism of reaction wave propa-
gation, which is essential in order to form materials with
tailored microstructures and properties. Based on the results
obtained in CS, a new field of fundamental research that
incorporates concepts and principles from various branches
of science and engineering, which investigates initial stages
of chemical reactions in the combustion front and materials
structure formation in the heterogeneous medium after pas-
sage of the front, is being developed.
In general, characteristics of the combustion wave are
determined by the processes occurring in the heating, reac-
tion, and post-combustion zones (see Figure 6). The length

microgravity c
0.20 terrestrial

0.15 -



0 2 4 6
Time, s

Figure 7. Evolution of TiB2 (dark phase) grain size in
NiAl (light phase) matrix during combustion synthesis
in the (Ti+2B) (3Ni+Al) system, in microgravity and
normal gravity conditions.


I -a- Ti
I1 -P -Ti
v Tiq~N x I
V -TiOxN ---
VI TiO2 high-temperature phase
VII TiO rutile
VIII solid solution

2 4 6 8 10 12 14 16 18 20
time, s
Figure 8. Dynamics of phase formation during combustion of
titanium in air.
Winter 2001

of the preheating zone varies from 0.05 to 0.3 mm, while the
total wave is generally 1 to 2 mm and may be as wide as 20
to 30 mm for multistage reactions. A variety of physico-
chemical processes occur in different portions of the CS
wave. The initial stage of structure formation is concurrent
with the chemical reaction, where the driving force of the
process is a reduction of Gibbs free energy resulting from
formation of new chemical bonds under non-equilibrium
conditions. In the final structure-formation process, physical
effects are predominant where the free energy reduces fur-
ther due to interfacial surface reduction, ordering of the
crystal structure, and other related processes that occur with-
out changes in the chemical composition under quasi-
equilibrium conditions.
To study the variety of structural transformation processes,
it is necessary to use a wide range of methods. For example,
the evolution and morphological features of the microstruc-
ture during CS can be identified using a layer-by-layer mi-
croscopic and composition analysis of quenched samples.[121
The dynamics of phase composition and crystal structure
ordering can be monitored continuously by time-resolved X-
ray diffraction (TRXRD).'"I Also, microstructural analy-
sis of the combustion front, using a digital high-speed
microscopic video recording (DHSMVR)[14' provides im-
portant information about the local conditions, which
affect the synthesis process.
An illustration of the quenching technique involves com-
parison of the dynamics of structure formation in terrestrial
and microgravity conditions. It has been shown that the
microstructure of CS products is finer and more uniform in
microgravity. For example, the growth rate of TiB2 grains in
molten Ni3Al matrix during CS was four times smaller in
microgravity (10-5g) as compared to normal (1g) conditions
(see Figure 7), thus yielding a final cermet product with finer
grains that provide superior mechanical properties.[151

The application of the TRXRD method, which allows one
to obtain X-ray patterns of several non-overlapping lines of
each phase in complex systems every 0.1 seconds
of observation, can be demonstrated on the com-
bustion of titanium in air.1161 In this system, a com-
plicated mechanism was observed involving four
intermediate phases that preceded the formation of
the final equilibrium TiO, product:

CTi -- TNTNx Ti305-xNx TiO2-xNx
Ti2 (high T phase) -- TiO2(rutile structure)

The kinetics of the appearance and disappearance
of each phase are shown in Figure 8.
Finally, by using the DHSMVR method of in-situ
observation of rapid processes, with a rate of re-
cording up to 12,000 frames/second occurring at
the microscopic level (spatial resolution of approxi-
mately 1.5 microns), significantly new information

Nb + 2B

Ti+ Si


X ,
rz:- .


I*, 14

a $~~
. 41.

reaction spot


along the front

reaction spot


along the front

Figure 9. Microstructures of reaction wave during combustion synthesis:
(a) quasihomogeneous reaction wave; (b) scintillating reaction wave.
Chemical Engineering Education

16; nis



about the microstructure of gasless heterogeneous combus-
tion waves was recently obtained. It was shown""71 that while
on the macroscopic length and time scales, the reaction
appears to move in a steady mode, on the microscopic level
it may have a complex unsteady character that is related to
the reaction mechanism. In general, we may classify these
waves into two types, according to their microstructures:
quasihomogeneous reaction waves and scintillating reaction
waves. In the former case, the wave moves steadily and there
is relatively little variation of temperature along the surface
of the front (see Figure 9a). In the latter case, however, a hot-
spot initiates the reaction ahead of the front and the wave
moves forward only as a consequence of the appearance of
the hot-spots (see Figure 9b). Thus, an essential temperature
nonuniformity exists in the reaction front that leads to
nonuniformity of the product microstructure formation. In this
case, the extent of nonuniformity can be controlled, i.e., the
number and frequency of scintillations decrease by using finer
reactant powders or samples with higher initial density.I41

There are some successful commercial applications of the
combustion-synthesis technique for production of advanced
materials. They include nitride ceramics, high-temperature
heating elements, shape-memory alloys, ceramic-lined pipes,
and high-performance composites. In my own laboratory at
Notre Dame, we are currently working on two major appli-
cations with industrial partners. The first involves enhance-
ment of cobalt-based alloys, where we are working with
Zimmer, Inc., a world leader in production of orthopedic
implants, to develop stronger alloys as well as new CS-based
technology that will eliminate several manufacturing steps.
The second project relates to improving production of emer-
gency oxygen by combustion of low-exothermic condensed
reactants, where we are collaborating with B/E Aerospace,
Inc., the leading manufacturer of chemical oxygen genera-
tors for passenger aircraft applications.
While combustion synthesis offers a number of potential
advantages over conventional techniques, some hurdles need
to be overcome before it can enjoy widespread use. The
foremost among these is a mechanistic understanding of
reaction and structure-formation processes, which are in-
deed complex owing to the heterogeneous nature of the
reaction media. This understanding is critical in order to
produce materials with tailored microstructures and proper-
ties. As discussed above, excellent analytical tools are now
becoming available that facilitate this understanding. Based
on such insights, combustion synthesis is expected to be-
come an important technique in the 21" century for the
production of a variety of advanced materials.

It is a pleasure to thank my long-term collaborators, Drs.
Alexander Mukasyan and Alexander Rogachev, for their
Winter 2001

help in preparing this manuscript. I am grateful to the Na-
tional Science Foundation (grant CTS-9900357) and the Na-
tional Aeronautics and Space Administration (grant NAG3-
2213) for support of my research in this field.

1. Munir, Z.A., and U. Anselmi-Tamburini, "Self-Propagating
Exothermic Reactions: The Synthesis of High-Temperature
Materials by Combustion," Mater. Sci. Reports, 3, 277 (1989)
2. Merzhanov. A.G., "Self-Propagating High-Temperature Syn-
thesis: Twenty Years of Search and Findings," in Combus-
tion and Plasma Synthesis of High-Temperature Materials,
edited by Z.A. Munir and J.B. Holt, VCH Publishers, New
York, NY (1990)
3. Varma, A., A.S. Rogachev, A.S. Mukasyan, and S. Hwang,
"Combustion Synthesis of Advanced Materials: Principles
and Applications," Adv. in Chem. Eng., 24, 79 (1998)
4. Varma, A., "Form from Fire," Scientific Amer., 283(2), 58
5. Calcote, H.F., W. Felder, D.G. Keil, and D.B. Olson, "A New
Flame Process for Synthesis of Si:N4 Powders for Advanced
Ceramics," 23rd Symposium (International) on Combus-
tion, 1739 (1990)
6. Davis, K., K. Brezinsky, and I. Glassman, "Chemical Equi-
librium Constants in the High-Temperature Formation of
Metallic Nitrides," Combus. Sci. Tech., 77, 171 (1991)
7. Patil, K.S., S.T. Aruna, and S. Ekambaram, "Combustion
Synthesis," Current Opinion in Solid State & Mat. Sci., 2,
8. Levashov, E.A., I.P. Borovinskaya, A.S. Rogachev, M.
Koizumi, M. Ohyanagi, and S. Hosomi, "SHS: A New Method
for Production of Diamond-Containing Ceramics," Int. J.
SHS, 2, 189 (1993)
9. Mukasyan, A.S., and I.P. Borovinskaya, "Structure Forma-
tion in SHS Nitrides," Int. J. SHS, 1, 55 (1992)
10. Sata, N., N. Sanada, T. Hirano, and M. Niino, "Research
and Development on Functionally Gradient Materials by
Using a SHS Process," in Combustion and Plasma Synthe-
sis of High-Temperature Materials, edited by Z.A. Munir
and J.B. Holt, VCH Publishers, New York, NY, 195 (1990)
11. Miyamoto, Y., K. Tanihata, Y. Matsuzaki, and X. Ma, "HIP
in SHS Technology," Int. J. SHS, 1, 147 (1992)
12. Rogachev, A.S., V.A. Shugaev, C.R. Kachelmyer, and A.
Varma, "Mechanisms of Structure Formation During Com-
bustion Synthesis of Materials," Chem. Eng. Sci., 49, 4949
13. Kachelmyer, C.R., 1.0. Khomenko, A.S. Rogachev, and A.
Varma, "A Time-Resolved X-Ray Diffraction Study of Ti5Si3
Product Formation During Combustion Synthesis," J. Mater.
Res., 12, 3230 (1997)
14. Hwang, S., A.S. Mukasyan, and A. Varma, "Mechanisms of
Combustion Wave Propagation in Heterogeneous Reaction
Systems," Combust. Flame, 115, 354 (1998)
15. Mukasyan, A.S., A.E. Pelekh, A. Varma, A.S. Rogachev, and
A. Jenkins, "The Effects of Gravity on Combustion Synthe-
sis in Heterogeneous Gasless Systems,"AJAA J., 35, 1 (1997)
16. Khomenko, 1.0., A.S. Mukasyan, V.I. Ponomarev, I.P.
Borovinskaya, and A.G. Merzhanov, "Dynamics of Phase-
Forming Processes in Metal-Gas System During Combus-
tion," Combust. Flame, 92, 201 (1993)
17. Varma, A., A.S. Rogachev, A.S. Mukasyan, and S. Hwang,
"Complex Behavior of Self-Propagating Reaction Waves in
Heterogeneous Media," Proc. Natl. Acad. Sci. USA, 95, 11053
(1998) 0

[f class and home problems

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



New Jersey Institute of Technology Newark, NJ 07102

Chemical engineers are quite familiar with the use of
the Laplace transform method for solving linear or-
dinary differential equations. Usually, the differen-
tial equation is converted to an equivalent algebraic equa-
tion, then the appropriate initial conditions are applied, and
the resulting algebraic equation is prepared for inversion in
order to recover the sought-after solution.
Frequently, the techniques to invert the resulting algebraic
equation involve the use of a table of Laplace transforms.
Most practitioners of this approach develop devices to ex-
tend their table of Laplace transforms when their particular
inversion is not listed.
There is an alternate technique, however, that is especially
useful when a difficult inversion is to be performed. This
method employs a concept that is fundamental in the theory of
functions of a complex variable-namely the residue theorem.
Following Mickley, Sherwood, and Reed,"' Churchill and
Brown,[2] and Dettman,13] the variable s in

F(s)= e-stf(t)dt (1)
can be interpreted as a complex number. Here, F(s) is the
Laplace transform of f(t). Further, except for singularities,
F(s) is usually analytic (has a Taylor series expansion).
A frequently encountered class of problems in chemical
engineering are the Sturm-Liouville problems, and it is use-
ful to know that the transform of a solution to a Sturm-
Liouville equation is analytic for all finite s except at the
singularities (poles) of the system.
When F(s) is analytic, except for poles, the inverse trans-
form is given by

f(t)= L- F(s)}= p (t) (2)
where pn(t) is the residue of F(s) at the pole Sn. Even though
this concept is firmly grounded in the theory of functions of
a complex variable, direct use of complex variables is not
always required. A procedure is given below that avoids the
direct use of complex variables.

Rewrite F(s) as a quotient

F( s) (3)

which enables us to quickly identify the singular points
(poles) of F(s) and to determine if the degree of Q(s) is at
least one greater than that of P(s). This procedure may re-
quire power series expansions of both P(s) and Q(s). If the
degree of the denominator is at least one greater than that of
the numerator, and the poles are simple (singularities of
order one), then

Pn(t) P(s (4)
where Q'(sn) is the derivative of Q(s) evaluated at the
simple pole s,.
Norman W. Loney is Associate Professor of Chemical Engineering at New
Jersey Institute of Technology (NJIT). He has studied chemical engineering
at NJIT and applied mathematics at Courant Institute of Mathematical
Science. In addition, Dr. Loney has practical experience in process develop-
ment, process design and inplant engineering. His new book entitled Ap-
plied Mathematical Methods for Chemical Engineers by CRC Press became
available in October of 2000.
Copyright ChE Division of ASEE 2001
Chemical Engineering Education

If the poles are of order m (multiple pole), then
t2 m-lA 1 m tJ-1
pn(t) =esnt A+tA2+ Am +...+( t esnt jA-t 1
2! (m-1 (j0)

Q(s) has repeated roots at +i,+i, so that the multiplicity, m, of
each root is two. Therefore, as singularities of F(s), these are
poles of order 2. Then using Eq. (5) with s=-i,
P-i(t)= e-it(Al + tA2) (16)

(5) since m = 2. Also,

The A's are defined by
1 dm-I
A -1 d [(s-sn)mF(s)] j=1,2,...,m (6)
(m-j)! dsm-J S=Sn
Three examples are presented below. Examples 1 and 2
are elementary problems that can be quickly inverted by use
of tables; they are presented here to illustrate the concept of
the residue method. The third example demonstrates a more
appropriate application of the method.


1. Simple Poles
Suppose we need to invert
5s2-7s+17 P(s)
F(s) (7)
(s l1)(s2 + 4) Q(s)
P(s)=5s2-7s+17, Q(s)=(s-1)(s2+4), Q'(s)=s2+4+(s-1)2s
The roots of Q(s) are the simple poles of F(s). Therefore, Eq.
(4) is the appropriate form with which to evaluate the resi-
dues since the poles are not repeated; that is

(s+i)2 S s
(s-i)2(s+i)2 (s-i)2
Application of Eq. (6) gives

1 d s -i-s
A= (2-1)! ds (S i)2 -2(s i)3

=0 (18)

A2 s i
- (s-i)2]s=-i 4



p-i(t)= e-it

t it
Pi(t)= --te

such that Eq. (2) results in
t (eit -e-it
f(t)= P-.(t)+pi(t)= 2= t sint (22)

in which use of the identities
ei0 -ie i0 _e-ie
e +e e -e
cos = and sin = --(23)
2 2i
are employed to express the final results of both examples.

3. Diffusivities of Gases in Polymers

Pl(t) = (1) Q 3e'

2it= P(2i) e2it (5 1+ 2it
P2 ( Q'(2i) -4
P i(t)= P(-2i) -2it ( 5i -2it
Q'(-2i) -l 4

so that use of Eq. (2) results in

f(t) = P(t)+p2i (t) +p-2i(t)
That is
f(t)= 3et +2cos2t--sin2t

2. Multiple Poles
Suppose we wish to invert

(2 +1)2

for which

Winter 2001

P(s)=s and Q(s)= (s2 +1)2

k ) Consider a model of diffusion through a membrane that
separates two compartments of a continuous-flow permeation
(10) chamber. Then, following Felder, Spence, and Ferrell,141 at
time t=0, a penetrant is introduced into one compartment (the
(11) upstream compartment) and permeates through the membrane
into a stream flowing through the other (downstream) com-
partment. Further, this model includes the following assump-
(12) Diffusion of the penetrant in the gas phase and absorption at
the membrane surface are instantaneous processes.
Diffusion in the membrane is Fickian with a constant
(13) diffusivity.
The concentration of dissolved gas at the downstream
surface of the membrane is always sufficiently low compared
to that at the upstream surface, such that the downstream
surface concentration may be set equal to zero.
(14) Then, diffusion through a flat membrane of thickness h is
described by
aC(t,x) a2C(t, x)
D (24)
at -D ax2
(15) subject to

C(0,x)=0 (25)
C(t,0)=C, (26)
C(t,h)= 0 (27)
Application of Eq. (1) to transform Eqs. (24 27) results in
the second-order constant-coefficient homogeneous differ-
ential equation
0=d2y(s,x) sy (28)
dx2 D

subject to



y(s,h)= 0 (30)
The solution to the boundary-value problem described by
Eqs. (28-30) is
( s 7

y() sinhD hjcsh xj-c x c h Jsinh| x -x
y(s X, )=C FD D F ] D D I
ssinh h) ]

Then, applying Eqs. (2-4) to invert y(s,x), we get

L-l[y(s,x)] =C(t,x)=C P(snsexp(s t)=
n= Q'(sn)

iP(s) sinh(h- x) h-
Po(t)= lim s = lims (38)
s-o Q(s) s-O h s ) h
ssinh( DhJ

gives the residue at s=0 using l'Hospital's rule. for Sn 0,

Q'(sn)= snh I h +sIs coshi j h (39)

The simplifying substitution i = s / D results in

.=n-, n=1,2,--- (40)
sn = -nD (41)
That is, when Eq. (37) is set equal to zero, either s=0 or
sinh /s D h =0. The case s=0 results in the residue po(t)
given above (Eq. 38), while the case sO gives sinh ik = 0, a
condition that is satisfied for the values of k as given in Eq.
(40). Finally, after performing the necessary algebra, we get
the result

(tx P(sn,)esnt 2 sin(nx 'Dt (42)
Q'(sn) --nn--m- h je
and the concentration profile is
h-xh 2- 1 (n2 2tD) ( (nx)
C(t,x)= C, -r exp 2 Jsin (43)

Then, the rate of penetrant across the surface x=h is given by

C iL- '' si (32)

Recall that

P(Sn) P(s) P(s)
Q'(sn) s-i [Q(s)-Q(sn m ss (33)
L s-sn
such that for so=0

po(t) lms (34)
Q'(0) s-o Q(s)
Then for

y(s,x) =

P(s,x)= sinh(h x) D

Q(s)= ssinh s Dh

J(t') -_ -DAI ac I

DACI (- nx 2 Dt (44)
1+2n(-1) exp h2 (44)

for a flat membrane with a surface area A. Also notice that
the steady-state rate, J,, is given by
Jss = D-A (45)
For an example involving cylindrical geometry, the reader
is directed to the recent literature where a model based on
membrane separation is treated by Ramraj, Farrell, and
Loney.151 Also, a model involving membrane separation with
chemical reaction in a flowing system is treated by Loney.[6'
Inversion by the residue method is not a new concept;
however, it can be very useful in efficiently solving systems
of non-homogeneous linear partial differential equations.

1. Mickley, H.S., T.K. Sherwood, and C.E. Reed, Applied Math-
ematics in Chemical Engineering, McGraw-Hill (1957)
2. Churchill, R.V., and J.W. Brown, Complex Variables and
Applications, 4th ed., McGraw-Hill (1984)
3. Dettman, J.W., Applied Complex Variables, Dover (1984)
4. Felder, R.M., R.D. Spence, and J.K. Ferrell, J. Appl. Poly.
Sci., 19, 3193 (1975)
5. Ramraj, R., S. Farrell, and N.W. Loney, J. of Memb. Sci.,
162, 73 (1999)
6. Loney, N.W., Chem. Eng. Sci., 15(6), 3995 (1996) J
Chemical Engineering Education


Random Thoughts...


North Carolina State University Raleigh NC 27695

Most of us in this business are periodically called on
to evaluate award nominations and applications
for faculty positions. It's not an easy job if you
take it seriously. You have to make intelligent judgments
about somebody's qualifications based on lists of articles in
journals you never heard of on subjects you're unfamiliar
with written by multiple authors whose roles in the work are
undefined. You also have to deal with collections of refer-
ence letters unanimously assuring you that the applicant is
currently or potentially the finest researcher, teacher, and
all-around human being the world has ever known. Interpret-
ing all that unfiltered verbiage to figure out just how good
someone actually is takes considerable experience and skill.
As a service to the profession, I've prepared a glossary of
classifications that will make interpretation of award and
application dossiers a snap once everyone agrees to adopt it.
Here, for example, is how a hypothetical citation might
appear in a dossier submitted by one Alvin Tsimmes, fol-
lowed by the glossary.

23. B.A. Mensch,[b" A. Tsimmes,"dl and O.Y. Gevalt,/"'
"The Environmental Impact of Biochemical
Superconductors,",'l Journal of the Environmental
Impact of Biochemical Superconductors,'PJ 27(3),
357-375 (1996).

Categories of authors: (a) did most of the research; (b)
contributed significantly to the research; (c) occasionally
attended meetings and had a vague idea what the research
was about; (d) clueless about the research but part of a
faculty group that automatically shared authorship of all
papers on the subject; (e) owned some equipment needed for
the research; (f) had nothing to do with the research but
famous enough to guarantee the paper's acceptance; (g)
department head
Categories of papers: (h) classic in its field-hundreds of
citations; (i) many citations by non-authors; (j) mainly or
entirely self-citations; (k) no citations; (1) named in a reader's
poll as the paper least likely to be cited by anyone ever
Categories of journals: (m) top of the line: non-National
Academy members can forget it; (n) prestigious and hard to
get into: referees are those people whose hobby is asking
Winter 2001

sarcastic questions at conferences to prove they know more
than the presenters; (o) refereed; (p) sort of refereed: condi-
tion for acceptance is lack of obvious gross errors; (q) non-
refereed: condition for acceptance is submission; (r) last
resort: authors who submit papers get a free lifetime sub-
scription, two large pizzas with toppings of their choice, and
a shoeshine; (s) accepts papers rejected by (r)
Next, here is the list of authors of Tsimmes's reference
letters: 1. Dr. R.U. Sirius;["' 2. Mr. I.C. Dimmly;t'` 3. Prof.
U.R. Toest'"'
Categories of reference letters: (t) completely sincere; (u)
exaggerated but more or less accurate; (v) politely camou-
flaged negative opinion; (w) form letter-respondent never
heard of the applicant and didn't look at the dossier; (x)
blatant lying
The creativity required to write good Category (v) letters
has led to such gems as
El "I cannot recommend him too highly,"
El "With her on your staff there's no telling what you might
EL "I am pleased to say that this candidate is a former
colleague of mine,"
E "I can assure you that no person would be better for this
E "I would waste no time in making this candidate an offer of
and the ever-popular
E "I've known Mr. Jones for many years, and you'll be lucky
if you get him to work for you!"
Dossier evaluators could get a serious headache trying to
interpret these remarks and possibly make a serious mistake
if they guess wrong, but a "v" by the authors' names would
make life easy for everyone (except perhaps Mr. Jones).
Okay, now that we've got all that worked out, let's discuss
applying the same technique to TV commercials, political
speeches, marriage proposals, and perhaps even journal
'" This column is a facetious and excessively cynical adaptation
of a clever idea suggested by Kay C. Dee of Tulane in a casual
conversation. The author should be ashamed of himself
Copyright ChE Division of ASEE 2001





Georgia Institute of Technology Atlanta, GA 30332

Rapidly escalating gasoline and heating oil costs in
the Spring of 2000 represented the first major en-
ergy consumption crisis in this country since 1973.
Nonetheless, this experience served once again to demon-
strate the vulnerability of the nations of the Western world to
production and marketing policies in the various oil-produc-
ing nations. One of the many energy conservation efforts in
the 1970s after that earlier crisis was associated with a search
for alternative methods of cooling, air-conditioning, and re-
frigeration. r One of these methods, using low-grade ther-
mal energy (e.g., solar or waste heat) to power the cooling
cycle, forms the subject of this article.
The most expensive step, corresponding to the greatest
amount of energy consumption, in conventional refrigera-
tion cycles is the mechanical compression step, wherein a
refrigerant vapor is compressed from a low pressure to a
higher pressure. It is then condensed to liquid form in, typi-
cally, an air-cooled heat exchanger before expansion back to
the same low pressure in an expansion valve, followed by
vaporization-whereby the refrigeration effect occurs. This
refrigerant vapor is then recompressed to the higher pres-
sure, and the cycle is complete. These mechanical compres-
sors are typically driven by electric motors or internal com-
bustion engines, the energy sources for which can generally
be traced back to fossil fuels or nuclear power.
For more than twenty years now,[21 the widespread use of
vapor-compression refrigeration for commercial and house-
hold air conditioning has caused a shift in the seasonal peak
for electric power production from mid-winter to mid-sum-
mer. This trend naturally suggests the possibility of match-
ing demand with availability, i.e., the use of solar thermal
energy in the neighborhood of 200'F to power the cooling
cycle rather than mechanical work, at least in certain climes.
Related to this possibility is the suggestion of heat-driven
mobile refrigeration cycles,[31 as in an automobile, wherein
waste heat from engine cooling water could serve as the

driving medium. All of these developments were natu-
rally spurred by various energy tax credits,[41 also inau-
gurated in the 1970s.

By contrast with the conventional vapor-compression re-
frigeration cycle with its two refrigerant pressure levels, in
this solar-powered refrigeration cycle there are three differ-
ent pressure levels: low, medium, and high. Here, the high-
pressure is achieved by pumping a portion of the liquid
refrigerant stream, and not a vapor stream. This high-pres-
sure liquid refrigerant stream is vaporized in a solar-collec-
tor heat exchanger. The high-pressure vaporized stream then
serves as the motive stream to a thermal (or jet) compressor,
wherein this stream sucks up the low-pressure stream from
the refrigeration coil, thereby creating a medium-pressure
stream (much like a laboratory aspirator). This latter stream
is then totally condensed, as in a conventional refrigeration
cycle, typically by heat exchange with ambient air. Part of
this condensed stream feeds through an expansion valve in
the low-pressure loop and then evaporates in the refrigera-
tion coil, again just as in conventional refrigeration. The
remainder of the condensed mixed stream is pumped in the
high-pressure loop and to the solar collector, thus complet-
ing the refrigeration cycle.
It is clear that the crucial piece of equipment in the above
cycle is the thermal compressor. With its lack of moving
parts, it is certainly an attractive alternative to the conven-
tional mechanical compressor. Rigorous mathematical mod-
eling of thermal compressors is a rather formidable task,
Jude T. Sommerfeld is a professor of chemical engineering at the
Georgia Institute of Technology, where he has also served as Associate
Director of the School. He received his BChE degree from the University
of Detroit, and his MS and PhD degrees (both also in chemical engineer-
ing) from the University of Michigan. His primary interest is computer-
aided design of chemical processes, and he has published approximately
150 papers in this and other areas.
Copyright ChE Division of ASEE 2001
Chemical Engineering Education

however. Earlier one-dimensional models
of such devices15'61 proceeded from first
principles in fluid mechanics and thermo-
dynamics. The model of DeFrate and
Hoerli6i was later coded in a FORTRAN
routine,[71 suitable for incorporation into
the early FLOWTRAN systeml81 for
CAPD (but never implemented therein).
Indeed, most of today's state-of-the-art
CAPD systems in chemical engineering
still do not have building blocks or mod-
ules for thermal compressors. This lack is
due to, among other things, the complex-
ity of the models themselves, difficul-
ties in generalization, and the need for
detailed specifications.

----- --- --- -- --C ..

-- i I TOCOND T T,-?L |
r- _J

Figure 1. HYSYS process flow diagram for a solar-powered
refrigeration cycle.

There was an early attempt at modeling a solar-powered refrigera-
tion cycle using the seminal FLOWTRAN system.[91 The module
used for the thermal compressor was an adiabatic flash block, in
which the high-pressure stream was simply mixed adiabatically with
the low-pressure stream to yield the medium-pressure stream. The
same effect could similarly have been achieved with a mixer module
also operating in adiabatic fashion. Thus, these modules in the
FLOWTRAN system would allow a pressure rise across them, with no
concern as to how this pressure increase was to be achieved. The same,
admittedly unrealistic, capability existed in the PRO/ITI system.["'0

Current CAPD systems in chemical engineering have a graphical
user interface (GUI) as their input/output medium, and operate on a
personal computer (PC) platform. One such modern system is
HYSYS-the precursor of which was HYSIM,1"' both of which
were developed by Hyprotech, Ltd., in Canada. This HYSYS system
is the one presently used in the chemical engineering instructional
program at Georgia Tech, and thus it is the one employed in this study.
The process flow diagram (PFD) for the solar-powered refrigera-
tion cycle, as constructed by the HYSYS system, is shown in Figure
1. The various streams and unit operations in this PFD, as well as
their functions, are described in Table 1. This HYSYS system is
more realistic in the sense that it will not allow a pressure rise across
either an adiabatic flashing or mixing operation. That is, the outlet
stream pressure from the unit cannot exceed the pressure of any one
of the incoming process streams. And, as with most present CAPD
systems in chemical engineering, there is no formal thermal com-
pressor module in the HYSYS system. Thus, an alternative method
must be developed to simulate this device in the refrigeration cycle, the
description of which follows.
A logical place to begin the description of this three-pressure-level
refrigeration cycle is with the mixed vapor stream exiting the MIXER
unit. This stream is at the medium pressure of the loop and feeds the
air-cooled condenser (duty of QCOND), as in a conventional air
conditioning cycle. It was assumed here that the exiting stream from
this condenser was saturated liquid refrigerant at 1250F. This as-

Winter 2001

Streams, Unit Operations, and Their Functions
in the HYSYS Simulation of a
Solar-Powered Refrigeration Cycle

Unit Operation

Operation Function
Mixes the vapor streams (both at the same
medium pressure) from the high-pressure
(FROMCMPRES) and the low-pressure

CONDENSER Rejects heat from the mixed vapor stream
(TOCOND) to the ambient air (duty =
SPLITTER Splits the condensed stream (TOSPLIT) into
the two parts feeding the low-pressure
(TOVALVE) and high-pressure (TOPUMP)
EXPANSION Reduces the pressure of the liquid stream in
the low-pressure loop (TOCOIL)
REFRIGCOIL Extracts heat from the environment to
vaporize the stream in the low-pressure loop




Varies the refrigerant flow rate in the low-
pressure loop to achieve the desired refrigera-
tion duty in the coil (QCOIL)
Compresses (work = WCOMP) the low-
pressure vapor stream to the cycle's medium
pressure (FROMCMPRESS)
Increases the pressure of the liquid stream
(TOPUMP) in the high-pressure loop (work =

COLLECTOR Vaporizes the liquid stream (FROMPUMP) in
the high-pressure loop with solar or waste heat
(duty = QSOLAR)
EXPANDER Reduces the pressure (work = WEXPAND) of
the vapor stream in the high-pressure loop
(TOEXPAND) to the cycle's medium pressure


Equates the work of compression (WCOMP)
in the COMPRESSOR with the work of
expansion (WEXPAND) in the EXPANDER

sumption thus determined the medium-pressure level in this
process for a given (pure component) refrigerant. A sum-
mary of the various assumed operating conditions for all of
these simulations is given in Table 2.
The liquid stream exiting the condenser is fed to a tee
module (named SPLITTER); here, the stream is divided into
the two parts, feeding the low-pressure and high-pressure
loops, respectively. The stream for the low-pressure loop is
fed through a conventional expansion valve and then to the
refrigeration coil. The duty (QCOIL) of this coil (and hence
of the refrigeration cycle) was set at

4 tons = 48,000 BTU/hr = 0.48 therm/hr
in all of these simulations. This duty is a typical value for a

Operating Conditions and Parameters in the Simulations
of a Solar-Powered Refrigeration Cycle
Condition/Parameter Value
Temperature of the saturated vapor refrigerant leaving
the refrigerant coil (TR) 40'F
Temperature of the saturated liquid refrigerant leaving
the air condenser (Tc) 125F
Temperature of the saturated vapor refrigerant leaving
the solar collector (Ts) 200F
Refrigeration duty of the coil and cycle (QR) 4 tons

Figure 2. HYSYS process flow diagram for a mechanical
vapor-compression refrigeration cycle.

Refrigerants and Operating Pressures and Flow Rates
(Refrigeration Duty (QR) = 4 tons = 0.48 therm/hr)

Low Medium High Cold Loop
Pressure Pressure Pressure Flow Rate
Refrigerant (PR), psia (Pc), psia (Ps), psia (M), lbs/hr

R-113 2.906 17.34 55.03 1014
R-134a 49.7 200 506.1 881
Propane 78.49 258.8 581.8 477
Iso-propanol (IPA) 0.25 3.96 22.12 179.5

modern residential dwelling of moderate size. This condi-
tion was achieved with the aid of an adjust module (named
REFRIGDUTY), which varied the flow rate of refrigerant in
the low-pressure or cold loop, much like a proportional
controller. The thermodynamic condition of the refrigerant
leaving the coil was assumed to be saturated vapor at 400F;
this condition then specified the operating pressure level in
the cold loop. The refrigerant vapor was next supplied to a
conventional mechanical compressor module (nonexistent
in the actual process itself, of course), which recompressed
this stream to the medium pressure of the process before
entering the mixer module. The power requirement of this
compressor is denoted as WCOMP.
Returning to the tee module (SPLITTER) following the air
condenser, the remainder of the saturated liquid refrigerant
was pumped (power requirement = WPUMP) to the high-
pressure level of the process. This latter value, for a given
refrigerant, was dictated by the condition that the refrigerant
exiting the downstream solar collector (duty of QSOLAR)
was saturated vapor at 2000F-a not unreasonable value for
modern solar collection systems. This collector, as well as
the air condenser and the refrigeration coil, were all modeled
by simple process-utility heat exchangers in these simula-
tions. Also, for simplicity, any process fluid pressure drops
in these exchangers were neglected.
Lastly, the saturated refrigerant vapor from the solar col-
lector is fed to a conventional mechanical expander (again,
non-existent in the actual process). This expander module
reduces the vapor refrigerant pressure down to the medium-
pressure level in the process and generates a work stream
denoted as WEXPAND. Another adjust module (named
EQUALWORKS) equates the power required by the me-
chanical compressor with that generated by the expander, by
varying the refrigerant flow rate in the high-pressure loop.
The process output stream from this expander, at the same
medium-pressure level as the vapor stream from the com-
pressor is then mixed with this latter stream to form the
input to the air condenser, thus closing the loops.
That part of the process flow diagram in Figure 1 repre-
senting the thermal compressor is enclosed within the dashed-
line envelope of that figure. The units enclosed therein in-
clude the mechanical compressor, mixer, expander, and the
adjust block to equate the work streams for the two mechani-
cal units. Also, for comparison purposes, the HYSYS pro-
cess flow diagram for a comparable and conventional vapor-
compression refrigeration cycle is shown in Figure 2. It is
obviously considerably simpler, in that the high-pressure
loop from Figure 1 is no longer present and there is no need
to construct an artificial representation of a thermal com-
pression unit here.
Four different pure-component refrigerants were investi-
gated in this simulation study. These are summarized in
Table 3. The first one of these, R-113, has been a popular

Chemical Engineering Education






refrigerant for many home air conditioning systems and was the refrigerant em-
ployed in earlier analyses 231 of refrigeration cycles using a jet ejector. This particu-
lar refrigerant is rapidly being replaced, however, with R-134a-a more environ-
mentally friendly species. The remaining two prospective refrigerants considered
(propane'" and isopropanol91) were similarly studied by earlier investigators of these
cycles. Once a refrigerant had been chosen and given the operating conditions
specified in Table 2, all of the remaining process conditions followed. These latter
conditions, such as the three operating pressures in the cycle and the refrigerant flow
rate in the cold or low-pressure loop, are also given in Table 3. Lastly, the Peng-
Robinson thermodynamic system for computing physical properties as implemented
in the HYSYS system was employed throughout this work.

The use of each of the above four refrigerants in a solar-powered air conditioning
cycle rated at 4 tons of refrigeration was investigated. Specifically, the effect of the
adiabatic efficiencies of the compressor/expander combination on the performance
of the cycle was determined. Five different values of this efficiency were
chosen: 100, 90, 75, 60, and 50%. The same value of the efficiency (e.g., 75%)
was applied to both the compressor and the expander in a given simulation.
These efficiency values bracket the compression ratio efficiencies determined
experimentally (56 to 74%) in an earlier study of jet ejectors using butane and
hexane as the process fluids.15'
The simulation results obtained for the four refrigerants are summarized in Tables
4-7, respectively. Some general observations from these tables may be made first.
Thus, all of the dependent parameters shown in the tables, save for the coefficient of
performance (COP), increase monotonically with decreasing compressor/expander
efficiency. For all practical purposes, the condenser duty (Qc) exceeds the solar
collector duty (Qs) by the assumed refrigeration duty (QR = 0.48 therm/hr). This is
readily apparent from the specific curves in Figure 3 for the condenser and collector
duties in the case of refrigerant R-134a. The small amount of thermal energy
contributed by the pump work (W,) is also rejected by the condenser. This latter
power stream is the only mechanical energy contribution to this loop, and varies
from fractions of a horsepower up to 4+ hp in the worst case of propane as the
refrigerant at 50% adiabatic efficiencies (Table 6). The magnitude of the pump
work stream is obviously directly related to the magnitude of the hot loop
circulation rate (Mh in these tables).
The coefficient of performance (COP) for a refrigeration cycle or heat pump is
generally computed as the ratio of the refrigeration duty or the amount of heat
pumped to the thermal energy or mechanical work supplied to the cycle."21 Thus, a
COP value in this study was computed as the quotient of the refrigeration effect (QR)
divided by the solar collector duty (Qs), or COP = QR/QS. Note that the small amount
of mechanical work contributed to the cycle by the pump (Wp) was ignored. This
COP quantity then varies in the range of 0.45-0.55 down to 0.1+ as the adiabatic
efficiencies decrease. The best values are observed in the case of isopropanol (Table
7). Among other deficiencies, however, this refrigerant suffers from the rather large
compression ratio (>10) required in the compression step (see Table 3). Isopropanol
is followed in performance by R-113, as shown in Table 4. Propane and R-134a
(Table 5) are virtually identical in displaying the poorest performance of the four
refrigerants; they also require the highest operating pressures in the loops.
The COP values calculated above may be compared with maximum theoretical
values. Thus, Chen[31 also begins his analysis of this cycle by ignoring the negligible
amount of work contributed to the cycle by the pump (Wp). The maximum attainable
coefficient of performance for the ejector-operated refrigeration cycle is then equal
Winter 2001

... this study

of a
represent an
aided design
project in an
It is in this
spirit that
this study

Effects of Compressor/Expander Efficiencies with R-113
(Refrigeration Duty (QR) = 4 tons = 0.48 therm/hr)

Expander Collector
Efficiency Duty (Q),
(E), % therms/hr


Duty (Qc),


COP Expander
(=Q/Qs) Work (W), hp



Hot Loop
Pump Work Flow Rate
(W), hp (M), lbs/hr


Effects of Compressor/Expander Efficiencies with R-134a
(Refrigeration Duty (Q,) = 4 tons = 0.48 therm/hr)

Expander Collector
Efficiency Duty (Q),
(E), % therms/hr


Duty (Qc),


COP Expander
(=QJQ) Work (W), hp



Hot Loop
Pump Work Flow Rate
(W), hp (M), bs/hr





S0 2 Q co ector

4 1

50 60 70 80 90 100
Compressor/Expander Efficiency, %

Figure 3. Air condenser and solar collector duties for a
4-ton solar-powered air-conditioning cycle, with R-134a
as the refrigerant, as functions of the adiabatic efficien-
cies of the compressor and expander.

to the coefficient of performance for a Carnot refrigeration
cycle (COPc) working between the temperatures of the re-
frigeration coil (TR) and the heat rejection temperature (To),
multiplied by the efficiency of a Carnot heat engine (Ec)
operating between the solar collector temperature (Ts) and
the rejection temperature of To. The above Carnot refrigera-
tion cycle can also be viewed as a heat pump operating in the
cooling mode between the two temperatures of TR and To. If
one selects the condenser temperature (Tc) as the heat rejec-
tion temperature, then

Effects of Compressor/Expander Efficiencies with Propane
(Refrigeration Duty (QR) = 4 tons = 0.48 term/hr)
Expander Collector Condenser Compressor/ Hot Loop
Efficiency Duty (Q), Duty (Q), COP Expander Pump Work Flow Rate
(E),% therms/hr therms/hr (=QQ) Work (W),hp (W),hp (M), lbs/hr
100 1.076 1.584 0.446 4.455 1.082 961
90 1.328 1.842 0.361 4.950 1.336 1187
75 1.913 2.442 0.251 5.940 1.924 1709
60 2.989 3.545 0.161 7.425 3.006 2671
50 4.304 4.894 0.112 8.910 4.329 3846

Effects of Compressor/Expander Efficiencies with iso-Propanol
(Refrigeration Duty (QR) = 4 tons = 0.48 therm/hr)

Expander Collector
Efficiency Duty (Q),
(E), % therms/hr
100 0.884
90 1.092
75 1.572
60 2.456
50 3.537

Duty (Q),

COP Expander
(=Q/Q8) Work(W),hp



Hot Loop
Pump Work Flow Rate
(W), hp (M), lbs/hr
0.010 271
0.013 335
0.018 482
0.029 753
0.041 1084

TR 500
COP = TR = 5.882
Tc TR 585 500

E Ts-Tc 660- 585 0.1136
Ts 660

from which COP = (COPc)(Ec) = 0.6684. The best COP
values (at adiabatic efficiency values = 100%) in Tables 4-7
are seen to approach this value.

The selection of the heat rejection temperature (To) is
clearly somewhat arbitrary. The selection of the condenser
temperature (Tc) of 1250F as this rejection temperature is
admittedly a very conservative choice, leading to the poorest
or lowest values for the theoretical COP. As this temperature
is reduced, the computed theoretical COP value improves, as
summarized in Table 8, wherein these values are calculated
for heat rejection temperatures of To = 125, 110, 100, 90,
and 770F. The value of To = 100F, for example, was
chosen by Chen"3' in his analysis. Of course, a common
value for this latter quantity is 77F, particularly in ther-
modynamic availability or exergy analyses."'21

In his more thorough analysis of this cycle, Hamner[21
computes various other theoretical coefficient of performance
values, which are generally less than those from the Carnot
analysis and thus somewhat more realistic. An analysis as-
suming an isentropic turbine-compressor combination with

Chemical Engineering Education

- -

no mixing losses yields COP values not much less than the
Carnot values. Once mixing losses are allowed to affect the
results, using either an ideal gas model or a real gas model,
the COP values drop markedly due to the internal
irreversibilities or lost work. Hamner also reports experi-
mental data on such an ejector-operated refrigeration cycle,
rated at approximately one ton of refrigeration and em-
ploying R-l 1 as the refrigerant. Experimental COP val-
ues of about 0.10 to 0.25 were obtained for pressure
ratios (Ps/Pc) of 5.0 to 7.5.


This article has demonstrated the applicability of the
HYSYS computer-aided process design system to the simu-
lation and analysis of a solar-powered refrigeration cycle.
While such a cycle consists of a number of standard chemi-
cal process equipment items such as heat exchangers, a
pump, and an expansion valve, the key hardware element in
this cycle is a thermal compressor or jet ejector. Models of
the latter item, while a relatively common piece of process-
ing equipment in the chemical and allied industries, are not
that extant in computer-aided process design systems such
as HYSYS or comparable software packages. The em-
ployment of an adjust or control module to balance the
work of a compressor and an expander in a cycle was
illustrated in this work.

The coefficient of performance (COP) values for refrig-
eration cycles driven by a solar collector and jet ejector are
admittedly much smaller than those of conventional cycles
employing mechanical compressors. As numerous authors' -
3] have pointed out, however, applications of the former may
be economical in cases wherein the required input heat is
very inexpensive (e.g., solar energy) or it would be other-
wise wasted, as from the cooling system of an automobile
engine. And there are certainly more than just technological
factors operative in this arena.14 Lastly, it should be remem-
bered that the energy input to a mechanical vapor-compres-
sion refrigeration cycle generally originates from an electri-
cal power plant. This power often derives from the combus-
tion of a fuel with a process efficiency of about 33%. Thus,
the ultimate
TABLE 8 amount of
Influence of Heat Rejection Temperature energy re-
(To) on COP and Efficiency Values quired in
(TR = 400F, Ts = 200F) such a me-
Rejection Efficiency Overall cycle is
Temperature Refrigeration of heat cycle COP rou g
(To), F Cycle (COP), engine (Ec) [=(COP)c(Ec)] r o u g h 1
three times
125 5.882 0.1136 0.6684 the amount
110 7.143 0.1364 0.9740
100 8.333 0.1515 1.2626 actually sup-
90 10.000 0.1667 1.6667 plied to the
77 13.514 0.1864 2.5184 compressor.

Winter 2001


I have written a little book especially designed for the first
engineering thermo course. It is called

Understanding Engineering Thermo

and it uses a radically different teaching approach. Students like
The OSU Bookstore (Box 489, Corvallis OR 97339) is dis-
tributing it at $20 plus mailing cost. If you are a thermo teacher
and want a desk copy, contact me at
Chemical Engineering Department
Gleeson 103
Oregon State University
Corvallis OR 97331
Octave Levenspiel
octave @

Perhaps the major contribution of this work is of a peda-
gogical nature. Thus, this study of a solar-powered refrigera-
tion cycle, exploring different refrigerants, efficiencies,
operating conditions, etc., could represent an excellent
computer-aided design project in an introductory engi-
neering thermodynamics course. It is in this spirit that
this study was formulated.

1. Heymann, M., and W. Resnick, "Optimum Ejector Design
for Ejector-Operated Refrigeration Cycles," Israel J. Technol.,
2, 242(1964)
2. Hamner, R.M., "An Alternate Source of Cooling: The Ejec-
tor-Compression Heat Pump," ASHRAE J., 22(7), 62 (1980)
3. Chen, L.-T., "A Heat Driven Mobile Refrigeration Cycle
Analysis," Energy Conserv., 18, 25 (1978)
4. Pavone, T., and G. Patrick, "Energy Tax Credit Aids Invest-
ment Projects," Chem. Eng., 88(4), 99 (1981)
5. Khoury, F., M. Heyman, and W. Resnick, "Performance
Characteristics of Self-Entrainment Ejectors," I&EC Proc.
Des. Develop., 6, 331 (1967)
6. DeFrate, L.A., and A.E. Hoerl, "Optimum Design of Ejectors
Using Digital Computers," Chem. Eng. Prog. Symp. Series,
55(21), 43 (1959)
7. Holldorff, H.F.W., J.D. Muzzy, and J.T. Sommerfeld, "Digi-
tal Computer Model of a Thermal Compressor," Proc. 1981
Summer Computer Simulation Conf., p. 247, July (1981)
8. Seader, J.D., W.D. Seider, and A.C. Pauls, FLOWTRAN
Simulation: An Introduction," 2nd ed., The CACHE Corp.,
Cambridge, MA (1977)
9. Clark, J.P., T.P. Koehler, and J.T. Sommerfeld, Exercises in
Process Simulation Using FLOWTRAN, 2nd ed., The CACHE
Corp., Salt Lake City, UT (1980)
10. PROIII Keyword Input Manual, Version 4.0, Simulation
Sciences, Inc., Brea, CA (Sept. 1994)
11. HYSIM: Special Features and Application Guide, Version
C2.50, Hyprotech, Calgary, Canada (March, 1994)
12. Kyle, B.G., Chemical and Process Thermodynamics, 2nd
ed., Prentice Hall, Englewood Cliffs, NJ (1992) J





Ben-Gurion University of the Negev Beer-Sheva 84105, Israel
Tel-Aviv University Tel-Aviv 69978, Israel

he application of citation statistics for evaluating and
comparing research groups has continually increased
in recent years. One reason for this trend is that such
statistics are perceived to be quantitative and objective. In
addition, the pertinent information is available through
the Internet and the evaluation can be carried out with
minimal expense.
Unfortunately, there are many pitfalls in using citation
statistics as a sole measure of research achievements. Some
of the pitfalls are mentioned, for example, by Grossman[11
and Angus, et al.121 Those pitfalls can often lead to absurd
results, as was recently shown in letters to the editor by
Braun[31 and Reedijk.[4] The response from the institutes that
provide the citation statistics and the citation analysts
(Blazick,']5 van Raan'6]) state that citation analysis should be
used only as an additional, supporting tool to peer review
and should never be used in "isolation."
In the 1995 NRC report,"' results of a research-related
comparison study of 93 chemical engineering departments
awarding PhDs in the US were reported. In that study, quali-
tative measures of research achievements (such as peer re-
view) and quantitative measures, both intensive (such as the
number of citations per paper, CPP) and extensive (e.g., total
number of publications and citations), were used. But even
such an extensive and thorough study cannot be completely
faultless (as was pointed out by Grossman[11 and Angus, et
al.[2]) because of technical difficulties in collecting reli-
able citation data and the very different nature of the
evaluated departments.
Furthermore, conducting both a thorough peer review and
a quantitative evaluation can be time consuming and expen-
sive. For this reason there is a growing tendency to rely on

quantitative measures alone in evaluating and comparing
research programs. Van Raan[16 has recently evaluated the
validity of a measurement that he calls the "impact of the
group." This measure is essentially the CPP divided by the
CPP world average of the field (WAF). Studying various
departments in The Netherlands, van Raan has found a strong
correlation between quantitative measures, such as the CPP/
WAF and results of the peer reviews. In spite of these
positive results, he does not recommend the use of quantita-
tive measures alone and calls for carrying out further re-
search in cases where considerable differences are detected
between results obtained by employing different measures.
In this paper the influence of the publication profile on the
CPP is investigated. In the next section an example is pre-

Mordechai Shacham received his BSc (1969)
and his DSc (1973) from the Technion, Israel
Institute of Technology. He is currently a profes-
sor of chemical engineering at the Ben-Gurion
University of the Negev, Beer-Sheva, Israel,
where he has served since 1974 at every aca-
demic level, including two four-year terms as
department head. His research interests include
analysis, modeling, and regression of data, ap-
plied numerical methods, computer-aided in-
struction, and process simulation, design, and

Neima Brauner is a professor and head of me-
chanical engineering undergraduate studies in
the Department of Fluid Mechanics and Heat
Transfer at the Tel-Aviv University, Tel Aviv, Is-
rael. She received her BSc and MSc in Chemical
Engineering from the Technion Israel Institute of
Technology, Haifa, Israel, and her PhD in Me-
h chanical Engineering from the Tel-Aviv Univer-
sity. Her research has focused on the field of
hydrodynamics and transport phenomena in two-
phase flow systems.
@ Copyright ChE Division of ASEE 2001
Chemical Engineering Education

...there are...pitfalls in using citation statistics as a sole measure of
research achievements [that] can often lead to absurd results....The response
from the institutes that provide the citation statistics and the citation
analysts state that citation analysis should be used only as
an additional, supporting tool for peer review...

sented where the CPP and CPP/WAF yield results that are
contradictory to the results of other measures of research
impact and productivity. In the third section a simulation
study of the influence of the publication profile on the CPP
is presented and the weakness of the CPP is demonstrated.
Finally, a new "weighted CPP," which is invariant of the
publication profile, is introduced.

In a recent evaluation (carried out in January 2000) of the
Chemical Engineering Department at the Ben-Gurion Uni-
versity of the Negev (BGU), an outside evaluation commit-
tee decided to use CPP/WAF as the only measure of research
productivity and impact. As a basis for the evaluation meth-
odology, the review produced by the Institute for Scientific
Information (published in Science Watch18 in 1992) was
used. In this review, chemical engineering departments that
had published more than a threshold value of 70 papers over
the period of 1984-90 were ranked. The BGU department
was ranked 24th in this list with 70 publications, 250 citations
and 3.57 citations per paper. During the period of that study,
there were twelve faculty members in the department, mean-
ing 1.2 publications per year per faculty (P/Y/F)
The evaluation committee collected data from the Science
Citation Index for the period of 1989-1999. Only citations of
papers published starting in 1989 were considered. The study
found that the total number of publications (cited at least
once) during this period was 397; thus, 2.78 P/Y/F (11 years,
13 faculty). The total number of citations during that period
was 1603; thus, 11.21 per year per faculty (C/Y/F). The CPP
was obviously 1603/397 = 4.04.
According to the evaluation committee, the WAF for the
same period for the field was 5.54, and thus the CPP for the
BGU department was significantly below the world average.
The committee based its conclusions regarding research pro-
ductivity and impact of the department at BGU on the CPP/
WAF alone, disregarding all the other available measures
and relevant information.
In order to assess the validity of the CPP/WAF as a sole
indicator in this case, we compared the P/Y/F and C/Y/F
with similar figures obtained from the 1995 NRC report'71
(Appendix P, p. 500). The publication and citation data in
the NRC report is for a period of five years (1988-1992). The
corresponding values of the P/Y/F and C/Y/F for the depart-

ments included in this report indicate that there were fewer
than ten departments in the US with values higher than 2.78
P/Y/F and 11.21 C/Y/F at the time of that study. Obviously,
there are some significant differences between the study
conducted by the NRC and the one conducted by the evalua-
tion committee at BGU. Still, this comparison strongly im-
plies that the CPP severely undervalues the research produc-
tivity and impact of the department at BGU.
The sharp increase in the publication rate between the
periods of 1984-90 (1.2 P/Y/F) and 1989-99 (2.78 P/Y/F) at
BGU provides a clue to the contradictory results obtained by
using various indicators. The publication profile over the
years may affect the CPP values and this could be the reason
for the poor performance indicated by the CPP. This as-
sumption is supported by the number of publications in peer-
reviewed journals during the last five years, as provided by
the faculty of the BGU chemical engineering department.
The number of publications was 36 in 1995, 53 in 1996, 57
in 1997, 61 in 1998 and 85 in 1999-a sharply increasing
profile indeed.
In order to check the effect of publication rate profile on
the CPP, a simulation study has been carried out. The meth-
odology used by the evaluation committee in the BGU
(evaluation period, papers, and citations included, etc.)
was used as the basis for the simulation model, as de-
tailed in the next section.

The details of this simulation are shown in Table 1. The
simulation covers the period from 1989-99. Only citations of
papers published starting in 1989 are considered. Obviously,
not all the papers have the same, constant citation rate. Also,
the CPP can change with time, with a maximum around the
5th year after publication (Grossman111). In order to isolate
the impact of the publication rate profile on the CPP, how-
ever, a constant citation rate of one citation per paper per
year, starting one year after the year of publication, is as-
sumed in all cases studied here.
Six different cases are considered. In the first case of "stop
publishing," one paper was published in 1989 and none
thereafter. In the second case of "slope: -2," ten papers were
published in 1989 and thereafter the number of papers has
been reduced by two every year, reaching zero publications
per year after 5 years. The additional cases ("slope: -1,"

Winter 2001

"slope: 0," "slope: +1," and ":slope: +2") are simi-
lar, except that the number of papers published per
year changes according to the specified slope.
The results of the simulation are summarized in
the "Total" column of Table 1 and in Figures 1 to
4. Figure 1 shows the total number of publications
at the end of the eleventh year for the various cases.
As expected, the number of publications increases
monotonically with increasing the publication rate.
There are 30 publications for the case of rapidly
decreasing production rate of "slope: -2," 110 pub-
lications in the case of constant production rate
("slope: 0") and 220 publications for the rapidly
increasing production rate of "slope: +2."
The total numbers of citations (see Figure 2)
show a similar trend. There are 10 citations for
case No. 1, 260 for case No. 2, and 880 citations
for the most rapidly increasing publication rate
of case No. 6.
The trend of the CPP values (see Figure 3) is
completely opposite to the trends of the total num-
ber of publications and citations. The CPP is the
highest (CPP=10) for the "stop publishing" case. It
decreases continuously with increasing the slope of
the publication rate, reaching the lowest value
(CPP=4) for the steepest increase of productivity
considered in the case of "slope: +2." The calcu-
lated CPP values can be compared with the true

citation frequency (one citation per paper per year) using CPP values
normalized by the number of citation years (=10). These are shown in
Figure 4. Obviously, there is no difference in the trends shown in
Figures 3 and 4; only the scaling has changed. The value of the normal-
ized CPP is one (as would be expected) only for the first "stop publish-
ing" case. For the more productive research groups, the normalized
CPP is significantly smaller than one, and it keeps decreasing with
increasing productivity.
It is evident from this study that the CPP based on averages, as



2150 -



Stop publ. slope: -2 slope: -1 slope: 0 slope: +1 slope: +2

Figure 1. Total number of publications in the various cases
(time period eleven years).

Year '89 '90 '91 '92 '93 '94 '95 '96 '97 '98 '99
Year no. (1) 0 1 2 3 4 5 6 7 8 9 10

Simulation Study of the Variation of CPP as a Function of Publication Rate
with Constant Value of One Citation per Paper per Year

1 Stop Publication 1 0 0 0 0 0 0 0 0 0 0 1 10
publishing Citations 1 1 1 1 1 1 1 1 1 1 10 10
CPP 10 1.00
2 Slope: -2 Publications 10 8 6 4 2 0 0 0 0 0 0 30 260
Citations 0 10 18 24 28 30 30 30 30 30 30 260 260
CPP 8.67 1.00
3 Slope: -1 Publications 10 9 8 7 6 5 4 3 2 1 0 55 385
Citations 0 10 19 27 34 40 45 49 52 54 55 385 385
CPP 7 1.00
4 Slope: 0 Publications 10 10 10 10 10 10 10 10 10 10 10 110 550
Citations 0 10 20 30 40 50 60 70 80 90 100 550 550
CPP 5 1.00
5 Slope:+1 Publications 10 11 12 13 14 15 16 17 18 19 20 165 715
Citations 0 10 21 33 46 60 75 91 108 126 145 715 715
CPP 4.33 1.00
6 Slope: +2 Publications 10 12 14 16 18 20 22 24 26 28 30 220 880

36 52 70 90 112 136 162 190


0 10 22

880 880
4 1.00

Chemical Engineering Education

Case Description

Total Weighted

Stop publ. slope: -2 slope: -1 slope: 0 slope: +1 slope: +2

Figure 2. Total number of citations in the various cases
(time period eleven years).

Figure 3. The CPP values in the various cases
(time period eleven years).

a Normalized E Weighted I




. 0.6



Figure 4. Normalized and weighted CPP values in the various
cases (time period eleven years).

Winter 2001

6 400
z 300

calculated in this study (and in Science Watch,'81 for
example) does not represent the "impact" or research
quality, since the impact is actually the same in all the
cases studied here-one citation per paper per year. It
definitely does not represent research productivity,
since increase of productivity actually reduces the
CPP. In the next section a modification of the CPP
that eliminates the influence of the publication pro-
file is proposed.

When the CPP is calculated for a given time period
(as in the example presented), the papers published at
the beginning of the period have a higher chance to be
cited than papers published in later years. This inequal-
ity can be eliminated by assigning to each publication a
weight according to the feasible number of years for its
citation. Accordingly, in an evaluation that considers a
specified time period of n years, a particular paper that
was published in the i' year is assigned a weight of (n-
i). To calculate the weighted CPP (WCPP), the total
number of citations is divided by sum of the weighted
publications (each publication is multiplied by its cor-
responding weight).
In the last column of Table 2 the calculation of the
WCPP for the various cases is shown. In case No. 1
there is one publication in the first year, which should
be included with the weight of (11-1)=10. Since there
are ten citations of this publication the resulting WCPP
is one. For the "slope: -2" case, there are ten papers in
the first year with a weight of ten, eight papers in the
second year with a weight of nine, and so on. The sum
of the weighted publications is:
(10*10+8*9+6*8+4*6+2*5) = 260
Since this is equal to the total number of citations, the
value of the WCPP is also one in this case.
In Figure 4 the value of WCPP is presented together
with the normalized CPP values. It can be seen that
WCPP obtains the expected value of one for all the
cases, in contrast to the normalized CPP, which obtains
different values for the various cases. Thus, in the
WCPP the influence of the publication profile on the
citation statistics is eliminated, and it correctly reflects
an "impact" of a single citation per paper per year.
Therefore, the WCCP is much more appropriate to
represent "impact" of research than the CPP.

Using a simulation study, it has been shown that a
CPP- based comparative evaluation of research groups
of different publication profiles may yield absurd re-

Continued on page 45.





slope: -1 slope: 0 slope: +1 slope: +2

slope: +2

Stop publ. slope: -2 slope: -1 slope: 0 slope: +1

Stop publ. slope: -2

n safety




The University of Iowa Iowa City, Iowa 52242-1219

Apathy towards chemical process safety in the United
States came to an abrupt end following the toxic
methyl isocyanate (MIC) release from the Union
Carbide India Ltd. pesticide plant in Bhopal, India, on De-
cember 2, 1984.[I'21 The resulting MIC spread over a heavily
populated area and resulted in the death of thousands. This
incident led to passage and implementation of the Emer-
gency Planning & Community Right-To-Know Act of 1987
and the chemical process safety amendments to the Clean
Air Act of 1990 in the United States. In addition, the Center
for Chemical Process Safety (CCPS), which is affiliated
with the American Institute of Chemical Engineers, was
founded in 1985 in response to the Bhopal incident. The
CCPS is committed to developing engineering and manage-
ment practices to prevent or mitigate the consequences of
catastrophic events at chemical plants.
Recognizing the need for educating undergraduate chemi-
cal engineering students, CCPS, in a cooperative effort with
engineering schools, initiated the Safety and Chemical Engi-
neering Education (SACHE) program in 1992. SACHE pro-
vides teaching materials (e.g., slide/lecture sets, video lec-
tures, problem sets, and instructional modules) to aid educa-
tors in incorporating safety into undergraduate chemical en-
gineering programs. The SACHE instructional materials can
either be used to incorporate safety into existing chemical
engineering courses, e.g., using safety related problems from
the SACHE problem sets, or as supplementary material for a
dedicated chemical process safety course.
At the University of Iowa we have a dedicated required
chemical process safety course that is taken by students
during their junior year. We also incorporate problems from
the SACHE problem sets into other chemical engineering
courses. We believe that a dedicated chemical process safety
course is highly desirable since it (i) allows coverage of
material that would not fit well into existing chemical engi-

neering courses, (ii) emphasizes the importance of chemical
process safety, (iii) reinforces the importance of all chemical
engineering fundamentals (e.g., thermodynamics, reaction
kinetics, transport, and material and energy balances) to
chemical process safety, (iv) provides an excellent review of
the other chemical engineering courses, and (v) better pre-
pares students for industrial employment. Major aspects of
the chemical process safety course offered at the University
of Iowa are summarized in Table 1. Additional approaches
to incorporating chemical process safety into the chemical
engineering curriculum can be found in the literature.13'4
The laboratory experiments listed in Table 1 give students
hands-on experience with such issues as flammability limits,
flash points, electrostatics, runaway reactions, explosions,
and relief design. All of these issues are important in indus-
trial processes. It is well known that most students are more

Brian D. Dorathy graduated from the University of Iowa in 1999 with a BS
degree in Chemical Engineering. He was an undergraduate teaching
assistant in the Chemical Process Safety course offered at the University
of Iowa during the Spring 1999 semester.
Jamisue A. Mooers graduated from the University of Iowa in 1999 with a
BS degree in Chemical Engineering. She was an undergraduate teaching
assistant in the Chemical Process Safety course offered at the University
of Iowa during the Spring 1999 semester.
Matthew M. Warren graduated from the University of Iowa in 2000 with a
BS degree in Chemical Engineering. He was an undergraduate teaching
assistant in the Chemical Process Safety course offered at the University
of Iowa during the Spring 2000 semester.
Jennifer L. Mich graduated from the University of Iowa in 2000 with a BS
degree in Chemical Engineering. She was an undergraduate teaching
assistant in the Chemical Process Safety course offered at the University
of Iowa during the Spring 2000 semester.
David W. Murhammer is Associate Professorin the Department of Chemi-
cal and Biochemical Engineering at the University of Iowa. He received
his BS degree in Chemistry and his MS degree in Chemical Engineering
from Oregon State University, and his PhD degree in Chemical Engineer-
ing from the University of Houston. He developed the Chemical Process
Safety course offered at the University of Iowa and has taught it annually
since its inception in 1996. His research interests include insect cell
culture and bioreactor monitoring.

Copyright ChE Division of ASEE 2001
Chemical Engineering Education

Details of Dedicated Chemical Process Safety Course
(Additional details: )

E Major Topics Covered
Government Regulations
Process Safety Management
Industrial Hygiene
Source Models
Dispersion Models

Fires and Explosions
Fire and Explosion Prevention
Relief Design
Hazard Identification
Risk Assessment/Reliability Engineering
Case Studies
Inherently Safer Design

E Textbooks
Daniel A. Crowl and Joseph F. Louvar, Chemical Process Safety: Fundamen-
tals with Applications, Prentice Hall (1990)
Safety, Health, and Loss Prevention in Chemical Processes: Problems for
Undergraduate Engineering Curriculum, AIChE Center for Chemical Process
Safety (CCPS) (1990)
1 Safety Essay
Students write an essay (1500-word maximum) regarding a topic relevant to
chemical process safety consistent with the format requested for the SACHE-
sponsored Student Essay Award for Undergraduate ChE Students.
] Project
Group projects are performed in which students analyze safety issues rele-
vant to the process of manufacturing the specific chemical assigned to them.
A major aspect of the final report and oral presentation regards how inherently
safer design concepts can be incorporated into the existing process.
E Homework
There are weekly homework assignments.
] Quizzes
There are weekly quizzes (generally 10-20 minutes). These seem to improve
the learning process and to discourage student procrastination.
U Exams
There are one or two midterm exams and a final exam.
I Laboratory Experiments
Flammability Characteristics Electrostatics Runaway Reactions/ Relief
Design Explosions/Relief Design

receptive to this type of hands-on learning than the tradi-
tional lecture format.15' Therefore, this laboratory benefits
all students by giving them a hands-on appreciation for
concepts introduced in the lecture and is especially benefi-
cial to the majority of students who are more receptive to
active learning approaches. While these experiments are
ideally suited to supplementing a chemical process safety
course, they can also be used in other courses, e.g., unit
operations laboratory, to demonstrate important chemical
process safety concepts.

The laboratory experiments discussed in this section are
now an integral component of the chemical process safety
course offered at the University of Iowa. The major equip-
ment used, along with other key aspects of the experi-
ments, are summarized in Table 2. In order to maximize
the learning benefits, these experiments are conducted in
small groups (typically 3 students) soon after discussing
the corresponding topic in the lecture portion of the course.
The laboratory reports consist of a title page, an abstract,
an introduction, materials and methods, results and discus-
sion, conclusions, references cited, and an appendix.
The introduction includes an overview of the purpose of
the experiment and discusses the relevant theory. The ma-
terials and methods section discusses the procedures used,
what data were collected and how, and diagrams of the
apparatuses used. The results and discussion section in-
cludes a discussion of the results obtained and their rel-
evance to industrial applications, including a comparison
of theory and experimental results, and answers to all ques-
tions asked about the experiment. The appendix includes
MSDS sheets for all chemicals used, sample calculations,

Summary of Equipment
Additional comments and equipment costs (1999 estimates) are also given. (Prices are for Advanced Reactive System Screening Tool (ARSST), successor to the RSST.)

Experiment Major Equipment Used Comments
Flammability Miniflash FLP Flash (Model 4200. Petrolab Company. Latham, NY: -$12.000) The flammability limits experiment also uses a personal compu-
Characteristics Flame Tec Flammability Limits and Data Acquisition System (Fauske and Associates, ter for data acquisition and gas cylinders with gauges to charge
Burr Ridge, IL; -$24,000) the apparatus with the desired amounts of propane and air. In
addition, a barometer can be used to correct for deviations from
atmospheric pressure.

Electrostatics Electrometer with accessories for high voltage (Model 6514, Keithley Instruments, Inc., These experiments are generally simple and require a few
Cleveland, OH; -$4,000). Accessories include a Model 6103C Voltage Divider Probe, additional accessories commonly available in laboratories, as
Model 6171 2 log to 3 log Triax Adaptor, and Model 6102 Triax to UHF Adaptor. described in the main text
Static Monitor (JCI-140. John Chubb Instrumentation. Cheltenham, England; -$1,000)
Van de Graaf Generator (Stock # CR52-587, Edmund Scientific, Barrington, NJ: -$500)
Runaway Reactive System Screening Tool (RSST) (Fauske and Assc., Burr Ridge, IL; -$19,600) These experiments also require a computer for data acquisition
Reactions/ Vent Sizing Software Program (Fauske and Associates; -$2,500) and for running the vent sizing software. In addition, a nitrogen
Relief Design gas cylinder and regulator are needed to pressurize
the RSST unit.

Explosions/ Flame Tec Flammability Limits and Data Acquisition System (Fauske and Assc.. The flammability testing apparatus requires a computer for data
Relief Design Burr Ridge. IL: -$24,000) acquisition. The modified Hartmann tube does not require any
Modified Hartmann Tube (Type MP-5, Adolf Kiihner AG. Switzerland: -$12,500) accessories other than the test materials.

Winter 2001 3;

and a copy of the raw data. Experimental write-ups
given to the students at the University of Iowa can be
found on the Chemical Process Safety course web
site (see Table 1).

Flammability Characteristics

Prior to conducting these experiments students
should be introduced to the importance of flash points
and flammability limits. In addition, they should un- I
derstand the relationship between these two param- 3
eters, i.e., the flash point of a pure substance is the
temperature at which the concentration of flammable
vapor in the gas phase is equal to the lower flamma-
bility limit (see Figure 1). These experiments will
acquaint the students with methods used to deter-
mine flash points and flammability limits.
In our experiments we collected flash points for
pure alcohols (methanol, ethanol, and 1-butanol), alcohol/
water mixtures, and alcohol/alcohol mixtures. Data were
collected with a Miniflash FLP Flash Point Tester (Table 2),
an automated instrument that determines closed cup flash
points. A manual Pensky-Marten flash point tester (Table 2)
was also used to observe when a substance flashes. Data
obtained from the automated instrument were used for analy-
sis. Note that the data should be corrected for deviations
from atmospheric pressure.
First, the measured flash points (FPs) for the pure alcohols
were compared with accepted literature values.[61 All of the
experimental values were within 2'C of the accepted litera-
ture values, with the ethanol measurement typically within
0.50C of the literature value (Table 3).
Second, the flash points were measured for different con-
centrations of alcohol/water mixtures. These values were
first compared with values calculated assuming an ideal
solution, i.e., determining the temperature at which the va-
por pressure of the flammable component (alcohol) is equal
to that of the pure component vapor pressurei7 at its flash
point[6 81 (Table 3). The experimental values for the alcohol/
water mixtures were consistently several degrees lower than
the values obtained assuming ideal solutions. The largest
difference between the experimental and calculated values
occurred for the mixture containing 33.3 mole% 1-butanol,
for which the calculated value was about 140C higher. It
should be emphasized, however, that in reality these solu-
tions are nonideal. Therefore, the FPs were also calculated
taking into account nonideal behavior. Briefly, liquid-vapor
equilibrium data191 for the alcohol/water mixtures were used
to calculate activity coefficients (y, =Ptoti / psatxi) at the
corresponding liquid concentration (x,). The FP was then
calculated assuming that the activity coefficient is a function
of concentration only. As shown in Table 3, the correspond-
ing calculated values are all within 20C of the measured

Flash Poit



Figure 1. Diagram demonstrating the relationship between
the flash point and lower flammability limit.

Typical Flash Point (FP) Results
The experimental FPsfor the pure alcohols were compared with litera-
ture values, while the experimental FPsfor the alcohol/water mixtures
were compared to calculated values assuming either ideal or nonideal

Typical Calculated FP (IC)
Material Tested (mole fraction) Experimental Ideal Nonideal
water methanol ethanol 1-butanol FP/(C) Solution Solution
0.000 1.000 0.000 0.000 9.0 11* ---
0.000 0.000 1.000 0.000 13.4 13*
0.000 0.000 0.000 1.000 38.6 37*
0.667 0.333 0.000 0.000 27.5 31.3 26.4
0.500 0.500 0.000 0.000 19.4 23.3 21.2
0.333 0.667 0.000 0.000 16.8 18.0 17.3
0.667 0.000 0.333 0.000 24.1 31.6 23.0
0.500 0.000 0.500 0.000 20.1 24.4 20.9
0.333 0.000 0.667 0.000 17.4 19.5 18.3
0.667 0.000 0.000 0.333 40.9 54.7 43.9
0.500 0.000 0.000 0.500 43.2 47.7 43.6
0.333 0.000 0.000 0.667 39.6 43.1 41.4
0.000 0.333 0.000 0.667 23.5 20.5
0.000 0.500 0.000 0.500 18.5 16.1
0.000 0.667 0.000 0.333 15.2 12.5
0.000 0.000 0.333 0.667 25.1 22.6
0.000 0.000 0.500 0.500 21.4 19.0
0.000 0.000 0.667 0.333 18.0 15.9
Literature values obtainedfrom reference 6.

values, with the sole exception of the 1-butanol (x = 0.333)/
water solution, for which the calculated FP was 3C higher
than the measured value.
Third, the FPs were measured for different concentrations
of alcohol/alcohol mixtures, specifically, methanol/ -butanol
and ethanol/l-butanol mixtures. These measured values were
compared to the theoretical values obtained from the inter-
section of the vapor pressure and lower flammability limits

Chemical Engineering Education


curves on a plot of flammable vapor concentration versus
temperature (Figure 1). For a given liquid phase concentra-
tion the vapor phase concentration of the alcohols can be
calculated by assuming an ideal solution, i.e., by using
Raoult's Law. The lower flammability limit (LFL) of the
mixture can be estimated by

LFL=mix = (1)

where LFLm,x is the LFL for the mixture, y'i is the mole
fraction in the vapor phase of component i on a combustible
basis, and LFL, is the LFL of pure component i at 25'C.[8
Thus, it is assumed that LFLmix is not a function of tempera-
ture, i.e., it is a constant in Figure 1. This is a good assump-
tion for the experiments discussed in this paper since the
temperature effect on LFL is small over the relevant tem-
perature ranges.18] It should also be noted that there is some
controversy regarding the equation commonly used to esti-
mate the effect of temperature on the LFL of a pure sub-
stance.[8,10 The "concentration of flammable vapor" given in
Figure 1 is the sum of the vapor phase mole fractions of all
of the alcohols. The calculated values given in Table 3 are
generally 2-30C lower than the measured values for both
alcohol mixtures. It is possible that the calculated values
would agree more closely with the measured values if treated
as nonideal mixtures. It is suggested that the students at-
tempt to develop other methods of estimating the FPs of
alcohol mixtures. In addition, the fact that the predicted FPs
agree so well with the data may lead to overconfidence in
this method by students. It is known that this method works
satisfactorily at room temperature and pressure, but degrades
considerably at increased temperature and pressure. Thus,
this approach is applicable for product or raw material ship-
ping (done at ambient temperature and pressure), but less
applicable for process conditions.110]
The lower and upper flammability limits of propane were
determined using a Flame Tec flammability limits and data-
acquisition system (Table 2) that collects pressure-time and
temperature-time data. Briefly, the FlameTec apparatus is
charged with a known mixture of air and propane and igni-
tion with a spark is attempted. If the mixture ignites, i.e., if
an appreciable rise in pressure and temperature occurs, then
it is within the flammability range. This process is repeated
with a variety of pro-
pane concentrations to
TABLE 4 determine the flamma-
Flammability Limits Results ability range. Typical
Typical Literature LFL and UFL values
Result Value' obtained using this pro-
Propane LFL 2.0- 2.6% 2.1% cedure are compared
Propane UFL 10.4- 10.9% 9.5% with literature values,[6]
as shown in Table 4.

The experimental LFL value is consistent with the literature
LFL, while the UFL literature values is approximately 1.0%
higher than the experimental UFL range. More precise re-
sults could be obtained by using smaller incremental partial
pressures of propane. In addition, it should be noted that
flammability results can vary slightly based on the ignition
Once the LFL for a substance is known, its minimum
oxygen concentration (MOC) can be estimated using

( Moles 02 (
MOC= LFL Moles Fuel) (2)

where LFL is the lower flammability limit, and the ratio is
the stoichiometric ratio of oxygen to fuel for complete com-
bustion of the substance.181
Students were expected to address the following in their
I Define the following terms and discuss how they are
relevant to the experiment and industry: LFL, UFL,
MOC, and flash point.
Calculate the LFL, UFL, and MOC of propane and
compare to literature values.
How do the flash points obtained in the laboratory
compare to expected values?
0 Under what conditions would the results from this
experiment not represent actual process results?


A series of experiments was conducted to provide stu-
dents with an improved understanding of the electrostatic
charge accumulation that can occur in industrial processes.
Most of these experiments were suggested by Crowl101 and/
or Liittgens and Glor.'"1 A high-impedance electrometer
(Table 2) was used to measure electrostatic potential (V) and
charge accumulation (Q). A high-impedance instrument is
required for these experiments since electrostatics is a low
current-high voltage phenomenon. Given V and Q, the accu-
mulated energy (J) and capacitance (C) can be calculated
J=QV/2 (3)
C=Q/V (4)
One set of experiments was conducted to demonstrate
electrostatic charge and/or voltage accumulation during the
transport of materials. First, the phenomena were measured
on a metal beaker isolated from ground (using a piece of
teflon or styrofoam) into which substances were added in a
variety of ways. Specifically, (i) water or a powder (e.g.,
cornstarch) was allowed to free-fall through air into the

Winter 2001

beaker, (ii) water or a powder was poured down a plastic
surface into the beaker, and (iii) water or a powder was
poured down a metal surface into the beaker (see Fig. 2a).
Second, the voltage and/or charge buildup resulting from
recirculating diesel fuel through a metal filter (7-micron
mesh size) isolated from ground was monitored (Fig. 2b).
Third, experiments were conducted in which the accumula-
tion of voltage and/or charge resulting from the agitation of a
liquid in a metal vessel isolated from ground was monitored.
Monitoring the dissipation of the accumulated charge fol-
lowing grounding of the vessel provides the means to evalu-
ate a liquid's relative ability to dissipate charge (conduc-
tive liquids versus non- and semi-conductive liquids).
This experiment was conducted with both tap water and
diesel fuel, which are conductive and semi-conductive
liquids, respectively.
The results of the above experiments can be analyzed
qualitatively or quantitatively, or a combination of both. For
example, monitoring charge and voltage accumulation for a
given experiment can be used to calculate energy accumula-
tion and capacitance, using Eqs. (3) and (4), respectively.
The basic principles involved, however, can be sufficiently
understood with a qualitative analysis. The dropping water/
powder experiments were performed to compare different
contact methods and the relative conductivities of different
materials. Water poured down a metal sheet dissipates po-
tential difference effectively due to the conductive nature of
both materials. Water poured through air or down plastic
demonstrates that when at least one material is a poor con-
ductor (air or plastic), a charge separation and potential
difference are observed. Since cornstarch is a nonconductive
powder, charge separation is observed regardless of what
additional materials or contact methods are involved. Recir-
culation of diesel fuel through an ungrounded metal filter
demonstrates how contact between a semiconductive liquid
and an improperly bonded and grounded filter can easily
result in charge separation. Similar phenomena are observed
in the agitation experiments, which directly contrasts the
dissipative qualities of water and diesel fuel and suggests the

S ---- Water or Powder
0 Metal or Plasbc Sheet I nlne Filter

--- a, | Pump


Insulating Block
Figure 2. Experimental apparatus for demonstrating
electrostatic charge accumulation in (a) pouring liquids
or powders, and (b) recirculating liquids.""'"

importance of understanding the conductivity of various
chemicals present in a given industrial process. An addi-
tional experiment could involve the addition of an antistatic
agent to the diesel fuel in the recirculation experiment in
order to observe ways to alleviate charge accumulation.
A second set of experiments was conducted to demon-
strate charge accumulation on humans and the potential haz-
ards thereof. First, the potential difference caused by remov-
ing a sweater or jacket and by using a Van de Graaf genera-
tor (Table 2) was measured and used to calculate the energy
and capacitance using Eqs. (3) and (4), respectively. In addi-
tion, in each of these cases it can be determined whether
enough energy is generated to ignite propane (see Fig. 3). In
general, any charge accumulation exceeding 350 volts and
0.1 mJ is considered dangerous in industrial operations where
flammable vapors are present.[81 It should be emphasized
that these experiments work best when conducted under low

Figure 3. Experimental setup to demonstrate that
electrostatic charge accumulation in humans stores
sufficient energy for propane ignition.1111

Figure 4. Experimental setup to demonstrate a
propagating brush discharge.'"I
Chemical Engineering Education


Grounded Electrode


Propane Torch

Metal Sheet
- Insulating Block

humidity conditions, and when the insu
effective in isolating the lab participant fro
particularly important for the Van de Graa
Finally, a propagating brush dis-
charge was demonstrated in a third ex-
periment. This provides a dramatic vi-
sualization of releasing charge buildup.
A propagating brush discharge is an
energy-rich energy discharge in which
a highly charged insulating surface (e.g.,
a film) is backed with a grounded con-
ductor. The "feathery" discharge shown
in Figure 4 is characteristic of this type
of discharge.JI21 The demonstration uses
a thin sheet of overhead projector plas-
tic, a sheet of metal (approximately 1/4
insulating block. Each of these listed mater
mately 1 foot by 1 foot. The three materials
shown in Figure 4a. The metal sheet was
insulated metal wire to the Van de Graaf g
connections were secured with electrical t
two electrodes with sharp tips were grou
lated metal wire.
The experiment was initiated by turning
Graaf generator, which resulted in a sep
between the metal slab (which becomes p
and the overhead plastic (which becomes ne
While this charge separation is occurring, ol
electrodes is held at a distance of appro
from the overhead plastic. Results are be
trode is moved laterally across the entire
sheet while the Van de Graaf generator is
While the charge separation is occurring
should look and listen for sparks along
particularly at sharp corners. These sparks
charge dissipation and can diminish expe
An easy solution to this problem is to line
metal sheet with electrical tape prior t
Charge separation on the experimental s
proceed for approximately one minute, a
Van de Graaf generator is turned off. T
electrode is centered over the plastic sheet
of approximately 1 cm), and the second gi
is brought into direct contact with the e
sheet, as shown in Figure 4b. At this point,
grounded, leaving a large charge imbal
plastic sheet and the metal sheet, resulting
brush discharge through the first ground
success of this experiment depends on dev
tent routine; therefore, it is most effective a
instead of having each lab group attemf
(though if time allows, students should a
Winter 2001

lating material is
m ground. This is
f experiment.

the experiment after the initial demonstration is complete).
Propagating brush discharges can occur with a voltage
difference of 200 kV per meter of distance between the

The experiments .. .give students hands-on
experience with such issues as flammability limits,
flash points, electrostatics, runaway reactions, explosions, and
relief design.... While these experiments are ideally suited to
supplementing a chemical process safety course, they
can also be used in other courses, e.g., unit operations
laboratory, to demonstrate important
chemical process safety concepts.

'" thick), and an surface where a charge imbalance exists and the potential
rials was approxi- dissipation route (e.g., the first electrode in the discharge
were arranged as experiment above). A simple experiment to investigate this
connected via an maximum potential difference involves unraveling a roll of
enerator (all wire overhead projector plastic. A field meter (see Table 2) or
ape). In addition, high-voltage probe used with a high-impedance electrom-
inded using insu- eter (Table 2) is placed near the expected trajectory of the
unrolling plastic. This distance should be estimated for cal-
g on the Van de culation purposes. The end of the rolled plastic should be
aration of charge pulled quickly and with some force in order to observe a
positively charged) significant amount of unraveling, and the maximum voltage
gatively charged). reading should be recorded. Results for the experiment var-
ne of the grounded ied between lab groups, but as much as 194 kV/m potential
ximately one cm difference was developed in some runs.
st when the elec- After completing the above experiments, students were
area of the plastic expected to answer the following questions in the results and
running. discussion section of the report based on data obtained from
,the experimenter the laboratory experiments:
the metal sheet, Describe the phenomenon of charge separation be-
lead to premature tween two materials in contact.
rimental success. What are the important material properties being con-
e the edges of the sidered in the experiment (i.e., how does one material
o conducting the vary from another, and how do these differences affect
experimental results)?
etup is allowed to > Relate each experiment to a typical situation in indus-
it which time the try, and discuss the difference of magnitude between
he first grounded industrial and lab scale scenarios.
(still at a distance What are other ways that charge accumulation can
rounded electrode occur in an industrial environment?
:dge of the metal
So t m What can be done to prevent and/or counteract charge
the metal sheet is bui
ance between the
in a propagating
ed electrode. The Runaway Reaction/Relief Design
eloping a consis-
s a demonstration This experiment provides the students with an understand-
,t the experiment ing of how to characterize a runaway reaction and how to
attempt to recreate collect and use data to size a relief vent for an industrial scale

reactor of a given volume and given charge of reactants. To
demonstrate these principles, the reaction of 2 moles of
methanol with 1 mole acetic anhydride, i.e.,

2CH30H +O(COCH3)2 2 CH3COOCH3 +H20 (5)

is investigated using a calorimeter (Reactive System Screen-
ing Tool [RSST], Table 2).113 Briefly, methanol and acetic
anhydride are charged into the RSST and the temperature is
increased at a constant rate of 2'C/min. Once an onset tem-
perature is reached, the exothermic reaction self-heats and a
runaway reaction occurs. Two separate tests are conducted:
the first is conducted at a low pressure (representing the set
pressure of a field scale vessel) and the second is conducted
at a high pressure (representing the maximum allowable
working pressure of a field scale vessel). The actual pres-
sures used for these tests can be changed according to the
type of vessel that is being simulated, and they also depend
on the pressure of nitrogen available. Results included in this
paper use pressures of 15 and 150 psig for the low- and high-
pressure tests, respectively. The RSST, interfaced with a
computer, collects temperature-time and pressure-time data.
When the reaction "takes off," a rapid increase in the tem-
perature and pressure are observed, as demonstrated in Fig-
ure 5 for the methanol/acetic anhydride system at an initial
pressure of 15 psig. It should be noted that, ideally, the low-
pressure test should be conducted at constant pressure for
the duration of the experiment (i.e., pressure should be moni-
tored and relieved under a chemical fume hood when neces-
sary in order to maintain a vessel pressure consistent with
the chosen set pressure of the relief device).
In the data presented above pressure was not regulated,
which affects the observed tempering temperature. These
pressure and temperature data can be used to generate heat
rate and pressure rate curves for relief vent sizing by hand
calculations or by the Vent Sizing Software Program (VSSP,
Table 2). Whether heat-rate or pressure-rate data are used in
vent-sizing calculations depends on the type of runaway
reaction that occurs (vapor, gassy, or hybrid)."'3I In a "vapor"
runaway reaction, the vapor is produced by components
vaporizing due to the heat of reaction; in a "gassy" runaway,
the gas is a product of the reaction; and in a "hybrid" reac-
tion, both phenomena occur. For the vapor system of metha-
nol and acetic anhydride, heat-rate curves for both the
high- and low-pressure tests are sufficient. Figure 6 shows
typical heat rate results for the methanol/acetic anhy-
dride reaction system.
Two calculation methods for vent sizing were considered.
The hand calculation was performed using

A=1.5x10-5 m (6)
FPs (6)

where A (m2) is the vent area, m (kg) is the charge of
reactants to the field scale vessel, Ts (C/min) is the self-


80 20

40 10

20 5
0 0
0 10 20 30 40
Time (min)

Figure 5. Typical temperature and pressure data for
the methanol/acetic anhydride reaction in the
RSST (15 psig).

1.8 Low Pressure Data A
SHigh Pressure Data
1.4 /
I 1
2 0.8
0 50 100 150 200
Temp (OC)
Figure 6. Typical heat rates for low (15 psig) and high
(150 psig) pressure obtained for the methanol/acetic
anhydride reaction in the RSST.

Summary of Typical Vent-Sizing Results
for the Methanol/Acetic Anhydride System

Tempering Rate (at Set Reactant VentArea VentArea
Temperature Tempering) Pressure Charge Hand VSSP
(C) ("C/min) (psia) (kg) Calc. (in) (in2)
100.5 20.54 29.7 1500 24.1 11.4

heat rate due to exothermic reaction at the specified set
pressure P, (psia), and F is the flow reduction factor that
accounts for piping connected to the relief vent (venting
directly to atmosphere has a flow reduction factor equal to
one).'31 Equation (6) assumes two-phase flashing flow and
assumes 20% overpressure (absolute) during venting. The

Chemical Engineering Education

second calculation method used the VSSP. The theory behind VSSP
calculations and Eq. (6) is similar and leads to similar results. Figure 7
shows typical VSSP input for the methanol/acetic anhydride system.
Inputs such as set pressure, back pressure, vessel volume and reactant
charge are defined in the laboratory objectives (and can be modified for
each lab group). Liquid specific heat and density are estimates based on
reactant composition at set pressure. Vapor density does not play a
significant role in calculations, thus it is assigned a value of 1 lbm/ft3.
The Clausius-Clapeyron equation is used by the VSSP when latent heat
is set equal to zero, and cross-sectional area and surface tension infor-
mation are not required for homogenous vessel venting (assumed here).
Pressure/temperature data are estimates based on Raoult's law, and dT-
dt data points are taken from the graphs generated in Figure 6. Percent
overpressure (relative in the VSSP) values are entered arbitrarily, but
care should be taken to include an overpressure that corresponds to
the hand calculation that assumes 20% absolute overpressure. One
of the advantages of using the VSSP software is that plots can be
generated, including overpressure vs. vent area and vent area vs.
maximum pressure.
The VSSP calculates both a vapor vent size and an ideal vent size;
only the former is used for comparison with the hand calculations. Table
5 summarizes the results obtained from the RSST data in Figures 5 and 6
for a field scale vessel with a volume of 1000 gallons and a methanol/
acetic anhydride charge of 1500 kg (2:1 mole ratio).
The students compared and discussed the results they obtained from
the hand calculation and the VSSP. Typical results for hand calculations
for this system (1500 kg reactants, 1000 gal vessel) ranged between 17
and 24 in2. VSSP results for the same system ranged between 11 and 13
in2. Error in the above results can be attributed to a number of factors.
For example, self-heat and tempering values are read visually from
charts, and are susceptible to some error. Also, the fact that the low-
pressure experiment was not conducted at a constant pressure introduces


set 140U1 1111ft 77=7.
Back Retwe
wvemal ds 18 Im 54

LqudSpedif heatI
Losa Dwlffy F7l
VdPa Demu%

%enal X-SecbmndAM j u UIbJ
Sdolace Teraq I _05




J/4 ,

ReWe BlehaIm I y
.4 r.L .9 U r Jr.

a R eaviD*Irqx
r rp'WWan sn a Q'
rRwoviNReachm DdaiMr--- -
r 1.0pa.,01 d74* data JXWA@
r ZwothaorhKkx hwans

;dzd2F,, Cini T2 r211005 C

seTe e o'Pervlr--
I parop dP-TdLapoMt 1 | 711,1 5 | 9 | .. : 1 nrUi 62 ) a 40.00 10 80.0-
P1 N4hIm22 N irP E.l NJm2 3 i 1 ::1 .o 11 9000
TI,-, C 12 f c 4 e.l ooB 12 100

Figure 7. Typical VSSP input data for the methanol/acetic
anhydride system.
Winter 2001

error in the tempering temperature, and subse-
quently the self-heat rates. The VSSP and hand
calculations both use the original system pressure
of 15 psig in the calculation.
The student can also complete a kinetics analy-
sis of the RSST data (this is especially applicable
to a course in chemical reaction engineering).
First, the students can determine the activation
energy using the straight-line data shown in Fig-
ure 6. Second, given the kinetics of the reaction,
the students can calculate the temperature at which
the runaway reaction should begin and compare
their calculated results with the experimental data.
The students are expected to answer the follow-
ing questions in the results and discussion sec-
> What type of system is the methanol/acetic
anhydride reaction (vapor, gassy, or
hybrid)? How are the data collected used to
determine this?
I What is the tempering temperature of the
O Which relief sizing results (hand or VSSP
calculation) should be used? Why?
What characteristics of the RSST design
might prevent the results from being used
for a larger reactor vessel?
> Will a larger vessel runaway faster or
slower than the RSST? Why?
Define the following terms and relate each
to the RSST experiment: set pressure, back
pressure, and maximum allowable working

Explosions/Relief Design

The purpose of this experiment was to demon-
strate the principles of dust and gas explosions.
First, a Modified Hartmann Apparatus (Table 2)
was used to demonstrate dust explosions and to
aid in understanding the dust classification sys-
tem. Briefly, powder is loaded into a tube and a
continuous spark is created from two electrodes

Dust Explosion Results

Powder Output Visual Observation
Flour 0 or I Explosion
Cornstarch I or 2 Explosion
Baking Soda 0 No Explosion

I- VSSP Giah I RD Pw Swi T SRN Ppe S*

placed approximately one-half inch apart. Compressed air
at seven bar (gauge) is suddenly released, which suspends
the powder inside the tube and over the spark. An explosion
(i.e., a sudden flash of fire) occurs under the proper condi-
tions. Measurements are taken during the explosion and the
output from the control box is 0, 1, or 2. If the output is a 1,
then the dust is classified as a St-1 class dust. Anything other
than a 1 means that more sophisticated equipment is needed
to determine the classification. Dusts are classified into St
classes based on their deflagration index, Kst. The Ks, for a
dust increases as the robustness of its explosion increases.181
Three powders were tested-flour, cornstarch, and baking
soda, in the amount of 1200 mg each. Results are summa-
rized in Table 6.
The following questions were answered in the discussion
part of the report.
1 When do dust explosions occur?
What characteristics must a dust have to be explosive?
What are the typical lower and upper explosion limits
for dusts?
What were the physical differences between the three
powders and how did these attribute to whether or not
they created an explosion ?
Describe the classification system of dust explosions.
The second part of the experiment used the Flame Tec
flammability limits and data acquisition system (Table 2) to
examine gas phase explosions. Pressure-time and tempera-
ture-time data were collected for a propane explosion. A
propane concentration of 4.5% was used, which is equiva-
lent to a partial pressure of 0.7 psi propane in the vessel. The
pressure-time data collected can then be used to calculate the
gas deflagration index K, for propane using the "Cubic
Kg = (dP/dt)axV"3 (8)
where V is the vessel volume and (dP/dt)mx is the maximum
rate of pressure increase. It should be noted that Kg for a
given material is dependent on several factors, including the
composition of the material, mixing in the vessel, the vessel
shape, and the energy of the ignition source.18 The pressure-
time data can also be used to scale up the effects of an
explosion to a field-scale vessel.
In addition to the above, the following questions were
answered in the laboratory report.
What is the concentration of the propane in the
vessel? Is it inside the flammability range for pro-
Was the explosion in the vessel a detonation or a
deflagration? Why?
Calculate the Kforpropane.
Under what conditions would the results of this
experiment not represent actual process results?

Chemical process safety has become a critical component
of chemical engineering education, e.g., it is very important
to industry and is required by ABET 2000. The importance
of chemical process safety in the undergraduate education of
chemical engineers is certainly consistent with having a
course, preferably required, dedicated to this topic. In the
absence of a dedicated course, however, it is essential that
safety concepts (flammability characteristics, electrostatics,
runaway reactions, explosions, and relief design, be incorpo-
rated somewhere in the undergraduate curriculum. The ex-
periments described herein provide a means for introducing
these important topics into the curriculum in a hands-on
manner that is known to enhance the learning process.

The authors acknowledge the financial support provided
by the National Science Foundation (DUE 96-50491) and
matching funds provided by Praxair, 3M, Monsanto, Cargill,
and the University of Iowa College of Engineering. They
also thank Daniel A. Crowl, Michigan Technological Uni-
versity, for his thoughts on electrostatics experiments and
for critical review of this manuscript.

1. Lepkowski, W., "Ten Years Later: Bhopal," Chem. and Eng.
News, 72, 8 (1994)
2. Heylin, M., "Bhopal," Chem. and Eng. News, 63, 14 (1985)
3. Sheppard, C.M., "Process Safety in the Curriculum: Explo-
sion Prevention Technical Elective," Chem. Eng. Ed., 32,
270 (1998)
4. Mannan, M.S., A. Akgerman, R.G. Anthony, R. Darby, P.T.
Eubank, and K.R. Hall, "Integrating Process Safety into
ChE Education and Research," Chem. Eng. Ed., 33, 198,
209 (1999)
5. Restructuring Engineering Education: A Focus on Change.
Report of an NSF Workshop on Engineering Education.
Division of Undergraduate Education, National Science
Foundation (1995)
6. Fire Protection Guide to Hazardous Materials, 11th ed.,
National Fire Protection Association, Quincy, MA (1994)
7. Felder, R. M., and R.W. Rousseau, Elementary Principles of
Chemical Processes, 3rd ed., John Wiley & Sons, Inc., New
York (2000)
8. Crowl, D.A., and J.F. Louvar, Chemical Process Safety: Fun-
damentals With Applications, Prentice-Hall, Inc., Englewood
Cliffs, NJ (1990)
9. Perry, R.H., D. W. Green, and J.O. Maloney, Perry's Chemi-
cal Engineers' Handbook, 7th ed., McGraw-Hill, New York
10. Crowl, D.A., Michigan Technological University, personal
communication (1995 and 2000)
11. Liittgens, G., and M. Glor, Understanding and Controlling
Static Electricity, Expert Verlag (1989)
12. Pratt, T.H., Electrostatic Ignitions of Fires and Explosions,
Burgoyne Incorporated, Marietta, Georgia (1997)
13. Creed, M.J., and H.K. Fauske. "An Easy Inexpensive Ap-
proach to the DIERS Procedure," Chem. Eng. Prog., 86, 45
(1990) 0
Chemical Engineering Education

Publication Rate Profiles
Continued from page 35.
sults. The CPP is the highest for the less productive depart-
ment and the lowest for a department with the most rapid
increase of research productivity. It is also evident from this
study that, in contrast to the popular belief and depending on
the publication profiles involved, the CPP does not necessar-
ily represent the "impact" and quality of research.
A new "weighted" CPP has been proposed, which elimi-
nates the influence of the publication profile on the citation
statistics and as such, is much more appropriate for measur-
ing research impact. Obviously, in order to yield valid re-
sults, the WCPP must be referred to the world average
WCPP of the particular research field.
The simulation study demonstrates once again that a com-
parative evaluation of research quality and productivity can-
not be based on a single criterion. While the WCPP has
herein proven to be more reliable in measuring "impact"
than the CPP, it still may rank a stagnant (or even a declin-
ing-production research group) and a group of a rapidly

increasing productivity the same. The WCPP can be heavily
influenced, just as the CPP can, by additional factors such as
the number of co-authors, the number of different research
groups involved, and hidden self-citation. Therefore, it is
always essential to look beyond various citation measures.

1. Grossman, I.E., "Some Pitfalls with Citation Statistics,"
Chem. Eng. Ed., 34(1), 62 (2000)
2. Angus, J.C., R.V. Edwards, and B.D. Schultz, "Ranking
Graduate Programs: Alternative Measures of Quality,"
Chem. Eng. Ed., 33(1), 72 (1999)
3. Braun, T., Letter to the Editor, C&EN, p. 6, Dec. 6 (1999)
4. Redijk, J., Letter to the Editor, C&EN, p. 6, Dec. 6 (1999)
5. Blazick, P., Letter to the Editor, C&EN, p. 6, Dec. 6 (1999)
6. Van Raan, A.F.J., "Advanced Bibliometric Methods as Quan-
titative Core of Peer Review Based Evaluation and Fore-
sight Exercises," Scientometrics, 36(3), 397 (1996)
7. Goldberger, M.L., B.A. Mahler, and P.E. Flattau (Eds), Re-
search Doctorate Programs in the United States: Continuity
and Change, National Academy Press, Washington, DC
8. "Chemistry that Counts: The Frontrunners in Four Fields,"
Science Watch, 3(3), 1 (1992) 0

SIn Memoriam

Sami Selim

Sami Selim was born in Cairo, Egypt
on January 24, 1942, and passed away
in Denver, Colorado, on June 27, 2000.
He was educated as an undergraduate at
the University of Alexandria (Egypt),
and earned his MS in chemical engi-
neering from Carnegie-Mellon and an-
other MS in mathematics and a PhD in
chemical engineering, both from Iowa
State University. Prior to joining Colorado School of Mines
(CSM) in 1982, he held faculty appointments at the University of
Petroleum and Minerals in Dhahran, Saudi Arabia, and at Texas
Tech. He was a member of the Chemical Engineering Education
Publications Board for the last decade.
While his health problems limited his physical activity in
later years, his mental activities remained extremely strong. We
at CSM will remember "Dr. Sami" as the consummate profes-
sor: a brilliant lecturer, superb theoretician, gifted problem solver,
and a world-class example of the phrase "absent-minded profes-
sor." He will be sorely missed by all of his friends and col-
leagues in the department, and by those students who were
privileged to have benefited from his wit and wisdom, both
inside and outside of the classroom.
A list of adjectives that might be used to describe Sami would
almost certainly include scholar, educator, mentor, leader, col-
league, and friend.
Scholar A "learned person one trained in a special branch
of knowledge." Sami's knowledge of transport phenomena and
applied mathematics was unparalleled in our department and

university, and he was highly respected throughout the world in
his areas of expertise.
Educator Sami taught essentially every course in our under-
graduate curriculum and every transport- and mathematics-ori-
ented course in the graduate program. In his teaching career, he
earned outstanding teacher awards four times.
Mentor Perhaps Sami's greatest contribution was in his
ability to mentor graduate students and faculty. He literally
poured himself into a quest to instill a love of knowledge and
discovery in our students.
Leader Sami emerged as a true leader at CSM as a result of
the work he did to introduce flowsheeting and computer-aided
chemical process design throughout the undergraduate curricu-
lum. During the mid to late 1980s, Sami spent a great deal of
time working with Aspen Technology and Phillips Petroleum,
learning how to perform computer-aided process simulation and
then helping the faculty introduce this technology in every course
in the ChE curriculum. Associated with this effort, he emerged
as a leader in defining the content of what we teach in chemical
engineering at CSM.
Colleague and Friend All of us at CSM have memories of
Sami that will last a lifetime. We all have treasured "Sami
stories" that we won't forget anytime in the near future.
Sami is survived by his wife, Barbara. A scholarship fund in
Sami's name has been established in the CSM Foundation. Please
contact Bob Baldwin for instructions
on donations.
Robert M. Baldwin
James F. Ely
E. Dendy Sloan

Winter 2001 4

L classroom



Intuition and Analysis

University of Sydney Sydney, New South Wales, 2006, Australia
University of California Davis, CA 95616
A common example in chemical engineering textbooks F4 I Tim.
on dynamic modeling and control is the stirred-tank control
heater illustrated in Figure 1. If the tank is treated as volme
perfectly mixed, it represents a simple process that allows
for a straightforward derivation of the governing equations,
and it helps to illustrate the use of constitutive equations in
developing dynamic process models. The assumptions in-
volved in the analysis are typically justified by common cog
sense and practical insight. With them, one arrives at a F.o
model that can be used for demonstrating dynamic be-__
havior through simulations and for testing feedback con- Figure 1. Stirred-tank heater.
trol concepts.
This problem has found its way into the majority of dy- Jose A. Romagnoli received his BS and PhD
degrees in chemical engineering from the
namics and control textbooks, and in these textbooks we Universidad Nacional del Sur (Argentina) 1973
observe three different approaches to the analysis. The first and the University of Minnesota, 1980, respec-
h is r d by t a in co t f tively. He currently holds the Orica-University of
approach is represented by the assumption of constant flow Sydney Chair of Process Systems Engineering
rates (thus constant holdup), which simplifies the problem at the University of Sydney, Australia. His re-
considerably." 1 Seborg, et al., 8 derive the model equations search interests include process monitoring, dy-
but also focus on the constant holdup problem in the text. nami, ad
The second approach is by Harriot[9] and by Pollard,1o] who
consider flow variations but still assume constant holdup. Ahmet Palazoglu received his BS degree from
The book by Stephanopoulos"''' provides a third approach in Middle East Technical Universty, Turkey, in 1978
which the holdup is not constant and perturbations are and his MS from Bogazic University in 1980.
He received his PhD in chemical engineering
present in both inlet and outlet flowrates. This leads to from Rensselaer Polytechnic Institute in 1984,
the multivariable control of both the holdup and the tem- and has been at the University of California,
Davis, since that time.
perature in the tank.
If the volume of fluid in the tank is allowed to vary (in
response to variations in the outlet flowrate), the coupling Stephen Whitaker received his undergradu-
between the mass and energy balances generates a curious, ate degree from the University of California,
non-trivial problem that has not been fully recognized in the Berkeley, and his PhD from the University of
Delaware. He is the author of two undergradu-
literature. Specifically, the dynamic behavior of the stirred ate texts, Introduction to Fluid Mechanics and
tank heater appears to contradict, at least initially, our intui- Fundamental Principles of Heat Transfer, and
of te a monograph, The Method of Volume Aver-
tive reasoning, leading to an incorrect interpretation of the aging.
expected behavior.
Copyright ChE Division of ASEE 2001
46 Chemical Engineering Education

In the next section, we first define the problem and offer
an intuitive description of the dynamic behavior of the pro-
cess. Next, in an effort to provide a rigorous explanation of
why we should expect a dynamic response contrary to our
intuition, we derive the dynamic model of the stirred-tank
heater process, carefully delineating all the steps and the
assumptions involved. Finally, we offer some insight into
the correct intuitive interpretation of the analytical result.

Figure 1 depicts a stirred-tank heat process. The purpose
of this unit is to provide heating for a process stream, thereby
increasing its temperature before being supplied to a down-
stream unit. The heat provided is denoted by Q. In the
schematic diagram, F and T represent a stream flowrate and
a stream temperature, respectively, and the subscripts "in"
and "out" refer to the inlet and outlet streams. We assume
that there is a valve placed at the outlet flow stream, which
can be adjusted to affect the tank level.
> Intuition
Now, let us perform the following thought experiment.
Assuming that the system is initially at steady state, we shall
increase the outlet stream flowrate, thereby creating a tran-
sient process in which the level in the tank decreases, as
illustrated in Figure 2. The heat supplied to the system, Q,
remains constant, and we want to know how the temperature
of the outlet stream responds to this change. Typically, intu-
ition leads to the suggestion that the outlet stream tempera-
ture will increase since less mass is now being heated in the
tank. If we decrease the outlet stream flowrate, the level in
the tank will increase, as illustrated in Figure 3. The typical

Figure 3. Decreasing the outlet stream flowrate.
Winter 2001

intuitive interpretation of this transient process is that the
outlet stream temperature will decrease because more mass
is now being heated in the tank.
Both these intuitive interpretations of the dynamic re-
sponse of the stirred-tank heater are incorrect, as can be
shown by a careful analysis of the system. We must note that
if one changes the inlet flowrate instead of the outlet flowrate,
the above reasoning leads to the correct answer, and thus
there is a fundamental difference in the way each flowrate
affects the temperature in the tank.
To summarize our survey of intuitive judgments concern-
ing the influence of changing the outlet flowrate, we note
that almost everyone believes that changing the outlet flowrate
will change the outlet temperature. In addition, most believe
that increasing the flowrate will cause an increase in the
temperature, while decreasing the flowrate will cause a de-
crease in the temperature.

> Modeling
To develop the dynamic model of the stirred-tank heater,
we make use of the macroscopic mass and thermal energy
balances"21 for a moving control volume. These are given by

d J pdV+ J p(v-w)-ndA=0 (1)
V(t) A(t)

d J pcp(T-Tref)dV+ J pcp(T-Tref)(v-w)ndA
V(t) A(t)

=- (q+qR).ndA+ (T O +T:Vv dV (2)
A(t) V(t)

in which the moving control volume, V(t), contains the fluid
in the stirred tank as illustrated in Figure 1. In terms of the
liquid depth, h(t), and the cross-sectional area of the tank, A,
the control volume can be expressed as
V(t) = Ah(t) (3)
At the gas-liquid interface, the kinematic condition requires
v-n=w-n (4)
while at the liquid solid interface we have
v.n=w.n=0 (5)
Use of these two conditions, and the assumption that the
density can be treated as a constant, allows us to express the
macroscopic mass balance as
A dh
A = Fin Fout (6)
in which Fin and Fou, represent the volumetric flow rates
entering and leaving the system. To be explicit, we express
Fou, according to

Fou= v ndA (7)

In our treatment of the thermal energy balance, we neglect
the reversible work, To Dp / Dt, and the viscous dissipation,
T: Vv, so that Eq. 2 takes the form

dt pcp(T-Tref)dV+ J pcp(T-Tref)(v-w)-ndA
V(t) A(t)
(q+qR).ndA (8)
If the only significant heat transfer to the system is caused by
the heater, the heat-transfer term on the right-hand side of
this result can be expressed as

J (q+qR)ndA=Q (9)
in which Q is positive when heat is transferred to the system.
Assuming that the variations of p and Cp are negligible, and
making use of Eqs. 3 through 5, allows us to write Eq. 8 in
the form

d [pcpAh(t)((T)- Tref)]
= pcpFin (Tin -Tref)-pcpFout (Tout Tref)+Q (10)
Here the volume averaged temperature, (T), is defined by

(T) =- Td (11)
V (t)
and since pcp can be treated as a constant, Eq. 10 can be
rearranged as


=Fin (Tin -Tref)-Fout (Tout -Tref)+ (12)
Carrying out the differentiation on the left-hand side leads to

Ahd(T) + (() f )A dh
Ah(t) dt ((T) Tref)
=Fin (Tin -Tref)-Fot (Tout -Tref)+ (13)
Equation 6 can be multiplied by (T) Tref, leading to

A dh((T)- Tref) Fin ((T) Tref )- Fout ((T) Tref) (14)
and when this result is subtracted from Eq. 13, we obtain a
simplified form of the macroscopic thermal energy balance.
The macroscopic mass and thermal energy balances repre-
sent the governing differential equations for the fluid depth
and the volume-averaged temperature in the stirred tank. We
list these two results and the initial conditions as
Adt = Fin Fout (15)
Thermal Energy

Ah(t)d(T = Fin(Tin-(T))+ Q (16)
dt pcp

Initial Conditions
h = h (T) = (T)" Fout = Fin t = 0 (17)
The driving force for the dynamic behavior is the outlet
stream flowrate, which is a function of time, i.e.,
Fout = Fin + AF(t) t>0 (18)
At this point, we identify the state variables and constant
parameters as
State Variables: h and (T)
Constant Parameters: A, pcp, Q, Fin, and T,n
It may not be obvious on the basis of Eqs. (15) through (18),
but the solution for the volume-averaged temperature for
this process is given by

(T) = (T)


- Solution
In practice, the change in the outlet stream flowrate, AF, is
a function of h-h(t), and thus the outlet stream flowrate can
be expressed by
Fot = Fin + AF[h h(t)] t 0 (20)
Use of this result in the macroscopic mass balance given by
Eq. (15) leads to
A dh = -AF[h- h(t)] (21)
and this can be solved subject to the initial condition
I.C. h = h t = 0 (22)
in order to determine h(t). The solution for the fluid depth as
a function of time can then be used with Eq. (16) to deter-
mine the volume-averaged temperature in the tank. The ini-
tial steady-state condition of the stirred tank must satisfy the
following form of the macroscopic thermal energy balance

0 =Fin (Ti (T))+ Q (23)
This result can be subtracted from Eq. (16) to obtain

Ah(t) dT = -Fin ((T)- (T)o)
If we identify the temperature difference according to
E = (T) (T)
the initial value problem for o takes the form

f(t) dt = -_
I.C. 0=0 t = 0
Here the time dependent function, f(t), is given by

f(t) = Ah(t)
and the solution to Eqs. (26a) and (26b) is






S= 0 t > 0 (28)
This leads to the result listed earlier as Eq. (19), which is
often considered to be counter-intuitive.
The forms of Eqs. (26a,b) and the solution given by Eq.
(28) usually create a little skepticism; however, we can

Chemical Engineering Education

transform the initial value problem to a more familiar form
in order to make the solution more appealing. We begin by
letting T be a function of time defined by

f(t) # 0

This transformation leads to
dO d_ dt dO f I
d ddr df(t) (30)
dt dT dt d(
and the initial-value problem given by Eqs. (26a,b) takes the
dO = -O (31a)



Clearly, the solution to this initial-value problem is given by
Eq. (28).

The source of the seemingly counter-intuitive behavior of
the tank temperature when the outlet stream flowrate is
changed lies in the key assumption of complete mixing in the
tank. To provide a purely intuitive confirmation of either Eq.
(19) or Eq. (28), we construct special processes in which the
fluid depth in the tank either decreases or increases.
Case of Outlet Stream Flowrate Increasing This situation can be best
visualized as illustrated in Figure 4 where the tank height is decreasing as a
secondary stream removes fluid from the tank while the primary outlet
flowrate remains equal to the inlet flowrate. Note that the secondary stream

Figure 4. Increasing the outlet stream flowrate

Fr. r,.

-I F I
= constan l .

Figure 5. Decreasing the outlet stream flowrate.
Winter 2001

'r ff(Tl)-'d?

flowrate is time-dependent and will vanish as time increases and a new
steady state is established. In this case. removing additional fluid from the
tank will only decrease the fluid level in the tank without changing the
temperature of the fluid in the tank. The temperature of the fluid in the tank
is determined by the rate of heat transfer, Q, that is delivered to the
incoming fluid.
Case of Outlet Stream Flowrate Decreasing This situation can be visual-
ized as in Figure 5. where the tank height is increasing as the flow is
diverted back to the tank. Again, the diverted stream is time dependent and
will vanish as time increases and a new steady state is achieved. Returning
some of the outlet stream to the tank will increase the fluid depth in the tank;
however, the temperature of the outlet stream is identical to the temperature
in the tank and returning a portion of this fluid to the tank will have no
influence on the temperature in the tank.
Rather than use the constructions illustrated in Figures 4
and 5 to enhance our intuition, we could simply observe that
the energy delivered to the system, Q, is used to raise the
temperature of the inlet stream from Tn to (T)o, and the
disposition of the outlet stream has no influence on this
energy-transport process. Thus, the heated fluid in the tank
can be disposed of more or less rapidly, giving rise to a
change in the fluid depth in the tank, without influencing the
temperature in the tank.

The dynamic behavior of a stirred-tank heater challenges
our intuition when mass and energy balances are considered
simultaneously. The perfect mixing assumption creates a
decoupled dynamic response when changes in the outlet
flowrate are considered. We have presented a rigorous model
of the process and have offered an explanation as to why we
should expect this seemingly counter-intuitive phenomenon
to occur.

1. Bequette, B.W., Process Dynamics: Modeling Analysis and Simu-
lation, Prentice Hall, Englewood Cliffs, NJ (1998)
2. Coughanowr, D.R., Process Systems Analysis and Control, 2nd
ed., McGraw-Hill, New York, NY (1991)
3. Friedly, J.C., Dynamic Behavior of Processes, Prentice Hall,
Englewood Cliffs, NJ (1972)
4. Luyben, M.L., and W.L. Luyben, Essentials of Process Control,
McGraw Hill, New York, NY (1997)
5. Marlin, T.E., Process Control: Designing Processes and Control
Systems for Dynamic Performance, McGraw Hill, New York,
NY (1995)
6. Ogunnaike, B.A., and W.H. Ray, Process Dynamics, Modeling
and Control, Oxford University Press, New York, NY (1994)
7. Smith, C.A., and A.B. Corripio, Principles and Practice ofAuto-
matic Process Control, 2nd ed., John Wiley & Sons, New York,
8. Seborg, D.E., T.F. Edgar, and D.A. Mellichamp, Process Dy-
namics and Control, John Wiley & Sons, New York, NY (1989)
9. Harriot, P., Process Control, McGraw Hill, New York, NY (1964)
10. Pollard, A., Process Control for the Chemical and Allied Fluid-
Processing Industries, Heinemann, London, England (1971)
11. Stephanopoulos, G., Chemical Process Control:An Introduction
to Theory and Practice, Prentice Hall, Englewood Cliffs, NJ
12. Whitaker, S., Fundamental Principles ofHeat Transfer, Krieger,
Malabar, FL (1983) 0




Oklahoma State University Stillwater, OK 74078-5021

Nationally, chemical engineering BS graduation rates
cycle with about a 13-year period and a 2.2-to-1
high-to-low amplitude. The past 1.5 cycles are shown
in Figure 1, a plot of the national production of BS ChE
graduates from the past twenty years.'" The numerical value
for the abscissa, Academic Year, is the calendar date for the
end of the academic year (1990 represents the academic year
from the fall of 1989 to the summer of 1990). This study also
considered the 20-year history of 30 individual U.S. ChE
programs,121 chosen to represent a diversity of program types.
All 30 schools cycle substantially in phase with the national
data, and each with about a 5-to-1 ratio. Figure 2 presents a
graph of the trends and visually suggests that the BS-ChE
rates at all schools appear to cycle in phase with the
national data. Local events and the statistics of small
numbers make the individual school amplitudes greater
than the national amplitude.
In a more quantitative analysis of the data, Table 1 pre-
sents correlation coefficients, r, of the BS graduation rate at
each school to the national rate. All "r" values larger than
0.34 are significant at the 95% level for the number of data
points, and all but 2 of the 30 schools observed have "r"
values larger than 0.34. Even schools with smaller "r" values
have a BS production rate that is somewhat correlated to the
national trend. Coefficient of Variation (CV) results are also

R. Russell Rhinehart is Head of the School
of Chemical Engineering at Oklahoma State
University, with former experience in both in-
dustry (13 years) and academe (15 years).
His primary research interests are in the prac- Al
tical application of advanced technology for
automated process management (control, op-
timization, monitoring), He received his BS
(1968) and his MS in Nuclear Engineering .
from the University of Maryland, and his PhD
(1985) in Chemical Engineering from North
Carolina State University.

Copyright ChE Division of ASEE 2001

presented in the table for each school. CV is the ratio of the
standard deviation to the average. All schools have a CV of
about 0.4, and this common value indicates that all show the
same relative cycling amplitude. There are no trends of
correlation coefficient, CV, or cycle amplitude with ChE
program size. The data reveal that the phenomenon is
national and affects all schools in unison and to the same
relative degree.
The cycling is a source of great discomfort, and it hurts
chemical engineering education. During periods of peak BS
production rates, students who graduate without job offers
feel betrayed. Parents also become upset and challenge the

6 7000 -- a
74000 t t .. 1
5000 -- -
04000 ..*.. ..
3000 t U-U .
1970 1980 1990 2000
Academic Year

Figure 1. National BS-ChE rates.

180 i 8000
160 --- -- ---- 7000
140 6000
120 -
60 a 300

40 -

o0 0
1965 1970 1975 1980 1985 1990 1995 2000
Academic Year

Figure 2. BS-ChE rates, national and various schools,
versus year.

Chemical Engineering Education

ChE program's adequacy. This
is not a good way to generate
the alumni allegiance necessary
for program development.

At the other extreme, in pe-
riods of low BS production
rates, State and University ad-
ministrations often question the
need for an expensive ChE pro-
gram. This often leads to inad-
equate resources for lab up-
grades and difficulty in retain-
ing faculty positions. Both lab
equipment and faculty positions
are critical for sustained pro-
gram excellence, especially
during peak enrollment periods.
Further, during low production
periods, industry must inflate
entry-level salaries and make
an excess number of job offers
to attract a sufficient number
of employees. While this is ad-
vantageous for the new gradu-
ates, it results in an uncomfort-
able salary compression and re-
source-allocation problem for
industry. Finally, during peri-
ods of low BS supply, compa-
nies withdraw from recruiting
at schools, making it difficult
to later reestablish on-campus
recruiting and to maintain con-
tinuity in other forms of in-
dustrial support.

Statistics on BS-ChE Data for Thirty Schools and the Nation


U Michigan
Arizona State
Cal Poly, Pomona
Cal State Long Beach
Cal Tech

Georgia Tech
Kansas State
North Carolina State
Oklahoma State
Oregon State

San Jose State
Texas Tech
U Arizona

U Florida
U Kansas
U Houston
U Oklahoma
U Southern Cal

U Washington
Cal Berkeley
Cal Davis
Cal San Diego
Cal Santa Barbara

U Texas- Austin
Washington State

The cycling has lasted as far Nat
back as data could be found.
Figure 3 shows the BS-ChE his-
tory at Oklahoma State Uni-
versity (OSU) from its first graduate in 1921 to the
present. The 13-year cycling period is evident since about
1930, after the start-up phase of the program.

It appears generally accepted that the number of stu-
dents who choose engineering is influenced by the job
opportunities and the salary levels. In his study of over-
all engineering enrollments from 1965 to 1995, Heckel'31
Engineering enrollment trends are shown to difter
significantly from those of undergraduates as a whole and
to exhibit little correlation with trends in high school
graduation data. Freshmen engineering enrollments show
very strong correlation with factors which might indicate to
high school students the magnitude of their personal
economic gain such as on-campus industrial recruiting

Max Min Ratio
Number Number Max
ofBS ofBS to Std.
ChE ChE Min. Dev. Average

158 41 3.85 39 85
48 12 4.00 11 27
56 11 5.09 12 32
60 13 4.62 14 28
20 3 6.67 4 9

165 55 3.00 37 101
38 10 3.80 9 22
133 30 4.43 37 75
62 10 6.20 15 33
61 5 12.20 13 33

116 27 4.30 28 63
20 8 2.50 4 14
34 6 5.67 9 16
51 4 12.75 12 28
46 11 4.18 11 25

88 9 9.78 18 39
57 11 5.18 13 31
55 16 3.44 10 33
71 17 4.18 18 36
45 8 5.63 10 21

89 31 2.87 15 60
141 30 4.70 33 78
61 21 2.90 14 37
27 3 9.00 6 19
52 8 6.50 11 22

69 11 6.27 19 32
165 46 3.59 32 91
45 9 5.00 11 24
34 10 3.40 7 22
20 2 10.00 6 11

7475 3070 2.43 1463 4859

Ratio of Number Correl.
Range of years Coef.,
CV to Avg. ofdata r

0.46 1.38 29
0.42 1.31 29
0.37 1.42 29
0.49 1.67 27
0.49 1.93 29

0.36 1.09 29
0.41 1.28 29
0.49 1.38 29
0.45 1.59 29
0.40 1.67 29

0.44 1.42 29
0.28 0.88 29
0.54 1.74 29
0.45 1.70 28
0.44 1.41 29

0.46 2.01 29
0.41 1.47 29
0.31 1.18 29
0.49 1.51 29
0.49 1.80 29

0.26 0.97 29
0.42 1.43 29
0.37 1.09 29
0.34 1.27 19
0.50 2.01 29

0.59 1.80 15
0.35 1.31 29
0.48 1.52 29
0.32 1.07 29
0.53 1.65 15

0.30 0.91 26








40 -
m i : 9"

'20 --W
10 ---
1U : .: "

i i

1920 1930 1940 1950 1960 1970 1980 1990 2000
Academic Year

Figure 3. BS-ChE production at OSU.
Figure 3. BS-ChE production at OSU.

Winter 2001

intensity, annual growth in starting salaries and starting salary
levels relative to average salaries of all undergraduates. No
correlation between engineering freshmen enrollments and
national economic conditions as measured by the Gross
Domestic Product or unemployment was found.
Most engineering students seem to make an economically
rational decision, seeking to maximize the probability of
attaining a high-salary job upon graduation, and this one
incentive seems to control trends in engineering enroll-
ments. It appears that in certain engineering programs
(aerospace, chemical, materials, and nuclear) the cycling
behavior is unusually large.
If we can understand the mechanism, then we can hope to
design a cure. This paper presents data that refute the com-
monly accepted mechanism, a novel view of the mechanism
for cycling, a model that expresses the same behavior as the
national data, needs for further study, and possible solutions
to the cycling problem.

Enrollment cycling is a symptom: there is a cause, and a
successful cure for the cycling must affect the cause. If we
accept that the cause is "A" when it is actually "B," and
invest efforts to cure "A," we will not alleviate the cycling.
This study appears to refute the commonly held cause.
The traditional view is that enrollment swings are demand
driven-they reflect student choices in response to the chang-
ing ChE job demands by the economy. In this view, low
matriculation rates are due to low numbers of job offers in
prior years (due to a low economy), and then low matricula-
tion leads to low BS production about four to five years later.
There is, however, no correlation of national engineering
enrollment to national economic factors.[2] Figure 4 provides
additional data and reveals that there is no correlation of the
national BS-ChE cycling to the national economy, as mea-
sured by the gross domestic product (GDP). In Figure 4, the
GDP in constant dollars is plotted with the national BS-ChE
production; it reveals that the BS-ChE oscillations are seem-
ingly independent of, and much larger than, changes in the
national economy. In addition, economic factors (that would
indicate job opportunities for chemical engineers) have shown
a relatively steady growth, with neither 2-to-1 cycling nor a
regular 13-year period.
One may suppose that while the overall national economy
does not cycle, sectors do-and that the employment de-
mand by specific industry sectors causes enrollment cycling
for schools that supply those sectors. But if this were true,
then Gulf-Southwest schools, which are significantly coupled
to the oil and gas industry, would cycle independently of
Northwest schools, which are significantly coupled to the
pulp and paper industry, etc. Enrollment at all schools, how-
ever, regardless of region or orientation, cycles in phase,
with the same relative amplitude and with a regular period.

As a supporting refutation of the impact of local school
data on school cycling, see Figure 5. Figure 5a plots the
ChE freshmen-class size versus the BS-ChE rate at OSU,
and Figure 5b plots the same data versus the national BS-
ChE rate. The four-year lag led to the strongest correlation
for both scatter plots. The correlation coefficient, r2, stated
on each graph, along with the observable closeness of the
data to the regression lines, indicates that OSU freshmen-
class-size data correlates much better to the national data
than to our local data. Figures 5c and 5d reveal similar
relationships for the total OSU undergraduate ChE enroll-
ment. Student choices at OSU correlate much better to na-
tional data than to local data. The same probably holds true
for other schools.
The commonly held view of the cause for cycling is in
contrast to data. Let's consider mechanism "B."

The mechanism for ChE enrollment cycling hypothesized
here is not based on employment demand. It is based on BS-
ChE supply. The cycling is supply driven. The cycling causes
the cycling. The job market is relatively steady, but when
there is an excess supply of graduates there is a low prob-
ability of finding a job, and high school students do not
choose ChE. Subsequently, when graduation rates are low,
there is a very high demand for ChEs, which is exaggerated
by industrial competition for the limited supply. This attracts
a flood of students. Further, the cycling mechanism hypoth-
esized here is not based on local events; it is based on the
national data.
Proposed here: the enrollment process is inherently un-
stable because of the controllers-the "high gains" in stu-
dent and recruiter choices.
The hypothesized incentive that drives enrollment for
chemical engineering is a perception, by high school stu-
dents, of the attractiveness and availability of jobs at the
national level. Attractiveness includes a combination of fac-
tors that appeal to young adults and includes both profes-
sional and personal attributes such as salary, social stature,


11000 ,
S7000 .
g 5000 -


1965 1970 1975 1980 1985 1990 1995 2000 2005
Academic Year
Figure 4. GDP and national BS-ChE rate.
Chemical Engineering Education


120 = 0.32

40 '

S0 4


0 10 20 30 40 50 60
OSU BS graduates in year I
Figure 5a. Freshman+other class size vs. OSU BS rate
(lag 4).

400 -R066

100 1 00

40 -

a ,00- --- -- -- -^ .
3000 4000 5000 6000 7000 8000
National BS graduates in year i
Figure 5b. Freshman+other class size vs. national BS rate
(lag 4).

350 -
S o R2- 0.57

soo *. >
| 250-

1 200 -

S10 --- ---
I- o
6 50
04 --------- -------------------
o 10

0 10 20 3D 40 50 60
OSU BS ChE Rate in year i

Figure 5c. OSU ChE total enrollment vs. OSU BS-ChE
rate (lag 5).

3R 0.89


150 -

so 00
o 50

3000 4000 5000 6000 7000 8000
National BS ChE Rate in year I

Figure 5d. OSU ChE total enrollment vs national BS rate
(lag 5).
Winter 2001

and academic challenge. Availability of jobs is indicated by
the fraction of students who obtain jobs upon graduation, but
it is based on recent-past data, not on the projected future.

Informal discussions and analysis with OSU ChE students
reveal several features of the decision process that control
enrollment. It appears that selecting ChE as a major is sensi-
tive to the perceived incentive of a high salary. Students who
can complete a degree in chemical engineering can probably
complete a degree in any of a number of chemical science
degree programs. These students probably had a chemistry
course in high school, but they are not familiar with any of
the professions within the chemical sciences, and the mar-
keting material for the various professions promises chal-
lenging, enjoyable, and socially important careers. As a re-
sult, their main incentive in the choice of a college major is
probable entry into a high-paying job upon graduation.

Within the chemical science careers, chemical engineering
is the degree program that leads to the highest employment
salary. When the probability of getting a ChE job is high,
students rush into the program. But the program is very
demanding, and when the probability of getting a job ap-
pears low, these same students choose other majors in an
economically rational decision. One source of data that stu-
dents might use to make that decision is the employment-
upon-graduation data that ChE schools submit each May/
June to AIChE, the Council for Chemical Research, and
other agencies. The data is nationally compiled and pub-
lished about a year later, through a variety of private, public,
and not-for-profit agencies. When high school seniors begin
to consider college choices they have access to this year-old
(and older) data. The mechanism proposed here is that em-
ployment-upon-graduation data from one year affects the
decision of matriculates two years hence and affects the BS
graduation rate four to five years later. This suggests a cause
of the six-to-seven year swing from high to low BS rate and
the 13-year cycle period.


A model of this mechanism is developed from a simple
population balance on students in each class-year category.
It reveals the natural instability of the process, the 13-year
period, the BS production rate cycling, and the national
coordination. It also produces statistics that match many
features of OSU ChE enrollment data. The model for a
single school is presented first.

One aspect of the model is the constitutive relation that
describes the number of high school students entering the
university who declare chemical engineering when they ma-
triculate. This number is a function of the incentive to study
chemical engineering. In what follows we use the term a to
represent the fraction of university matriculates declaring
ChE. In this study we use the number of job offers per BS-
ChE graduate as the incentive. We feel fairly certain of three

points on the a-versus-incentive curve. When there is zero
incentive (zero job offers for any ChE graduate), then there
will be no ChE matriculates. When the incentive is at a
maximum (experience indicates that this might be three job
offers per BS-ChE) then matriculation rates will be at a
maximum. At OSU this maximum in matriculation is about
twice the nominal value that seems to be 0.006 freshmen
ChEs per OSU freshman. So, a(incentive=3)=0.012 at OSU.
Finally, a nominal matriculation rate occurs when the incen-
tive is nominal, defining a(incentive=l)=0.006 (at OSU).
We also feel fairly certain of the general shape of the a-
versus-incentive curve. Regardless of the incentive, a will
not be much greater than 0.012, the apparent maximum.
There is only a portion of the population that leaves high
school with either the preparation or willingness to study
chemical engineering, and regardless of the incentive the
remainder will not choose ChE as a major. Consequently, a
should asymptotically increase to its upper limit of about
0.012. This feature and the three points define an S-shaped
curve, which we model as

a= 0.01679.exp(-1.0397/ r) (1)

where the variable "r" represents the incentive, the ratio of
job offers per graduating BS. The relation is illustrated in
Figure 6.
There is no pretense that this equation for a is precise, or
that the ratio of job offers per BS is the true and exclusive
measure of the incentive to study ChE. It appears to be
reasonable. Neither is there a claim that this equation is
necessary to stimulate cycling. One could certainly argue
that alpha should be based on a moving average of job-offers
to BS ratio, and the reader is encouraged to explore such
options. In this study any number of reasonable models for
a seem to lead to cycling. The feature that is important,
however, is that this is a nonlinear, high-gain relation.
The model is composed of a population balance on the
number of students in each class-year. The number of enter-
ing ChE freshmen is a multiplied by hs, the number of high
school students matriculating at the university. In what fol-
lows, the symbol "N" will represent the number of ChE
students, and the modifiers, "l," "2," ..., "5," will represent
the class year (freshman, sophomore, ..., through second-
year senior). The modifier "R" represents students who are
repeating the class-year. Accordingly, Eq. (2) presents a
population balance of students taking freshmen courses. The
subscript "i" represents the academic year. In Eq. (2), the
number of people in the freshman ChE class is the sum of
the number of new matriculates plus the number of students
who were freshmen last year and who are still taking fresh-
men-level courses this year. Some are repeating; some are
still ramping up from inadequate high school preparation;
some are on a reduced course load plan. To reflect both the
communications delay in presenting the "job market" to

Figure 6. Alpha as a function of jobs offers per BS.

high school seniors who are choosing a major and the persis-
tence of the recent-past "job market" history, we use the
average value of alpha of the two and three preceding years.
Nli = 0.5(Xi-3 + (i-2)hs + N1Ri_1 (2)
Based on historical data at OSU, the number of students
remaining in the freshmen-level ChE courses in year, N1Ri,
is approximately 20% of the freshmen in the previous year.
The formula used is
N1Ri =0.2 Nl (3)
It appears, from data from various sources, that approxi-
mately 60% of the ChE freshmen would progress into the
sophomore-level class; and at OSU, this number seems inde-
pendent of either class size or incentive. Transfers into the
sophomore level is conceived as students who have spent a
year in another major and have been enticed by the incen-
tives to switch into chemical engineering. It appears reason-
able to model this as 10% of previous year's a multiplied by
also, the total number of sophomores at the university.
Including a term for students that repeat the sophomore
class, the population balance for students in the sophomore
class is

N2i =0.1ci_- allso+0.6Nli_i +N2Ri_l (4)
The number of people that remain in the sophomore-level
classes was also modeled as 20%. The equation is

N2Ri =0.2N2i (5)
The equation for the number of students in the junior year
is similar to that for the sophomores. OSU data shows that
about 65% of the sophomores proceeded into the junior year,
and that this proportion, too, is independent of events. It was
also assumed that transfers only accounted for 1% of the
previous year's alpha multiplied by alljr, the total number of
juniors at the university. The equation is

N3=0.01ai1l alljr+0.65N2i_1 +N3R,_I (6)
OSU data indicate that about 5% of the juniors repeat their
junior year. The equation for the number of repeating juniors is
Chemical Engineering Education

N3Ri=0.05N3i (7)
It was assumed that there are no transfers in the senior
year. Approximately 93% of the OSU ChE juniors enter into
the senior year. The equation that describes the population
balance around the senior class is

N4i =0.93N3i-1 +N4Ri_- (8)
It was assumed that only one percent of the senior class
would have to repeat the year. The following equation was
used to describe the number of seniors that repeat:
N4Ri=0.01N4i (9)
OSU data shows that approximately 54% of the seniors
would go on to a fifth year where they would take courses
off of the critical path that they postponed earlier. The fol-
lowing equation is used to describe the number of students
entering the 5th year:
N5i=0.54N4i_1 (10)
The number of graduating seniors, NG, is modeled as the
number of 5th-year students plus 42% of seniors:
NGi =N5i +0.42N4i (11)
The missing 4% of the seniors seem to drop out of the OSU
program for varying reasons unrelated to academic ability.
The number of available employment positions, Jobs, is
modeled using a random walk starting from a nominal steady-
state number. The driver for the walk is a Gaussian-distrib-
uted random variable, NID( J.=0, c), where 3 =0.04* Jobs.
Generated by the Box-Mueller method, the job market equa-
tion is

Jobsi = Jobsi1 + (y-2 log(r) sin(2rr2) (12)

Here, r, and r, are independently and uniformly distributed
random numbers on the interval 0 to .999... This is a simple
model, and to prevent some realizations from leading to a
negative job market, the lower value is limited to at least one
job per year.
There seem to be any number of disturbances or distur-
bance models that lead to the limit cycling. These include
perturbations on the number of high school students or the
model's retention coefficients. This simple economic driver
was chosen for this example.
The number of jobs offered to BS ChEs may be different
than the number of positions available. In a "buyer's mar-
ket," in times when there are more graduates than there are
available positions, companies make one offer for each posi-
tion. But, in a "seller's market," in times when there are
more job openings than there are available graduates, com-
panies often make more offers than there are positions in an
attempt to fill all of their positions. The bigger the supply
deficit, the more aggressive are recruiting efforts; however,
Winter 2001

in fear of over staffing, it appears that companies limit
themselves to no more than two outstanding offers per open
position. Here the variable 3 represents the ratio of job
offers per available position. If it's a "buyer's market," 3=1.
If it's a "seller's market," this study models 0 as rising with
the square of the ratio of jobs per BS degree, but with an
upper limit of 2.

( Jobsi `2
B- i = b) (13)

The number of job offers is then P multiplied by the
number of jobs available for that year.
Joffersi = Jobsi Pi (14)
The number of job offers per BS ChE is then used to
calculate the incentive, alpha. Other models for 3 that ex-
press similar behavior produce equivalent cycling behavior.
Initial values for N1, N2, N3, N4, and N5 were chosen to
start the class sizes at a steady state, with NG equal to the
number of employment positions.
These equations lead to a fractional number of students in
any category. In this simulation all values that represented
the number of people were rounded to an integer.

Figure 7 shows the dynamic response of a single-school
model, from the initial steady state to the limit cycle, as
instigated by one realization of the random perturbations in
the job market. Notice that the variation in BS production
rate cycles regularly, irrespective of the number of jobs.
Notice also that the amplitude of variation in the BS rate is
much greater than the amplitude in the number ofjobs. This
was one realization; any other, independent of driver, shows
the same eventual behavior.
Figure 8 reveals the behavior of the number of students in
each class year for a smaller school. The population of each
class shows a phase lag of one year from the previous class
and the expected reduction in numbers due to attrition.

The single-school model was expanded to ten schools. In
one study, shown in Figure 9, each of the ten schools is
independent in the sense that: 1) the "local" student's incen-
tive was purely based on the local incentive (job offers to BS
ChEs at that particular school), and 2) the perturbation in the
local job market for each of the ten schools was independent
of the nine other local markets. The sizes of the ChE pro-
grams ranged from an average BS-ChE rate of 12 to 45 per
year, and no significant impact on the period of the oscilla-
tion was observed. All were about 13 years. Noticeable,
though, is that all schools found their own phase in response
to the independent realizations of the disturbance, cycling
was not coordinated.

To model a national influence, to generate correla-
tion between schools such as is observed in the na-
tional data, the calculation of a is weighted to repre-
sent both "local" and the ten-school "national" job h ,0
offers per BS incentive. It was found that the more
national influence, the more the cycles became "in 0
phase ." At only a 20/80 national/local weighting, Fig- Z |,
ure 10 shows that all ten schools become "locked" in A
phase in spite of independent random behavior in the 10 -
local job markets, and independent initial behavior.__
0 10 20 30 40 50 60 70 80 90 100
DISCUSSION Academic Year
Results of model explorations not shown here re- Figure 7. BS degrees awarded and jobs available versus
veal that the cycling phenomenon is independent of academic year: single-school model.
all reasonable adjustments in model coefficients. There
were only two features that seem to have a tempering ,5
influence on the cycling phenomena; these being the o
students' and the employers' response to the jobs/BS
ratio-modeled here as alpha and beta. If there is a M
cure for the cycling phenomena, it seems grounded |
in tempering human reaction to a perceived per-
sonal impact.
This is a simplistic, deterministic model that as-
sumes all students share a common, time independent, o o 70 o ,0
region independent stimulus-response mechanism. Academic Year
While this study does not answer many important Figure 8. Class size vs time: single-school model.
questions, and while it cannot claim to be a definitive
exposition on individual student behavior, it does pro-

nism that leads to enrollment cycling. I.
Perhaps characteristics of a degree program that ..
exhibits cycling are: 1 40
A difficult program, both conceptually and time 2
demanding, that can only be passed by a small 10
number of the population who have the innate 0
ability, self discipline, and adequate preparatory 0 10 20 30 40 o50 6 70 80 90 10
Academic Year, After Steady State
A program with specialty subject matter that is of Figure 9. BS production rate simulation: 10-school model,
interest or attraction to a small portion of the local influence only.
A degree that leads to very high starting salary,
social status, and/or secure lifestyle. 6o
A degree that leads to a career that is relatively non- -- ---
understood by the student, and where salary
therefore dominates the possible influence of all
other possible incentives or connections to the -
student's personal values. .

If these characteristics lead to cycling, then there
may be actions that we can take to alleviate the cy- 0
0 10 20 30 40 50 60 70 80 90 100
cling. One approach would be to collect data on em- Academc Year, After Steady State
ployment based on graduate status four months after
graduation, and exclude data on graduates without Figure 10. BS production rate simulation: 10-school
permanent work status. It appears not that we produce model, 20% national influence on enrollment.

56 Chemical Engineering Education

too many BS graduates; it appears that we produce more
than are demanded by on-campus recruiters. Most of those
unemployed upon graduation find jobs within several months
after graduation, when they are free of the capstone course
demands and have the time to search for employment. Some
others choose graduate or professional school. If the em-
ployment data reflected employment rates about four
months after graduation, then the apparent probability of
landing a job might always be attractive, and cycling
might be tempered.
Additionally, perhaps professional societies and govern-
ment agencies that are accountable for rational system be-
havior and educational excellence, could shape the aware-
ness of the societal benefits that come from chemical engi-
neering, and thereby have the career appeal to personal
values other than just economic gain. Then, fewer matricu-
lates would be so greatly influenced by the apparent job
demand, and enrollments would be more level.
Another tempering action might be to reveal this cycling
phenomenon to high school students and counselors so that
they make matriculation decisions based on the future, not
the recent past, job market.
The model and data presented here provide indirect evi-
dence to support the mechanism; but there is no direct,
credible evidence to validate the hypothesized mechanism,
the human response, or to evaluate possible cures. It appears
that questions such as the following need to be answered.
What are the general characteristics of degree pro-
grams, or characteristics of the people, or nature of the
environment that lead to cycling? Is there a commonal-
ity of cycling phenomena for degree programs with
those common characteristics?
What is the primary incentive for high school and
transfer students to choose ChE as a major?
What is the model for the fraction of college-bound
students choosing ChE as a function of that incentive
(alpha in this study)?
Is it possible to shift the ChE enrollment incentive to
other professional or conscience attributes (social
stature, commitment to the environment, commitment to
human health, commitment to improve U.S. competitive-
ness, etc.) by a marketing campaign, and thereby reduce
the number who choose ChE for probable salary, and
thereby reduce the cycling amplitude?
What mechanisms create the information that becomes
the basis for student choices? Are they professional
society surveys of job salary and job satisfaction,
professional society surveys of the number of students
with jobs upon graduation, etc?
What mechanisms convey the information that becomes
the basis for student choices? Are they hearsay, older
siblings, friends, family, local employers, high school
counseling pamphlets, Internet data banks, Department
of Labor statistics, Society Publications, etc?
Winter 2001

What is the industrial recruiting response to over- and
under-supply of BS-ChE candidates for the number of
job openings (beta in this study)?
How do students, parents, faculty, administrators, and
employers view the effects of the cycling? What is the
magnitude of the cycling problem (perhaps as indicated
by the economic and personal impacts on society)?

Maximization of probable high-salary employment re-
mains accepted as the driver for student choice for enroll-
ment in chemical engineering. However, data refute the com-
monly held view that the "economy," the industrial job
market demand for BS-ChEs, is responsible for ChE enroll-
ment cycling. Hypothesized here, the perception of job op-
portunities by college matriculates is influenced more by
enrollment swings than the economy. Changes in the supply
of BS-ChEs dominates the supply-to-demand statistics, and
hence enrollment. Enrollment cycling is supply driven. A
primitive model of the mechanism expresses multiple as-
pects of national and local data. The model reveals that the
dynamic system is inherently unstable, that enrollments tend
to a limit cycle, and that student and recruiter response to the
supply-to-demand for BS-ChEs is the source of the instabil-
ity. A better understanding of this mechanism may lead to
solutions; but, since national statistics appear to drive the
perception, it appears that any cure must be implemented by
the agencies that create national perception.

The author appreciates the generosity, rapid response, in-
terest, and contributions of: Jacob T. Dearmon (Senior, OSU,
Chemical Engineering), who substantially contributed to the
data gathering, data workup, and model coding; Dr. Richard
W. Heckel (emeritus, Michigan Tech), who kindly provided
the historical data on the 30 schools; Drs. Melanie L. Page
(OSU, Psychology) and John R. Cross (OSU, Sociology)
who provided insight and encouragement; Randy S. Sharp
(M.S. Candidate, OSU, Chemical Engineering), who helped
with the original model coding; Drs. Heckle, Robert L.
Robinson (OSU, Chemical Engineering), and three anony-
mous CEE reviewers who provided helpful comments on the
manuscript. The OSU Office of University Assessment pro-
vided partial support for this work.

1. U.S. Department of Labor, Engineering Workforce Commis-
sion ( 1999036> accessed 5/18/99, National Center for Education
Statistics, Digest of Engineering Statistics) (1998)
2. Heckel, R.W., Professor Emeritus, Department of Metallur-
gical and Materials Engineering, MichiganTechnological
University; private collection
3. Heckel, R.W., "Engineering Freshman Enrollments: Criti-
cal and Non-Critical Factors," J. Eng. Ed., 85(1), 15 (1996)

[e, -learning in industry

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


A Key Link Between

Industry and Engineers in the Making

2053 21st Street SE, Apt. S Hickory, NC 28602

A complex metamorphosis must take place before an
engineering student understands what it really means
to be an engineer. Like most processes, there are
many paths to achieving the desired final result. This article
considers not only my actual experiences in that respect, but
also reflects on my discussions with many peers over the
years-so it is more than just a memoir of my passage along
this path to understanding.
In high school, the process toward understanding often
begins upon learning that mathematics and science are tools
of a mysterious profession called "engineering." It is not
unusual to find that when some of the "smart" people in the
graduating class announce their plans to pursue that profes-
sion, other students will often find themselves pulled in the
same direction. In my case, by the beginning of my senior
year in high school I had decided that engineering was the
path I wanted to follow. Despite this early decision, when I
entered college I still had only the slightest idea of what an
engineer's real role in society was or what the typical day-to-
day duties of an engineer really were.
The second stage of metamorphosis occurs in the first
engineering class when one is challenged to "think like an
engineer." At this juncture, the class divides into two types
of students. Type I students are extremely intuitive and seem
to be able to see processes in their minds without ever

having been in a plant, while Type II students can do the
math and have intuitive skills, but can't visualize the pro-
cess. I was definitely a Type II student.
Cooperative education, summer internships, or part-time
undergraduate research can eventually help Type II students
such as myself to connect what we study to what it is that
real engineers do, e.g., solve problems. Most universities
have a freshman class that considers and explains some of
the differences between engineering fields. In my case, how-
ever, this introductory course was extremely vague. For such
a course to be really useful to the student, I feel it should
thoroughly define each engineering discipline. For each de-
partment, there should be specific examples of the types of
assignments a student might expect. Tours of companies
employing various types of engineers could be included to
allow students to see where they might expect to work after

Tanya Bradburn graduated from North Caro-
lina State University in May of 2000 with a BS
,in Chemical Engineering. She is currently work-
ing as an engineer in the Research, Develop-
ment, and Engineering Division of Coring
Cable Systems in Hickory, NC.

Copyright ChE Division of ASEE 2001

Chemical Engineering Education

graduation, and a chance to hear an engineer from each
environment explain what he or she does would be ex-
tremely useful students by helping them make informed
decisions about the course of study to follow. The co-op
experience naturally incorporates such opportunities for those
students, like myself, who participate. Practically speaking,
however, I believe co-op opportunities come too late to be
useful to many of the Type II students who have become
disenchanted in the first three semesters on campus.
Stopgap measures incorporating some of the most desir-
able components noted above are needed in order to reduce
the loss of students experienced as a result of the nonspecific
introductory courses offered at the beginning of their educa-
tional experience. A good beginning would be to simply
show the students pictures of equip-
ment that engineers use in each of
the disciplines. For example, chemi-
cal engineering students could be
shown pictures of separation equip-
ment, computer control equipment,
and microelectronic and bio-process- en there c
ing equipment. This would help pro- sli tes
vide students with a visual idea of Slg t t
the huge range of opportunities avail- real role
able in the field. Also, at least one
field trip, complemented by on-cam- typi
pus presentations by articulate prac-
ticing engineers, could and should be
included in such a course.

A key common element in all these
options is the opportunity for a student to see, firsthand, the
exciting impact an engineer can have in the workplace. The
co-op option appears to be the most practical approach for
most students. This conclusion is based on the reality of
operation in the time-constrained environments present in
companies and universities. The companies that have com-
mitted to a bona fide co-op program understand its value as a
screening and recruiting tool and are willing to invest in it.
Such companies actually have a framework in place to sup-
port the program.
For any off-campus opportunity, knowing the answers to
the following questions will help students know what to
expect and how to prepare for their work assignments:

Is the potential assignment part of an on-going
established program in the company, or will you
be the guinea pig?
Has there ever been a co-op before in the area of
the company where you will be working-can you
contact prior participants ?
Does the company have a year round co-op
program, so you aren't disadvantaged in access to

class sequences needed for your graduation?
Will you be paired with an actual mentor?
Are there housing accommodations? If the
company reserves apartments for co-ops/interns,
are they furnished?
How will your salary typically change as you
complete rotations?
Do you get vacation days?

Some of the questions may sound inconsequential, but
they aren't. Shuffling back and forth from campus to plant
to campus without breaks gives students very little time to
digest their experiences. Fitting into an existing optimized
structure designed to accommodate students is a huge ad-

Despite this early decision
o into engineering], when I
I college I still had only the
idea of what an engineer's
in society was or what the
ical day-to-day duties of an
engineer really were.

vantage in terms of the ultimate benefits to participants.
Maintaining such a framework is a significant contribu-
tion made by companies that are truly committed to co-op
programs. Frameworks for moving and housing, as well as
for other placement issues, is less common in so-called
internships, since they are typically discontinuous and are
centered around the summer months. Moreover, comments
by internship participants suggest that such assignments of-
ten involve more ad hoc "mentor" relationships, as com-
pared to co-op assignments.
The third option, undergraduate research, is another area
of opportunity I have had experience with, and I will con-
sider its benefits after focusing on the off-campus co-op
benefits. While undergraduate research is highly desirable,
the time commitment involved in trying to accommodate all
undergraduates on campus via this route seems to be unrea-
sonable in typical educational institutions.
As noted above, co-op assignments are quite effective
because they give students an empowering insight into what
can ultimately be done with their degrees. Such insight can
give students the momentum to overcome hurdles along the
way to acquiring their degree and can ultimately increase the

Winter 2001

number of students who actually graduate.
Co-op work can also have second-order advantages. In my
case, it forced me to leave the safe surroundings of familiar
people and places. It put me into a new state, into a new
home, and into a completely different environment. Without
a doubt, pursuing a co-op job at Eastman Chemical Com-
pany in Kingsport, Tennessee, was one of the best decisions
I had ever made. I had always lived in the Raleigh-Durham
area, and moving to another state to work was an unsettling
decision, to say the least. I was five hours from home when I
drove by Eastman, or the "Big E," on the way to my first
apartment, and my mouth hung open the entire time as I
drove past the plant. I had never seen such a plant before-
the Kingsport Division is one of the largest chemical plants
in the US. I remember thinking as I drove past it, "I have
just made the biggest mistake of my life!!!" I simply wasn't
prepared for the smells of a chemical plant, the sight of the
looming distillation columns, the reactors, and what seemed
to be smoke (actually steam) swirling in the air.
To be honest, at the time I didn't know what a distillation
column or a reactor looked like. The boxes that I had drawn
on engineering paper to represent such equipment in no way
approximated the sights, sounds, and smells of an actual
plant. I often think of how disconcerting it would have been
if I had never worked on a co-op job, and if the first time I
saw a chemical plant was after graduating as a BS engineer.
Even if introductory classes had included pictures of a distil-
lation column, I believe a Type II student such as myself
would be unprepared for what it is really like to see one in a
plant. In any case, however, it would have been more reas-
suring to at least know what I was looking at.
During my first co-op rotation, I worked with a separa-
tions group on a high-priority adsorption project for feed
purification. Sometimes, I would work from 7 a.m. to 11
p.m. to meet our project objectives, but in spite of the long
hours, it was exciting and challenging to be working on a
real problem. For my second term, I worked in Cellulose
Esters doing compressibility and percent solids analysis, and
for my third rotation, I worked on optimizing the sludge
system for the polymer division, working with several crews
of operators. The reason for describing my co-op experience
here is to show that no college course could possibly simu-
late what it is like to work with other operators, to actually
be in a plant, or to work on a truly stressful project where the
results mattered. It is a far different experience from just
working for a grade!
Contact is made with students from many different schools
during one's work cycles, and based on conversations with
these peers over the years, I have reached a number of
conclusions about the strengths and weaknesses of educating
chemical engineers. The following paragraphs elucidate a few
of my thoughts about steps that could be taken that would
greatly enhance the educational experience of students.

After students select an engineering specialty, regardless
of whether or not they know what they are getting into, more
practical applications should be included to complement the
theoretical backgrounds they receive in the classroom. While
in some cases, such practical teaching may not be eco-
nomically possible, more can usually be done and it can
be done much earlier.
Specifically, simple qualitative experimental illustrations
would be useful, even when the students don't completely
understand every detail and equation needed to describe the
situation. For instance, seeing a distillation column concen-
trate a dye from a feed stream is impressive and offers
motivation to learn why it works; seeing a microelectronic
part made or a bio reactor work can inspire the student to
learn more about complex reactions and mass transfer phe-
nomena. Connecting these exciting processes to the need to
learn about chemical engineering fundamentals can make
the student want to learn about these subjects. It seems to me
that there is no real reason why some of these practical
motivational aspects cannot be integrated more effectively
into our formal education.
I know that senior-level engineering courses demonstrate
some of these practical applications. In my senior design
class, we were assigned to work in groups of 3-4 and were
given a chemical to produce. Our assignment was to re-
search the chemical in patents and textbooks and even to
contact companies to develop a process for making the chemi-
cal. While designing the process, we had to perform energy
and mass balances to track our components in producing the
desired product at the necessary rate. We also had labs
where we had hands-on experience with reactors, distillation
columns, heat exchangers, etc. The point is that students
would benefit more from some type of practical-application
exposure throughout their education, rather than packing it
all in at the end of their schooling. Even short demonstra-
tions and group projects, when done earlier in the curricu-
lum, would be more satisfying and valuable.
Can something be learned and transported back to campus
from industrial, co-op, and intern experiences to enrich the
learning experience of all students-even those who never
leave campus? I believe so-especially in the area of com-
puter skills and the need to better integrate their develop-
ment throughout the curriculum, principally through the use
of homework assignments.
There are at least four computer programs that are essen-
tial for a practicing chemical engineer to be familiar with:
Excel, Aspen, PowerPoint, and Word. Most employers, even
those found in co-op jobs, expect an employee to be compe-
tent with Excel. It seems to me that skills taught on campus
in the use of these software programs should lead, rather
than lag behind, those that are demanded in an industrial
setting. This is often not the case, however.

Chemical Engineering Education

Throughout their campus careers, students should build
steadily on the sophisticated application of tools such as
Excel. A good practice to follow is to teach the use of Visual
Basic programming-that skill can then be used with Excel
in later courses after students master the simple elements of
Excel in their freshman course. While co-oping, I often
needed to add data to a spreadsheet with over 700 rows, and
it goes without saying that editing a spreadsheet that long
can be very time-consuming. But (fortunately) an engineer
in my group showed me some simple programming in Vi-
sual Basic that showed that deleting spaces between data on
such a huge spreadsheet need not be done by hand at the
expense of hours. From my co-op experiences, I learned
that being a good engineer requires being able to opti-
mize time expenditures. Spending hours editing spread-
sheets that can otherwise be done more efficiently is not
good time management.
Besides learning basic spreadsheet skills, students should
learn how to analyze large data sets. During two of my co-op
terms, I had to analyze someone else's data, even though I
didn't know all the details about the process related to the
data. I now understand that one doesn't need to be an expert
in all of the process details to gain insight about why some-
thing is not going well with the process. A systems engineer-
ing perspective should be introduced and emphasized early
in the curriculum in my opinion. Assignments should be
given where only a brief description of the process is pro-
vided with a spreadsheet containing a large amount of data.
The assignment should be to analyze the data and then
decide which data should be considered. I believe that
most practicing engineers would agree that statistics
(which is necessary in data analysis) should be required
for all engineering students.
In addition to Excel, Aspen (the unit operations simulator)
is a key program and should be integrated into more of the
chemical engineering curriculum. I had two weeks of Aspen
during my second engineering class, and after that class, I
never used it again in school. Perhaps it will surface again in
the final design course, but Aspen would have fit well in
several courses prior to that. Ideally, assignments could be
given requiring integration of Aspen and Excel-this does
happen in the real world of co-ops.
Two additional easy-to-use tools, PowerPoint and Word,
also deserve some consideration. Both of these programs
can be used by anyone who can turn on a computer, but to
use them effectively takes practice. Given the intense pace in
industry, imprecise communication guarantees that prob-
lems will occur. At the end of each of my rotations, I had to
give a formal presentation in front of my peers, hiring man-
agers, and whoever else wanted to attend. It takes experience
to organize and present information in front of a large audi-
ence when you are under a spotlight and have to use a
microphone, but the thought that a good idea that can't be

communicated is wasted must always be at the forefront
when planning one of these presentations. I must also add
that hearing your own voice echo without quivering in
such a situation gives you confidence in your ability to
give a presentation anywhere and to anyone after that, so
the effort is worthwhile.
Before closing, I would like to mention an additional
experience vehicle that already exists on campuses that could
be invaluable in assisting the student's metamorphosis to
engineer. Increasing student-faculty interactions may sound
like a simple idea, but while it may be obvious, it is not
simple to do well. Our student chapter of the AIChE had a
program in which juniors were paired with seniors in their
major, and seniors were paired with a professor to create a
mentor-mentee relationship. The problem with this simple
but excellent idea is that often nothing happened. For
instance, when I participated in it, a meeting was set up
for students to meet the professors who had been as-
signed to them as their mentor-but none of the profes-
sors showed up!
We all accept that there are many demands on a professor's
time, and such a mentorship program is simply one more
demand that the professor didn't volunteer for, which prob-
ably places it low on the priority list. Nevertheless, if profes-
sors more actively participated and actually showed interest
in programs such as this, students would benefit enormously.
Ideally, the contact would provide exposure to the professor's
research program, and could possibly include some actual
research experience for the student. This experience could
give the student a more balanced perspective, enhancing the
industrial co-op and internship experiences, to draw upon
when considering the options available after graduation.
It took a bit of determination on my part to find that such
an "on-campus co-op" experience was available. After my
industrial co-op, I eventually worked for one of my engi-
neering professors over the summer. The project involved
growing bacteria that harbor a specific esterase that we wanted
to recover. The professor took an active interest in my career
and educational goals, which was an attitude that I had not
seen in a teacher since high school. Such a personal relation-
ship goes a long way toward allowing students to discover
what they really want to do in life and what they need to do
in order to develop the skills for pursuing their goal. I found
that simply having an intelligent, experienced, and patient
individual to bounce questions off of was enormously help-
ful in allowing me to frame my long-term plans. In fact,
this role is the common element present in both success-
ful industrial or successful campus co-op mentoring. This
component catalyzes the organization of all the facts and
experiences that are jumbled together in a typical "stan-
dard" undergraduate experience. It is one truly essential
element that turns a bewildered student into a self-as-
sured graduate engineer. 7

Winter 2001

, M. classroom



Notes From a Novice Teacher

Universidad Nacional Aut6noma de Mexico Mexico D.F., Mexico

Professors are faced with outstanding-, average-, and
poor-performance students every day in every class,
but it seems that only a few try to understand what is
wrong with the lower-end students and do something to help
them. The natural tendency is to think of them as lazy or
immature, or perhaps to view them as students who are in the
wrong discipline. While some of them might indeed be
immature or lazy, or misplaced, others could be failing courses
in spite of their best efforts and are often in danger of
dropping out of the course, or even dropping out of school.
When I think back to my days as an undergraduate and
remember some of the smart and capable classmates who
dropped out of school, I marvel that I managed to continue
and to succeed when they did not. While the reasons are
many and complex, some thoughts come to mind immedi-
ately-I was lucky enough to have several excellent teachers
at different stages in my education, and they had a great
influence on my academic and professional choices. I was
exposed to different teaching styles throughout my college
years and was able to take the best from each of them.
Some of my friends and relatives had the misfortune of
having extremely poor teachers or badly organized courses
at crucial stages of their education, with the result that they
became discouraged or confused and eventually lost the will
to continue their education. One friend from high school
who had extremely good grades, dropped out of the univer-
sity after one term in chemistry. She found that the imper-
sonal teaching style of her college professors contrasted
poorly with the cooperative teachers we had in high school,
and it made a big difference in her own attitudes. While she
was doing well (better than average), she was frustrated and
none of the professors took the time to give her the reassur-
ance she needed.

I am frequently impressed with the insight and intuitive
understanding my own wife has of physical and chemical
phenomena, in spite of her proclaimed dislike of "theoreti-
cal" explanations. She could have been an excellent chemi-
cal engineer, but she had "boring" chemistry teachers in high
school and then in college she came face to face with the
attitude (still prevalent) that women are not good at sci-
ence-so she chose a different educational path. There are
still professors who have different standards for men and
women, claiming that "female chemical engineers will most
likely end up staying home" while male chemical engineers
would "most likely become field practitioners" and thus
require a much more rigorous education. The sad truth is that
many students with the potential to succeed are discouraged
from doing so. In Felder's words,t1l "There is nothing wrong
with the raw material" (when talking about the quality of
American students)-it seems that the educational system
itself has been adversely affecting many of them.
This situation does not have to continue. Professors can do
many things to promote students' learning and to help them
discover the best in themselves. Professors should focus on
motivating students, adapting their teaching styles as neces-
sary to nurture and promote the intellectual growth of their

Copyright ChE Division of ASEE 2001
Chemical Engineering Education

Eduardo Vivaldo-Lima received his BSc in
chemical engineering from the National Au-
tonomous University of Mexico (1989) and
his MEng and PhD, both in chemical engi-
neering, from McMaster University. He has
had industrial experience as a Consultant
and as Supplier Quality Engineer for Gen-
eral Motors of Mexico, and has been an
External Instructor at the Universidad LaSalle,
Mexico (1991) and a Teaching Assistant at
McMaster University (1994-1997).

students. They should strive to help students rise to the top.
This is easily said, but how can it be accomplished? To
answer that question, a short literature review on motivation,
learning, and teaching styles follows, and a compilation of
recommendations from distinguished chemical engineering
educators is given. Finally, my personal perspective will be
given in two ways: first, the description of my strategy for
effective teaching, despite being new in the profession, and
second, a retrospective analysis of the experiences gained
during my first two years as an associate professor at the
National Autonomous University of Mexico.

It is usually accepted that motivated students
are easier to teach-that students who are in-
terested in learning do, in fact, learn more.
Professors should encourage and take advan-
tage of students' motivation to learn. An opti-
mal motivational level should be sought and
established right from the course beginning.
Since motivation and commitment are personal
matters, what can be done to motivate the class
as a whole? Cashin[2] proposes capitalizing on
the students existing interests; relating the
course to the students' interests whenever pos-
sible; explaining in detail the relevance of the
course, using problems, case studies, examples,
etc.; discussing the ways in which the teacher
finds the course interesting; finding out which
topics are of most interest or value to the stu-
dents; including some optional or alternative
units; encouraging alternative learning meth-
ods (e.g., lectures, discussions, independent study, etc.) when
Ericksen131 notes the negative effect on motivation when
there is a conflict between the teacher and students over
course objectives and content. In small classes it is not
particularly difficult to negotiate this conflict, but in large
lecture courses the objectives are operationally defined by
the procedures used in evaluation and grading. In large classes,
the teacher makes most of the decisions as to what is learned
and when it is learned. The principle of a performance-
oriented course is not compromised and the advantage of
participation is gained when students mark out a subset of
objectives with which they want to become involved. Initial
responsibility for drafting objectives, however, rests with
the teacher, who has the knowledge of both the discipline
and the available resources.
The learning environment present in a university can shift
students from an interest focus to a course focus. In other
words, students who are used to a high level of commitment
and have high expections for intellectual stimulation can be
disappointed with the actual levels of each course offered in
Winter 2001

their university, to the point that getting "passing grades"
becomes their main objective.['4
Professors have to deal with a wide range of students;
some are highly motivated and will succeed regardless of the
subject matter, teaching effectiveness, or class size. But in
nearly every class there are also those students who are "at
risk," who suffer from poor motivation, low self-esteem, or a
low sense of personal control. Menec and Perry'51 propose
two ways to help these students: (1) professors should func-
tion as attributional retrainers; i.e., they should modify stu-
dents' (mis)perceptions of their successes and failures by
using an attributional retraining technique, and
(2) there should be a proper use of comments
about students' performance and abilities.

Students learn in many different ways. Some
prefer abstraction, others prefer facts. Some
want details, others want the entire picture.
Some think in terms of pictures and charts,
others in terms of words, and still others in
terms of symbols and equations.161 Every hu-
man being has a preferred style. Woods161 has
recently published a detailed and interesting
review on personal preferences in learning
and provides a number of ideas on how to
improve learning by accounting for the
student's personal preference in learning.
Felder171 has also critically reviewed the lit-
erature on learning and teaching styles. More-
over, he stresses the importance of helping students build
their skills in both their preferred and less-preferred modes
of learning. To do so, it is important to teach using tech-
niques or material adequate for all the categories of learning
styles. This is called "teaching around the cycle."'71
Different authors classify learning styles in slightly differ-
ent ways. Woods161 proposes three categories to classify
learning styles. In the first category he includes those ele-
ments of learning style that are "robust," those that cannot be
changed: (1) Jungian or Myers Briggs with their four dimen-
sions for processing information; (2) visualizer/symbolizer/
verbalizer, or drawing/equation/words (DEW), which refers
to the preference for memorizing and thinking about infor-
mation; and (3) Kirton's adaptor-innovator options for ap-
plying creativity. In a second category, he includes those
styles that may or may not be possible to change: (4) serial
versus holistic preference for processing information; and
(5) inductive versus deductive preference for processing. In
a third category he includes those styles that can be changed
through training or experience: (6) Piaget's levels of devel-
opment; (7) Perry's levels of attitude toward learning; (8)
Kolb's learning cycle; and (9) Entwistle and Ramsden's

deep-versus-surface processing.
The explanation of these different learning styles and ex-
amples of how they manifest among students can be found
elsewhere.17-121 What is worth mentioning here is Woods' list
of ideas on how to facilitate learning in a way that takes into
account learning preferences.171 To develop awareness, he
proposes we
Identify our own learning preferences since teaching and
testing will be highly influenced by our own style
Help students identify their own learning preferences and
discuss the implications of each style
Help students become more understanding of learning-style
To cater to individual student's learning styles, Woods pro-
When presenting information, forcing oneself to present the
key concepts of the course in different styles
Impossible, sectioning classes to match learning styles
between teachers and students
Writing course material in a way that it appeals to different
Allowing different required texts
Identifing the learning styles of colleagues and recruiting
their help in the creation of exams, weekly assignments, and
the review of course texts
Having learning-style ombudspeople in the class to help the
professor prepare and assess class material
During assessment, providing flexibility in exam questions
Whenever possible, working with smaller groups and using
co-operative learning (with group membership influenced by
knowledge of students' learning styles)
Indivualizing teaching
Individualizing testing
Using the Keller "Proctorial System of Instruction" (PSI)
To develop students' skill and flexibility in appreciating
differences, Woods additionally recommends that we
Plan activities to explicitly help develop those elements of
style that are less robust and create learning environments
to promote a target style
Use small-group "self-directed learning" (SDL) or "prob-
lem-based learning" (PBL) with smaller (5 or 6) groups of

Junior professors might feel tempted to try some of the
ideas suggested by Woods right away and to make changes
in their teaching styles without full consideration. Although
it is true that favoring one's own learning style by teaching
in that style could be detrimental to students with different
styles, it is also possible to create conflict with a higher
proportion of students if they are not properly informed and
involved in the change process.

There are several cases reported in the literature where
conflict arose when changes were made to the instructor's
teaching style. Saunders"4' noted the case of reluctant be-
havior of some participants in role-play simulations. Ander-
son and Adams'8" report conflict when students experience
teaching styles that do not match their expectations. Miller,
et al., "' mention some of the disadvantages of active learn-
ing, such as poor structure or lack of structure. The dynamics
of interpersonal relationships in co-operative learning groups
and lack of individual responsibility on the part of group
members are potential sources of conflict, even in carefully
structured groups. Woods"[61 explains the problems (and so-
lutions) encountered when incorporating PBL components
(two courses) into the otherwise traditional chemical engi-
neering program at McMaster University. In the paper he
mentions nine issues of concern; one of them is gaining
student acceptance.
A common remedy to eliminate or attenuate the conflict is
to clearly inform the students of the objectives and benefits
of the new approaches and to explain how they can manage
conflict-prior to making the changes. In some cases it is
important to previously train the students on necessary
skills, such as problem-solving, interpersonal and group
behavior, and learning approaches."6] Some students may
be so reluctant to change that the use of different teach-
ing styles and even different (yet known) assessment
procedures may be needed.

There are many things that can be done to accommodate
differences in learning styles, levels of motivation, levels of
intellectual development, and judgment modes among stu-
dents. One approach is to try to follow the "seven principles"
for good practice in undergraduate education:r'71
Encourage contact between students and faculty
Develop reciprocity and cooperation among students
Use active learning techniques
Give prompt feedback
Emphasize time on task
Communicate high expectations
Respect diverse talents and ways of learning

Wankat1 8I proposed "ten learning principles." The first six
of them coincide with the previous principles, and the re-
maining four are:
Develop a structured hierarchy of content and guide the
Develop images and use visual modes of learning
Challenge the students
Ifpossible, separate teaching and evaluations
Similarly, Bird"91 also proposed "seven rules" for teaching:
Do not show off
Chemical Engineering Education

Do not bluff
Do not intimidate
Do know what you are going to teach
Do know why you are going to teach it
Do know how you are going to teach it
Remember that the teacher's job is to serve the student
(keeping in mind that serving does not mean being "crowd-
pleaser" or rewarding poor performance).

Felder and Silverman[91 proposed several teaching techniques
to address all learning styles. They include
Motivating learning
Providing a balance of concrete information and abstract
Balancing the material that emphasizes practical problem-
solving methods with material that emphasizes fundamental
Providing explicit illustrations of intuitive and sensing
Following the scientific method in presenting theoretical
Using graphical material (pictures, plots, figures) before,
during, and after the presentation of verbal material
Using computer-assisted instruction
Providing (short) active intervals during lectures
Giving the students the option of cooperating on homework
assignments to the greatest possible extent
Applauding consistent effort
Talking to students about learning styles

The comfort level of feelers (in the Jungian typology) in
technical courses can be raised by (a) bringing out the social
relevance of the course material, (b) addressing some non-
technical topics, and (c) using student-centered instructional
approaches such as cooperative learning.'"
To promote a deep approach to learning, the following
conditions should be present:"" student-perceived relevance
of the subject matter; clearly stated instructional objectives,
practice, and feedback; appropriate tests; reasonable
workload; and choice over learning tasks.
To help the students move up the intellectual-development
ladder, it is necessary to provide an appropriate balance of
challenge and support, occasionally posing problems one or
two levels above the student's current position. Instructors
should assign open-ended, real-world problems throughout
the curriculum, but should not make course grades heavily
dependent on them. Providing feedback on performance with
these types of problems is very important.1121
Felder, et al.,120.21] and Stice, et al.,[221 have prepared a
comprehensive study of teaching methods that promote learn-
ing and how to train teachers to apply those methods. The
emphasis is on engineering courses, but most of the ideas are
also applicable to other disciplines.
Winter 2001

A junior professor might be overwhelmed by the amount
of information cited in this paper for improving student
learning by accounting for personal preference and imple-
menting changes in their style of teaching. Even some ex-
perts on education recommend that junior professors in re-
search-intensive institutions give priority to getting started
with their research projects and obtaining funding; that they
start modestly in their first teaching efforts, trying to get as
much help as possible from senior professors when it comes
to course design and material.231
But what about those students in our classes during our
first few years in academia? Is it okay to "lose" some of
them during this early period of learning to teach? The
answers to these questions should be answered by the new
professors themselves. My first-year-as-a-professor answers
are provided in the following paragraphs.
I envision teaching as a continuous attempt to promote
students' growth, both intellectual and personal, by using
optimal mixtures of the three components in play: the stu-
dents themselves, the teacher (myself), and the subject disci-
pline or course topic.
As to the first ingredient of the learning mixture-the
students- there are several aspects that I try to keep in
mind. The first is that students are human beings, and as
such they deserve respect and consideration. The second
important aspect is to recognize that students have different
learning styles, attitudes, and motivations for learning. To
get the best out of them (to be used in the "learning mix-
ture"), it is necessary to motivate them (to an appropriate
level), and to be able to adapt my teaching style to their
learning styles, or to convince them of the convenience of
being exposed to learning styles not initially compatible
with theirs. Whenever possible, I try to allow my students to
negotiate learning objectives and the assessment procedure.
The second ingredient in the mixture is the teacher-
myself. Although most of my professional life has been
focused on industrial engineering practice and applied re-
search in polymer reaction engineering, I have always con-
sidered teaching as a noble and important activity. Much of
what I am today was inspired by former teachers of mine and
my parents. My love of mathematics, chemistry, and physics
(key components in chemical engineering) was triggered
with the help (and promoted by the example and teaching
styles) of several excellent professors. Many of my teaching
strategies were stolen from those teachers.
Some of the positive early aspects of my teaching style
have been my respect for students (as a student, I was turned
off by those who made students feel less intelligent, mature,
or trained) and responsibility in my activities. As a teaching
assistant at McMaster University, I always tried to fully

understand the material before a tutorial took place or when
talking to students (most of the time I tried the solutions to
assignments and exams myself, even when the solutions
were available from the instructor). When I did not know the
answer to a student's question, I tried to find the answer and
get back to the student as soon as I could. Whenever pos-
sible, I tried to design assignments that were connected to
actual problems and situations that the students would face
in their professional lives.

Teaching is sufficiently important th
need to continuously refine it through
systematic and scholarly approach.
That is the very least our students
and our society deserve.

Despite the fact that I followed the example of inspiring
teachers in my early teaching experiences, I discovered that I
was not as effective as I had wished to be. Being somewhat
introverted and a "feeler" (in the Jungian typology), I real-
ized that my teaching style was biased toward students with
learning styles similar to mine. I did not use many group
activities, my lecture material was abstract (mostly text,
equations, and diagrams), and I used to speak softly. I knew
something was not working the way I wanted it to. In this
area, I received some valuable help from a credit course I
decided to take while I was a graduate student (titled "Prin-
ciples and Practice of University Teaching").
The last ingredient in my teaching mixture is the course
subject. If I want to teach something, I should understand it
deeply and I should like it. I should be able to make it
appealing and interesting to the students. I should be able to
make the students understand why it is worth studying the
discipline. To that end, I spent three years as a practitioner in
order to garner first-hand experience in an actual work envi-
ronment. To help my understanding of the area, I enrolled in
graduate school where I obtained my MEng and PhD de-
grees. Throughout this time, I detected areas of opportunity
for research and technological improvement and learned
many things that I would like to share with students. I may
even be able to help some individuals to find and exploit (or
reinforce) their potential as professionals in this area.
In short, a young professor can improve student learning
by liking to teach, establishing interesting and novel re-
search lines, providing multiple real-life applications of top-
ics taught in the course, taking advantage of the instructional
development courses and facilities where and when avail-
able, and never ceasing to learn.

After two years of research and teaching experience as

Associate Professor of Chemical Engineering, I can add
some additional thoughts on the subject. It is not easy to be a
"quick starter," even if one tries to follow Brent and
Felder's[241 advice on how to become one, or even if one
reads what Felder1251 wished he had been told.
I still made mistakes or inadequate choices during my first
two years as a professor. In my first semester I offered a
graduate course with "cutting edge" content and found I had
to spend an average of six hours of prepara-
tion time per lecture hour (in the second
at we semester I reduced that to a two-hour prepa-
rh a ration per lecture hour, but was responsible
for three courses). I put too much emphasis
on the course content at the cost of little
active student participation. As a thesis su-
pervisor I gave too much freedom to my
undergraduate and graduate students (I my-
self was the independent type who did not
like my supervisors over me all the time) until I realized that
most of them liked and needed closer supervision. I did not
start manuscript writing early, so I had to spend several
weeks of twelve to fifteen working hours in order to reach
my own research goals during my first year as a professor.
Not everything could be classified as a "failure," however.
I also implemented some good ideas from the experts.122-241
tried to incorporate much of my own research into my course
materials; I started several collaboration projects with expe-
rienced researchers from industry; I asked colleagues to read
and provide the toughest critique to my research proposals
(so far, funding has been granted in all cases); I applied for
course load reduction (standard optional procedure for the
first two years of new faculty) in order to concentrate on
research funding activities, etc.
In order to motivate students, the teacher does not neces-
sarily have to be easy-going and charismatic (although it
helps). Even introverted or absent-minded teachers can mo-
tivate students. In my courses I try to motivate the students
in a number of different ways. For instance, as mentioned
before, I use my research projects to provide examples of
material covered in the course. This helps make matters
more interesting and, at the same time, it helps attract stu-
dents to my research projects. Sometimes there may be
topics in the course that I do not feel at ease explaining (in
advanced or graduate courses, for example)-in those situa-
tions, in addition to spending time reading and updating my
knowledge, I try to invite other professors or respected in-
dustry researchers to give short lectures within the course.
Students usually respond positively to these special lectures.
For one month during this school year I had a distin-
guished professor from Canada who participated in some of
my courses and co-supervised some of my graduate stu-
dents. They already know some of his work since I use it in
my courses, and knowing he would be here had a positive
Chemical Engineering Education

effect on my undergraduate and graduate students. They
waited anxiously for his visit and felt they would learn
something new and useful while he was here, which they
indeed did. I am also arranging the visit of another well-
known and well-respected Canadian professor next year who
will co-chair with me the organization of a microsymposium
within a major conference in my field.
My students, both undergraduate and graduate, are aware
of my efforts to establish a research group. In a department
where the average age of the professors is over 45, I find that
some of the undergraduate students are eager to get involved
in projects significantly different from the more traditional
ones. Graduate students are more reluctant to change, though,
and they usually prefer to work for senior professors. None-
theless, they also acquire a high degree of motivation from
courses where content is related to research as much as
possible. Innovation and change gets students' attention and
encourages them to go beyond the contents of textbooks.
In other words, young professors can concentrate on set-
ting up their research programs (what counts for promotions
and stimuli in our academic world) and still be able to
promote learning in their students by sharing with them the
enthusiasm for this activity. Students who get involved with
actual research projects in their courses develop a higher
degree of motivation than students who just follow the book.
Students are individuals, and as such they have different
learning styles, motivations, and expectations for learning. It
is very important that when implementing "non-traditional"
teaching methods in their courses, professors let their stu-
dents know the objectives and the value of the new teaching
approaches. It helps to stress the usefulness of these ap-
proaches. There may be students who remain reluctant to
accept them, even after the purposes and benefits of non-
traditional techniques have been explained. The instructor
should be prepared to use alternate teaching styles or have
alternate assessment procedures (known to the student).
Students are a precious resource that should not be squan-
dered through ineffective teaching. Teaching is sufficiently
important that we need to continuously refine it through a
systematic and scholarly approach. That is the very least our
students and our society deserve.

Financial support as a graduate student at McMaster Uni-
versity and for research activities as Associate Professor at
UNAM from the Science and Technology National Council
(Conacyt) of Mexico is acknowledged. Financial support
from DGAPA UNAM for research activities is also ac-
knowledged. The detailed proof-reading and critique of the
first version of this manuscript by Professor Alexander
Penlidis (University of Waterloo, Canada) and Mr. Dale Roy
(Instructional Development Center of McMaster University)
is gratefully acknowledged.
Winter 2001

1. Felder, R.M., "There's Nothing Wrong with the Raw Mate-
rial," Chem. Eng. Ed., 26(2), 76 (1992)
2. Cashin, W.E., "Motivating Students," Idea Paper No. 1,
Center for Faculty Evaluation and Development, Division
of Continuing Education, Kansas State University; August
3. Ericksen, S.C., Motivation for Learning: A Guide for the
Teacher of the Young Adult, The University of Michigan
Press, Ann Arbor, MI (1974)
4. Gibbs, G., Teaching Students to Learn: A Student Centered
Approach, The Open University Press, Milton Keynes (1981)
5. Menec, V.H., and R.P. Perry, "Disciplinary Differences in
Students' Perceptions of Success: Modifying Misperceptions
with Attributional Retraining," New Direct. for Teach. and
Learn., 64, 105 (1995)
6. Woods, D.R., PS News No. 104, Department of Chemical
Engineering, McMaster University, Hamilton, Ontario,
Canada; May-June (1996a)
7. Felder, R.M., "Matters of Style," ASEE Prism, 6(4), 18 (1996)
8. Anderson, J.A., and M. Adams, "Acknowledging the Learn-
ing Styles of Diverse Student Populations: Implications for
Instructional Design," New Direct. for Teach. and Learn.,
49, 19(1992)
9. Felder, R.M., and L.K. Silverman, "Learning and Teaching
Styles in Engineering Education," Eng. Ed., 674, April (1988)
10. Felder, R.M., "Meet Your Students: 3. Michelle, Rob, and
Art," Chem. Eng. Ed., 24(3), 130 (1990)
11. Felder, R.M., "Meet Your Students: 5. Tony and Frank,"
Chem. Eng. Ed., (1995)
12. Felder, R.M., "Meet Your Students: 7. Dave, Martha, and
Roberto," Chem. Eng. Ed., 31(3), 106 (1997)
13. Keller, F., "Goodbye Teacher," J. Appl. Behavior Analy., 1,
79 (1968)
14. Saunders, D., "Reluctant Participants in Role Play Simula-
tions: Stage Fright or Bewilderment?" Simulation/Games
for Learning, 15(1), 3 (1985)
15. Miller, J.E., J.E. Groccia, and J.M. Wilkes, "Providing Struc-
ture: The Critical Element," New Direct. for Teach. and
Learn., 67, 17 (1996)
16. Woods, D.R., "Problem-Based Learning for Large Classes in
Chemical Engineering," New Direct. for Teach. and Learn.,
17. Chickering, A.W., and Z.F. Gamson, "Seven Principles for
Good Practice in Undergraduate Education," AAHE Bulle-
tin, 39(7), March (1987)
18. Wankat, P.C., "What Works: A Quick Guide for Learning
Principles," Chem. Eng. Ed., 27(2), 120 (1993)
19. Bird, R.B., "Seven Rules for Teaching," Chem. Eng. Ed.,
27(3) 164 (1993)
20. Felder, R.M., D.R. Woods, J.E. Stice, and A. Rugarcia, "The
Future of Engineering Education: Part 2. Teaching Meth-
ods that Work," Chem. Eng. Ed., 34(1), 26 (2000)
21. Felder, R.M., D.R. Woods, J.E. Stice, and A. Rugarcia, "The
Future of Engineering Education: Part 3. Developing Criti-
cal Skills," Chem. Eng. Ed., 34(2), 108 (2000)
22. Felder, R.M., D.R. Woods, J.E. Stice, and A. Rugarcia, "The
Future of Engineering Education: Part 4. Learning How to
Teach," Chem. Eng. Ed., 34(2), 118 (2000)
23. Woods, D.R., "The Four Dimensions of an Academic," inter-
nal seminar organized by the Department of Chemical En-
gineering, McMaster University, Hamilton, Ontario, Canada;
Summer (1997)
24. Brent, R., and R.M. Felder, "The New Faculty Member,"
Chem. Eng. Ed., 32(3), 206 (1998)
25. Felder, R.M., "Things I Wish They Had Told Me," Chem.
Eng. Ed., 28(2), 108 (1994) 0

es Oclassroom



And Its Implementation by Meta-Computing Software

University of Guelph Guelph, Ontario, Canada N1G 2W1
Academy of Sciences 165 02 Prague 6, Czech Republic
University of Toronto Toronto, Ontario, Canada M5S 3E5

ne way of using an equation of state (EOS) for pure
fluids is to determine the EOS parameters by fitting
them to experimental data for each individual fluid.
Instead of a particular parameter set for each fluid, in a
variant of this approach, an EOS is extended to broad classes
of fluids by expressing at least one of the parameters in
terms of critical properties, Pc and T,, and at least one addi-
tional parameter such as the acentric factor, o. The result-
ing generalized form of EOS, while less accurate for each
individual fluid, is intended to provide a compact and rea-
sonably accurate representation of the volumetric (and ther-
modynamic) properties of the entire class of fluids.
An important aspect of chemical engineering education is
the study of this strategy of EOS construction and generali-
zation, and its extension to fluid mixtures. A recent example
of this strategy for pure polar and nonpolar fluids and their

mixtures is described by Platzer and Maurer.[1"21
A different and less common way of using an EOS for
classes of fluids is to follow an approach originally proposed
by Pitzer. It uses parameter sets corresponding to accurate
representations of the behavior of selected reference fluids,
and then approximates the properties of a class of fluids by
incorporating an additional parameter such as o in the form
of corrections to the principle of corresponding states, in
terms of an expression for z(P,,Tr) (where T, = T/To is the
reduced temperature and P, = P/Pc is the reduced pressure).
In this article, we explore the general basis for, and the
utility of, this latter approach, which we believe has been
insufficiently exploited in the pedagogical literature.
The correlation of Lee and Keslert13 for volumetric and
thermodynamic properties of normal fluids, based on the
acentric factor ( o) of Pitzer, et al.,14 71 (PLK strategy), is the
most favored three-parameter corresponding states correla-
tion (Smith, et al.,'18 p. 88). In spite of this accolade, almost
all introductory thermodynamics books either ignore it or
deal only with its implementation in graphical or tabular

Such pedagogical treatments of the correlation for the
most part ignore the underlying basis, thus obscuring its
possible extension to other classes of fluids, and by default
its dependence on a particular representation of PvT behav-
ior. Such treatments also disregard the emphasis placed by
Lee and Kesler on an analytical implementation in a form
convenient for computer use. The graphical form is useful
for a qualitative representation of the behavior of pure fluids,

ChE Division of ASEE 2001
Chemical Engineering Education

William R. Smith is Professor of Engineering and of Mathematics and
Statistics at the University of Guelph. He received his BASc and MASc in
chemical engineering from the University of Toronto, and his MSc and
PhD degrees in applied mathematics from the University of Waterloo. He
is co-author of Chemical Reaction Equilibrium Analysis (1982, 1991). His
research is in classical and statistical thermodynamics.
Martin Lisal is Researcher at the E. Hala Laboratory of Thermodynam-
ics of the Institute of Chemical Process Fundamentals, Academy of
Sciences, Prague, Czech Republic. He received his MSc in mechanical
engineering and his PhD in thermodynamics from the Czech Technical
University in Prague. His research is in classical and statistical thermody-
Ronald W. Missen is Professor Emeritus (chemical engineering) at the
University of Toronto. He received his BSc and MSc degrees in chemical
engineering from Queen's University and his PhD in physical chemistry
from the University of Cambridge. He is co-author of Chemical Reaction
Equilibrium Analysis (1982, 1991) and Introduction to Chemical Reaction
Engineering and Kinetics (1999).

but both tables and charts are inadequate for the best quanti-
tative results. The extension of the PLK strategy by Teja, et
al.,[19'31 (PLKT strategy), applicable to broader families of
fluids, is not discussed at all in introductory texts.
The somewhat complex computer implementation neces-
sary at the time of its original development has probably led
to emphasis on the use of graphs and tables in teaching the
PLK approach. But current computer implementation tech-
nology has advanced considerably beyond what was avail-
able twenty-five years ago. Implementations are now avail-
able in the form referred to by Edgarr[4' as "meta-comput-
ing" software, involving the use of packages such as Maple,1"51
Mathematica, 16' Mathcad,117] and MATLAB.[11s
We agree with Sandler[19'20 that this type of software is
especially useful in a pedagogical setting, since in addition
to allowing the treatment of problems previously considered
too complex at the undergraduate level, it "can let the stu-
dent concentrate on the subject matter at hand ., rather
than being distracted by computational methods, algorithms,
and programming languages."1201 This type of software is
becoming increasingly accessible to engineering undergradu-
ates and lends itself to efficient implementation of the PLK
and the PLKT approaches.
Our purpose in this paper is two-fold:
(1)To describe and emphasize the pedagogical impor-
tance of the PLKT strategy, both as a setting for
understanding the essential basis of the PLK strategy
for normal fluids and for extending it to other classes
of fluids.
(2)To describe an efficient analytical implementation of
the PLKT strategy using meta-computing software.
We first describe the Pitzer-Lee-Kesler-Teja (PLKT) strat-
egy and give an example of its application to two families of
non-normal fluids. This description points up the generic
nature of the original PLK strategy so as to remove any
dependence on a particular equation of state and choice of
reference fluids. We then describe, with an example, its
implementation using meta-computing software. Through-
out, we focus on representation of the compressibility factor,
z = PV/nRT, for pure fluids, but the determination of ther-
modynamic properties follows from this, as outlined, for
example, by Lee and Kesler.'31

In 1955, Pitzer, et al.,14'5] added a third parameter, co, to
the two-parameter (P,,T,) principle of corresponding states
for determining the thermodynamic properties of "normal
fluids." This was based on the concepts of
(1) "simple" fluids with spherically symmetric intermo-
lecular potentials/ shapes, and
(2) "normal"fluids with moderate departures from
simple-fluid behavior,
Winter 2001

and expressed as a linear relation for z in terms of 0o,

z(Pr,Tr;o)= z(0)(Pr,Tr)+ 0z(I)(Pr,Tr) (1)
where P,(=P/Pc) and T,(=T/Tc) are the reduced pressure and
temperature, respectively, and co is defined by

co = -loglo p;(Tr = 0.7)- 1.000 (2)
where pr is the reduced vapor pressure (=p*/Pc) of the
substance at Tr = 0.7. Since the analytical representation of
z'O)(P,,T,) and z''(P,,T) is not feasible, Pitzer, et al., provided
tables of their values based on analysis of experimental data.
Curl and Pitzer[211 provided a criterion for normal fluids
based on surface tension.
In 1975, Lee and Kesler,[31 seeking to improve the Pitzer
results for fluids involved in hydrocarbon processing, devel-
oped an analytical implementation for z(v, Tr) using a modi-
fied Benedict-Webb-Rubin (BWR) equation of state (EOS)122'
to represent the behavior of two reference fluids that served
to determine z") and z"' in Eq. (1). Lee and Kesler[3] de-
scribed a procedure to implement this strategy to obtain z for
a hydrocarbon fluid of interest at a given (P,T), and they
provided tables and charts of values of z(0' and z1' as func-
tions of (P,,T,). We call this the Pitzer-Lee-Kesler (PLK)
In the 1980s, Teja and co-workersl9-131 generalized the
PLK approach in three ways. First, they considered it as a
special case of interpolation/extrapolation involving z(P,,T,)
using two arbitrary, but conveniently chosen, reference flu-
ids, and thus extended its use to "families" of fluids other
than normal fluids. Second, they allowed the reference fluids
to be represented by any convenient EOS. Third, they ex-
tended their approach to mixtures and to other properties,
including viscosity,[23-26] surface tension,[271 and thermal con-
ductivity.[28] We call this the Pitzer-Lee-Kesler-Teja (PLKT)
strategy. No extensive published calculations show the util-
ity of this approach for pure non-normal fluids, perhaps
because the primary focus of their work was on mixtures.[29]
The basis of the PLKT strategy is to recognize that the
essential assumption underlying the Pitzer approach is that
z(P,,T,) is represented as a linear function of the acentric
factor (Eq. 1). To determine this linear relationship with 0,
we may select, from a family of fluids, two appropriate
reference fluids, r, and r, (according to some specified crite-
rion), with corresponding acentric factors o('r) and C(r2)
The equation for the linear z( o) relation for any member of
the family can be determined from the two points (o(r1), z(r)
and (co(2), z(r)) as

z(Pr,Tr,o))= z(r. )(Pr,Tr)+ Z-(r2 e )(p m)r )PrT
)(r2) --CO1'
Lee and Kesler used Eq. (3) for the family of normal

fluids, with the reference fluids r, and r2 chosen as a simple
fluid (o(r0)=0) and (essentially) n-octane (o(2 )=0.3978),
respectively. Equation (3), however, allows the use of any
two reference fluids within a family. The original PLK strat-
egy was developed for normal fluids; the following example
illustrates the appropriateness of Eq. (3) for two families of
non-normal fluids.

Example 1
(a) the family of halogenated hydrocarbon refrigerants, and
(b) the family of normal alkanols.
Investigate whether the PLKT strategy of Eq. (3) can be
applied to these fluid families.

(a) Figure 1 (similar to Figure 2 of Pitzer, et al."51) shows
values of z(Pr,Tr) for nine halogenated hydrocarbon refriger-
ants (ranging from C, to C4) as a function of o at four values
of (P,,T,), together with the corresponding results for normal
fluids. The z points for the refrigerants were calculated using
the NIST REFPROP software package;'321 the full lines are
least-squares fits through these points and are for compari-
son only. The dotted lines for normal fluids were calculated
using a quantitative numerical implementation of the PLK
strategy, as described in the next section. Figure 1 indicates
that the linear relationship of Eq. (3) holds for the family of
refrigerants and that it is somewhat different from that for
normal fluids (the agreement of the results at T, = 1.30 is
(b) Figure 1 also shows values of z(Pr,Tr) for the normal
alkanols from methanol (CH4O) to n-eicosanol (C20H420).
Experimental datat331 were used for z, (at P, = Tr = 1), and at
the other state points z was calculated for the first five n-
alkanols using the Patel-Teja EOS.135'361 The full line at P, =
T, = 1 is a least-squares fit through the points and is shown
for comparison only. Away from T, = 1, there is perhaps an
insufficient number of family members to draw a conclusion
(and the first member, methanol, is anomalous according to
the particular EOS used). But the data at T, = 1 indicate that
the linear relationship of Eq. (3) likely holds for the family
of n-alkanols and that it is also somewhat different from that
for normal fluids.
Finally, we remark that another way to extend the PLK
approach to non-normal fluids is to incorporate parameters
in addition to o. For example, if parameters ((o, ) are
used, z(Pr,T,) may be fitted to the plane through three points
corresponding to three reference systems

d (rle), (rl),(r )), z(r), ) E ( r3) i ), 3),s

and the analog of Eq. (3) is

z= Z +a (-(O(rl)) +b(-(rl)) (4)

where a and b are determined from the three reference sys-
tems. Essentially, this approach has been considered by Wu
and Stiel,1301 by Platzer and Maurer,111 and by Rowley and co-
workers,1311 who selected as reference systems two specific
nonpolar fluids and either water or methanol as the third
reference system. Platzer and Maurer1ll compared their imple-
mentation of this approach with the alternative approach of
expressing the EOS parameters in terms of Co and This
last approach is beyond the scope of this paper.

In the typical case when the reference fluid EOS is ex-
pressed in terms of v and T, implementation of the PLKT
strategy for calculating z or P, given (v ,T), is an explicit
calculation. In contrast, calculating z or v given (P,T) in-
volves an implicit calculation entailing the solution of cer-
tain nonlinear equations. The underlying structure of this

1.0 T,=1.50, P=2
0.9 o T,=1.30,P,2 .......
0.8 ...9. ....
0.7 -
T=1.15, P,=2
0.6 9 CQ .
z .. .
ST 1.00, Pr1
halogenated hydrocarbon refrigerants
01 o& n-alkanols
PLK strategy for normal fluids
0.0. .. .

0.2 0.3 0.4 0.5 0.6

0.7 0.8 0.9 1.0

Figure 1. Compressibility factor, z(Pr, T), as a function of
acentric factor, 0 for a family of refrigerants (filled circles),
some members of the family of n-alkanols (open circles
and triangles), and the family of normal fluids (dotted
lines). Where present, the full lines are linear fits to the
data points and are shown for comparison only. For the
refrigerants, z was calculated using the NIST REFPROP
software package;321 for the n-alkanols, at Tr = 1.00, experi-
mental critical data were used'33" for C1 to C20; at other
temperatures, z was calculated only for C1 to C5 using the
Patel-Teja EOS'35'36 and values of parameters provided;for
the normal fluids, the dotted line is the original PLK strat-
egy.'3' The refrigerants on the graph (in order of their o
values) are R13, R21, R22, R23, R125, R218, R134a, RC318,
R236ea; the corresponding order for the n-alkanols is Cl,
C6, C7, C11, (C5, C8, C9, C4, shown as one point), C10,
C13, C3, C2, C12, C14, C16, C17, C18, C20, C19, C15 at
T, = 1, and C1, C5, C4, C3, C2 at other temperatures.
Chemical Engineering Education

calculation is made transparent by the use of meta-comput-
ing software. We describe the approach in both cases and
provide an example calculation for the latter situation.
When the EOS for the reference fluids is given explicitly
in terms of v and T, we write
z = f(v,T; p) (5)
where p denotes a set of parameters that take on particular
values for individual fluids. The particular EOS that Lee and
Kesler used in the implementation of their approach'31 is
written in the form
z = f(v,Tr;p) (6)

where v' is the ideal-gas reduced volume defined by

Vr = c- ZcVr (7)

and vr is the actual reduced volume, v / vc. Now
Pv Pr Pcv Pr Pr
Z- = r = Vr -ZcVr (8)
RT Tr RTc T r T
Equation (8) shows that, in the use of an EOS to calculate
z at a given (P,,T,), the set of reduced variables {P,,T,, v; } is
more appropriate than the set {P,,Tr, vr }, since the former
requires neither an assumption about the constancy of zc nor
a knowledge of vc. Since the usual form of an arbitrarily
chosen EOS, Eq. (5), involves v, v must first be converted
to v' to use the PLKT strategy in conjunction with it. This is
a subtle point that is not apparent in the pedagogical litera-
ture; it affects the calculation of z both from a given (v ,T)
and from a given (P,T).

Calculation of z(v,T) via Eq. (5)

To calculate z(v,T) for a substance with critical constants
(P,,Tc) and acentric factor co, the following (explicit) equa-
tions are used when the EOS is expressed in terms of
(v,T)(i.e., via Eq. 5)

Z(ri) RT-( rT (9)

( (pr2)
z(r2) c v Tr (r2) ) (10)
Sr2 c

where v; and Tr are calculated using the properties of the
fluid of interest. The value of z is then obtained from Eq. (3).

Calculation of z(P,T) via Eq. (5)

To calculate z(P,T) when the EOS is expressed in terms of
(v,T), Eq. (5), an implicit calculation must be performed
involving the solution of nonlinear equations for the refer-
Winter 2001

ence fluids as follows:
1. Calculate Tr and P, using the properties of the fluid of

2. Calculate z(ri) as the solution of the nonlinear equation

fr T zT; ) T
z e f R c z, T; P(l
pZ n P

( la)

4. Calculate z(r2) as the solution of the nonlinear equation


( T(r2) T (
z=f RTc Tz,T;p(r
z= R pr2) P (r2

5. Calculate z from Eq. (3).

Calculation of z(v,T) and z(P,T) via Eq. (6)

When the EOS is expressed in terms of (v;,Tr) (as is the
case for the Lee-Kesler EOS[31), Eqs. (9,10) and (lla,l lb)
are, respectively,

z(rl) =f(v ,Tr;p(r)) (12)

z(r2) = f(vr,Tr;P(r)) (13)

for z(v,T), and

(zT )
z = Tr ,Tr;P(r)) (14)

z=f ,Tr;p(r2) (15)

for z(P,T).
The calculation procedure is illustrated in Figure 2 (next
page), which shows a Maple"51 script for calculating z(P,T)
using the PLKT strategy in conjunction with an EOS ex-
pressed in terms of (v,T). To make the approach itself
transparent, a simple technique is incorporated into the script
to calculate only the largest value of z; this suffices for the
supercritical case, but for the subcritical case only the
"vapor-like" root is found. It is left as a student exercise to
modify the script to calculate the appropriate result in any
given circumstances, which requires either a (somewhat com-
plicated) calculation of the vapor pressure using Eq. (3), or
use of the vapor pressure correlation of Lee and Kesler.131
Corresponding scripts for Mathematica,'16' Mathcad,"17' and
MATLAB'181 can be obtained from the web site at http:// Scripts for the cal-
culation of z given an EOS expressed in terms of (v; ,T,) are
also available at this location. Use of the Maple"15 script is
illustrated by the following example.
Example 2
n-propanol (C3HgO) is to be stored in a 200-liter cylinder

at 2300C. What is the maximum amount (kg) that can oc-
cupy the cylinder as vapor? For n-propanol, P, = 5170 kPa;
T, = 536.71 K; co = 0.628; M = 60.10; and p*(2300C) =
2996 kPa. (Data are from Yaws,1331 except for the vapor
pressure, which is from the DIPPR Student Chemical Data-
base web site.1341) Use the PLKT strategy with ethanol (m(rl)
= 0.637) and n-pentanol (m(r2) = 0.594) as reference fluids r,
and r2, respectively, and the Patel-Teja EOS'35'36] in conjunc-
tion with Eq. (3). Values of the parameters p(1), p(r2) are
given by Patel and Teja.
For the amount of vapor to be a maximum, the highest
pressure is P = p*(2300C) = 2996 kPa. The amount is

p* VM
m=nM= RVM (16)
All quantities in Eq. (16) are known except z(p*,T). Using
the PLKT procedure described above and the Maple script
shown in Figure 2, we obtain the values
z = 0.6279 m = 13.71 kg
(For comparison, direct use of the Patel-Teja EOS[35,361 gives
z = 0.6283 and m = 13.70 kg.)

1. We believe that it is important in teaching the PvT
behavior of fluids to emphasize the general basis for
the Pitzer-Lee-Kesler-Teja (PLKT) strategy as an

# PLKT strategy for z(Pr,Tr) using Patel-Teja EOS for n-alkanols
# Reference fluids are ethanol, n-pentanol
# Example 2 re n-propanol
if Tr>1 then
while x-zPT(x*R*Ti/P1,Tl,al*alpha(F1,Tr),bl,cl)> 0 do x:=x-0.01 end do;
while x-zPT(x*R*T2/P2,T2,a2*alpha(F2,Tr),b2,c2)> 0 do x:=x-0.01 end do;
end if;

Figure 2. MAPLE"1I script for Example 2. This and scripts in Mathematica,'6"
Mathcad,t'7 and MATLAB"i' can be obtained from the web site at
Chemical Engineering Education

implementation of the three-parameter principle of
corresponding states. This makes clear the underlying
basis for the Pitzer corresponding-states approach and
makes extensions more self-evident; we have provided
an example of possible extensions to the family of
halogenated hydrocarbon refrigerants and to the family
of n-alkanols.
2. We believe that it is important pedagogically to
emphasize the quantitative (analytical) implementation
of the PLKT strategy rather than the use of tables and
charts, although the latter are useful qualitatively. This
quantitative implementation is easily carried out using
meta-computing software.
3. We have shown an implementation for a three-
parameter corresponding states prediction of the
compressibility factor to calculate z(P,T) when the EOS
is given in the form z( v ,T) using meta-computing
software. We have illustrated this with an example
using MAPLE;[`5 files for Mathematica,116' Mathcad,1"7'
and MATLAB1I8 can be obtained from the web site at Also
contained on this web site are four corresponding files
for calculating z(P,T) when the EOS is given in the
form z(vr,Tr).

Financial assistance has been received from the Natural
Sciences and Engineering Research Council of Canada. We
are grateful for discussions with A.S. Teja.

1. Platzer, B., and G. Maurer, Fluid Phase Equilib., 51, 223
2. Platzer, B., and G. Maurer, Fluid Phase Equilib., 84, 79
3. Lee, B.I., and M.G. Kesler, AIChE J., 21, 510 (1975)
4. Pitzer, K.S., J. Am. Chem. Soc., 77, 3427 (1955)
5. Pitzer, K.S., D.Z. Lippmann, R.F. Curl, Jr., C.M. Huggins,
and D.E. Peterson, J. Am. Chem. Soc., 77, 3433 (1955)
6. Lewis, G.N., and M. Randall, Thermodynamics, 2nd ed.,
revised by K.S. Pitzer and L. Brewer, McGraw-Hill, New
York, NY (1961)
7. Pitzer, K.S., Thermodynamics, 3rd ed., McGraw-Hill, New
York, NY (1995)
8. Smith, J.M., H.C. Van Ness, and M.M. Abbott, Introduction
to Chemical Engineering Thermodynamics, 5th ed., McGraw-
Hill, New York, NY (1996)
9. Teja, A.S., AIChE J., 26, 337 (1980)
10. Teja, A.S., and S.I. Sandler, AIChE J., 26, 341 (1980)
11. Teja, A.S., S.I. Sandler, and N.C. Patel, Chem. Eng. J., 21,
21 (1981)
12. Wong, D.H.S., S.I. Sandler, and A.S. Teja, Fluid Phase
Equilib., 14, 79 (1983)
13. Wong, D.H.S., S.I. Sandler, and A.S. Teja, Ind. Eng. Chem.
Fundam., 23, 38 (1984)
14. Edgar, T.F., Chem. Eng. Progr., 96(1), 51 (2000)
15. MAPLE is a registered trademark of Waterloo Maple, Inc.
16. Mathematica is a registered trademark of Wolfram Re-
Winter 2001

Use CEE's reasonable rates to advertise.
Minimum rate, 1/8 page, $100;
Each additional column inch or portion thereof, $40.

Ben-Gurion University of the Negev. Israel
The Department of Chemical Engineering of Ben-Gurion Univer-
sity of the Negev invites applications for tenure-track faculty positions
to start September, 2001. Duties consist of teaching at the undergradu-
ate and graduate levels and conducting innovative research. Applicants
should have a PhD degree in engineering or sciences (a degree in
chemical engineering will be considered an advantage) and should
have demonstrated his/her potential for excellence in teaching and
The Department of Chemical Engineering has four tracks of special-
ization at the undergraduate level:
Advanced materials (microelectronics, catalysis, and polymers)
Biochemical engineering
Chemical process engineering
Computer applications in chemical engineering
It is expected that the selected candidate will contribute to teaching
in some of the specialization tracks and will interact with faculty
Applications should be addressed to:
Prof. Jose C. Merchuk, Head
Department of Chemical Engineering
Ben-Gurion University of the Negav
PO Box 653
Beer Sheva 8415, Israel

search Inc.
17. Mathcad is a registered trademark of MathSoft, Inc.
18. MATLAB is a registered trademark of The Mathworks, Inc.
19. Sandler, S.I., Chem. Eng. Ed., 31, 18 (1997)
20. Sandler, S.I., Chemical and Engineering Thermodynamics,
3rd ed., John Wiley & Sons, Inc., New York, NY, (1999)
21. Curl, Jr., R.F., and K.S. Pitzer, Ind. Eng. Chem., 50, 265
22. Benedict, M., G.B. Webb, and L.C. Rubin, J. Chem. Physics,
8, 334 (1940); 10, 747 (1942)
23. Teja, A.S., and P. Rice, Ind. Eng. Chem. Fundam., 20, 77
24. Teja, A.S., and P. Rice, Chem. Eng. Sci., 36, 7 (1981)
25. Willman, B., and A.S. Teja, Chem. Eng. J., 37, 65 (1988)
26. Willman, B., and A.S. Teja, Chem. Eng. J., 37, 71 (1988)
27. Rice, P., and A.S. Teja, J. Colloid Int. Sci., 86, 158 (1982)
28. Teja, A.S., and P. Rice, Chem. Eng. Sci., 37, 788 (1982)
29. Teja, A.S., private communication, July (2000)
30. Wu, G.Z.A., and L.I. Stiel, AIChE J., 31, 1632 (1985)
31. Rowley, R.L., Statistical Mechanics for Thermophysical Prop-
erty Calculations, PTR Prentice Hall, Englewood Cliffs, NJ
32. Gallagher, J., M. McLinden. G. Morrison, and M. Huber,
NIST Thermodynamic Properties of Refrigerants and Re-
frigerant Mixtures Database (REFPROP), NIST Std. Ref.
Database 23, Version 4.0 (NIST, Boulder, 1993)
33. Yaws, C.L., Chemical Properties Handbook, McGraw-Hill,
New York, NY (1999)
35. Teja, A.S., and N.C. Patel, Chem. Eng. Comm., 13, 39 (1981)
36. Patel, N.C., and A.S. Teja, Chem. Eng. Sci., 37, 463 (1982)

MW laboratory





University of California, San Diego La Jolla, CA 92093-0411

he Chemical Engineering Program at the University
of California, San Diego, has successfully operated a
unique Chemical Engineering Process Laboratory
since the early 1980s'"l that is taught at the senior level over
two quarters for a class size under forty. The emphasis of the
laboratory is to develop skills in planning, designing, and
building an experimental apparatus, performance of experi-
mental work, analysis of data, and making proper interpreta-
tion and decisions. Learning how to make effective oral
technical presentations and developing report-writing skills
are also integral parts of the course.
The projects of this course range from the more traditional
areas of kinetics and transport phenomena to applications in
microelectronics, environmental engineering, and biotech-
nology. The uniqueness of this process laboratory class is its
attempt to emulate industrial process development projects
with one in-depth project rather than rotating through a set
of unit operation equipment each quarter for each small
group of students.

Although the philosophy of the class has remained the
same since its inception,"' some changes have been made in
the details of its execution. For each ten-week project, a
proposal and a safety memorandum must be written. In the
project proposal, the proposed plans and methods form the
main body of the proposal and the background section must
have a critical review of at least five to six pertinent research
articles. A proposal should clearly identify the objectives of
the project, demonstrate an understanding of the relevant

literature, establish the significance of the proposed work,
and outline the approach.
The safety memorandum should be a summary of perti-
nent information from the MSDS of any chemicals that may
be used in the specific project, should contain specific safety
precautions and lab procedures, and should address the issue
of waste minimization. With a better appreciation of the
importance of planning, the students are required for the
second quarter to prepare and use a project schedule (a so-
called Gantt) chart. During each quarter, students must dem-
onstrate experimental design on their new project and scale-
up calculations as specified by their project director.
We have recently introduced a group rotation project (e.g.,
a cooling-tower or heat-exchanger project) that is ongoing
each quarter and that involves the whole class. The objective
of the project is to learn group-to-group communication to
achieve a common goal by having each group work on the
project for one week. The students analyze their data and
then write a memo that summarizes their progress and ad-
vises the next group of the new tasks to be accomplished. To
communicate the success of the laboratory projects, the final

Jan B. Talbot is Professor of Chemical Engi-
neering and Materials Science at the University
of California, San Diego. She received her BS
and MS in Chemical Engineering from Penn-
sylvania State University. She worked as a de-
velopment engineer at the Oak Ridge National
Laboratory for six years before returning to
academia and receiving her PhD from the Uni-
versity of Minnesota in 1986 Her current re-
search interests are in the areas of information
display technology and electrochemical engi-
neering, particularly electrodeposition.

Copyright ChE Division of ASEE 2001

Chemical Engineering Education

This laboratory course has
been an excellent vehicle for teaching and
demonstrating electrochemical engineering principles.

presentation for the second quarter is a poster session meet-
ing presented to the local professional AIChE Chapter and a
Web page for their project is designed (http://

This laboratory course has been an excellent vehicle for
teaching and demonstrating electrochemical engineering prin-
ciples. The applications of electrochemistry are diverse and
the projects in the laboratory have spanned the gamut from
energy conversion (fuel cell) to electroplating to environ-
mental engineering. The electrochemical engineering projects
typically involve thermodynamics, kinetics, transport pro-
cesses (diffusion, convection, and electromigration), reactor
design, and scale up.
The students readily determined, after reviewing the elec-
trochemistry from their physical chemistry class, that elec-
trochemical processes can be analyzed using chemical engi-
neering principles. The main difference is the addition of the
effects of a potential gradient. But for the experiments and
analysis of data, the ability to control and measure voltage or
current can often be an advantage as it gives direct informa-
tion about the process. The minimum voltage may be related
to AG of the reaction. The current density may be propor-
tional to the rate of heterogeneous reaction. The limiting
current density can be related to the mass transfer rate. The
design of an electrochemical reactor is often analogous to a
heterogeneous reactor in which the surface area to volume
needs to be maximized. Attention must be given to potential
and current distributions, however.
Table 1 (next page) lists the project statements that have
been assigned in the class over the past decade. Note that the
statements are all the students receive, which emphasizes the
open-ended problem at hand.
The main piece of necessary laboratory equipment is a
potentiostat to control voltage and measure current (average
cost is about $12,000). Also, reference electrodes are used
in most experiments. The most expensive materials were
ion exchange-membranes or dimensionally stable elec-
trodes, which were used in as small amounts as possible
when required.
A generic electrochemical reactor that could be used re-
peatedly was a project one year and it has been used by other

groups for many experiments. Note that four of the projects
(6 through 9) are of an environmental engineering nature,
which is of great interest to the students. There is the added
benefit to the professor in actually observing how these
various processes work.
It is also possible to incorporate the experiments given in
Table 1 into a more traditional unit operations laboratory
with a focus on kinetics or mass transport aspects. The
experimental stations, such as the reciprocating-paddle sys-
tem to be discussed, were built in our shop with inexpensive
materials such as plexiglass and used simple pumps and
motors (at a total cost of less than $200).

To illustrate the development of one of the lab projects, a
study of copper electrodeposition with a reciprocating-paddle
system will be used as an example to show how the prin-
ciples of electrochemical engineering were explored by the
students in a design context. It was one of the most success-
ful projects; this very motivated and industrious group wrote
a journal note121 from the results of their project.
The group of students was given the problem statement
(#3, Table 1), and they proceeded to design and build
(with help from a technician) their experimental appara-
tus based on information from a patent'3 in which a
reciprocating paddle cell had been used to deposit mag-
netic alloy films. A schematic of their experimental ap-
paratus is shown in Figure 1.
The plating cell was a plexiglass tank to be filled with

Figure 1. Reciprocating paddle cell
experimental apparatus.

Winter 2001

Motor and
paddle tracking gear

Project and Scale-Up Design Problem Statements

Project Statement

1) Methanol-air fuel cell
Fuel cells can be useful power sources (particularly for operation in remote areas)
due to their high energy density and quiet operation. However, in a methanol-air
fuel cell, methanol is soluble in the aqueous electrolyte and can diffuse to the
cathode where it can oxidize. Our client is interested in an application of a
methanol-air fuel cell, but is aware that there are problems. The problem of
diffusion of methanol to the cathode is an important problem to our client that
needs to be investigated. Please develop an experimental plan and design an
appropriate apparatus to conduct your study.

2) Electrolytic gas evolution
In conventional bioreactors, oxygen is provided to the cells by bubbling air
through the growth medium. This method has several disadvantages, however, in-
cluding nonuniform distribution of oxygen, destruction of fragile cells from induced
shear, and risk of contamination. In past studies of mass transfer of oxygen in both
air-lift and bubble-column reactors, our development engineers have determined that
achievement of an adequate gas-liquid mass transfer coefficient was not possible.
However, a novel electrochemical aeration system has been suggested to enhance
mass transfer by evolving oxygen electrolytically as finely dispersed bubbles. Please
develop an experimental plan and design a system to conduct your investigation.

3) Reciprocating paddle electroplating cell
3a) An electroplating process is being used by our client to produce thin alloy films
for magnetic recording heads. Our client would like us to investigate the use of a re-
ciprocating paddle as a means of electrolyte agitation. Please develop an experi-
mental plan and design an appropriate apparatus to conduct your investigation.
3b) Our client was satisfied with the investigation of the use of a reciprocating pad-
dle as a means of electrolyte agitation for copper deposition which was initiated
during the last quarter. However, we have been informed that organic agents are
routinely added to the plating baths to enhance the quality of the electrodeposition.
We need to understand if these additives affect the mass transfer and the kinetics of
deposition. Please develop an experimental plan to conduct your investigation.

4) Electroplating of through-holes
Plated through-holes provide electrical connections between the different layers of
printed circuit boards. However, the trend in the development of multilayer printed
circuit boards is towards more layers and narrower diameter through-holes. There-
fore, as through-hole aspect ratio (length-to-diameter) increases, uniform electro-
deposition of a through-hole becomes very challenging. We wish to study the
electrodeposition of copper in a through-hole in order to characterize the impor-
tant process variables that affect plating uniformity. Please develop an experi-
mental plan and design an appropriate apparatus to conduct your investigation.

5) Electroforming process
An electroforming process is being used by our client in making orifice plates for
their thermal ink jet printer. Our client wants a study of an electroplating process
to better understand the effects of various important system parameters, such as
anode-cathode orientation, applied potential drop, and electrolyte concentration.
Develop an experimental plan and design an appropriate apparatus to conduct
your investigation.

Scale-Up Design Problem

la) It has been proposed to use a methanol fuel cell system remotely in small cabins
for weekend recreational power usage. Please design a fuel-cell system for this
purpose. Discuss the power requirements and costs.
Ib) Our client was very pleased with the design of the batch apparatus for the
methanol-air fuel cell study. However, for continuous use, our client would like
the fuel cell to be designed as a continuously stirred tank reactor with recycle of
the electrolyte.
Ic) Our client would like a design of a cross-flow array assembly of methanol-air fuel
cells to supply back-up power to a personal computer for eight hours. Base this de-
sign on the results of your experimental study of a methanol-air fuel cell.

Our client would like a design of a 10-liter aerobic bioreactor to produce penicillin
using electrolytically evolved oxygen and air sparging in a bubble column. Compare
the costs associated with using electrolysis and compressed oxygen.

3a) Our client would like a design of a copper-plating system using a reciprocating-
paddle means of electrolyte agitation to plate multilayer circuit board
3b) Our client would like a design of a system for uniform electroplating of a 1-p.m
film of copper on a Im x Im device. Please design a commercial-scale electro-
plating process that uses the reciprocating paddle for agitation. Also include a pro-
cedure for operating the electroplating system that you design.

Our client would like a design of a copper-plating system to plate the through-holes of
a multilayer circuit board. The 0.381-cm (150 mils) thick circuit board has an array of
100 through-holes, each with a radius of 0.064 cm (25 mils) with a hole-to-hole distance
of 0.254 cm. The minimum thickness of 0.0025 cm (1 mil) of copper is required at the
center of each through-hole. In your design, consider agitation of the electrolyte solu-
tion by pumping the solution. Estimate the cost of power required for electroplating
and pumping of the solution.

Our client was very impressed with the design of the parallel-plate electroplating
process considering the effects of the primary current distribution. However, our
client would like to consider a system for uniform electroplating with flow through
parallel plate electrodes. Please design a commercial scale electroplating process for
a flowing electrolyte.

Continued on next page

Chemical Engineering Education

about 10 liters of the plating solution. A 1-cm2 copper cath-
ode was mounted in the center of a platform raised above the
tank floor. The raised platform was added to allow fluid
displaced by the paddle to flow underneath it to minimize
waves induced from the tank walls. Flow visualization was
performed by the students with this system.
A simple rectangular block fixed on a tracking system was
used as the paddle driven by a motor in a reciprocating
motion at various velocities and heights above the cathode
surface. A copper anode was placed parallel above the cath-
ode support. The anode area was about 250 times that of the
cathode, to allow it to function as a relatively unpolarized
reference electrode as well as a counter electrode. The solu-
tion was either a binary electrolyte 0.05M CuSO4 or a
supporting electrolyte solution of 0.05M CuSO and 1.7M
H2SO4. One of the students mentioned that this project
kept him in style with "acid-washed" jeans, which were
very popular at that time.

The quality of the 1-pm thick film was determined by
using microscopy and profilometry. Applied potential
and current were controlled and measured, respectively,
by a potentiostat, connected by copper-wire leads to the
Before building the apparatus, the first step for this
project was to learn about the thermodynamics, kinetic,
and mass transport of electrochemical systems. Often, a
comfortable place for a student (or a faculty member) to
begin is the chapter on electrochemistry in a physical
chemistry textbook.
There are also several good basic books on electrochemis-
try and electrochemical engineering (see Table 2) appropri-
ate for the novice. Extra time has not been needed to
lecture to the students on electrochemistry; this informa-
tion is garnered through specific technical problems the
students encountered either in reading literature or in
conducting an experiment.

Table 1, continued.

6) Packed-bed electrode system
A local manufacturer of printed ciruit boards requires a waste treatment system to
recover plating wastes generated in their copper electroplating process. The man-
ufacturer is interested in a high-surface area rotating electrode cell metal recovery
system. Please develop an experimental plan and design an appropriate apparatus
to conduct your investigation.

6a) A typical average discharge rate from a plating shop is 50,000 gal/day. Based on
your bench-scale packed-bed electrode system, design a waste-treatment system
to accommodate this flowrate that meets the EPA discharge standards.
6b) ELTECH Systems Corporation (Sugarland, TX) sells a heavy-metal recovery
electrochemical cell with an extended surface area provided by use of a "reticu-
lated" metal sponge cathode. The cathode presents an actual surface area almost
15 times its geometrical area. The cathodes are used in a cell as porous flow-
through electrodes that are placed in series with flow-through anodes. The
typical cell operating conditions for a copper sulfate stream are a feedrate of
3 gal/min with an inlet concentration of 250 ppm and an outlet concentration of
5 ppm. The average rate of removal is 0.36 Ib/hr. The process conditions are a cell
current of 200 amps and a voltage between 1.5 and 5.5 V. Please design a packed-
bed cathode cell (or a series of cells) that can compete with the ELTECH system.
Discuss the power requirements.

7) Electrokinetic soil remediation
Our company has been asked to evaluate an electrokinetic processing system for Geokinetics, Inc.. claims to have used the electrokinetic soil remediation to reduce
the treatment of pollutants in sand. Your group needs to determine the appropriate the copper concentration of 8100 ft' of soil from a former paint-factory waste facility
technology, a chemical system to study, and a plan of development for a sand- from 1200 ppm to <200 ppm. Please design an electrokinetic soil remediation system
treatment system, for this purpose, considering there is 10% clay by weight in the soil. Discuss the
power requirements and costs.

8) Electrochemical oxidation of organic
Our company has been asked to evaluate using an electrochemical oxidation system
for the treatment of organic constituents in a wastewater stream. A recent
article is attached that briefly reviews both indirect and direct oxidation pro-
cesses. Your group needs to determine the appropriate technology, a chemical
system to study, and a plan of development for a wastewater treatment system.

9) Electrochemical ion exchange
Our company has been asked to evaluate an electrochemical ion exchange
system for the treatment of toxic metal ions in a wastewater stream. Your group
needs to determine the appropriate technology, a chemical system to study, and a
plan of development for a wastewater treatment system.

8a) Based on your bench-scale experimental results and information from the litera-
ture, design a scaled-up system to treat phenol at 200 ppm from 1000 gal/day of
wastewater to U.S. effluent water standards. Please consult with the other two
groups (Fenton's Regent Group, UV Oxidation Group) in order for us to make a
"fair" comparison of these technologies.
8b) Design a cell (or a series of cells) with the ELTECH-type system (see 6b) with an
inlet concentration of 150 ppm of phenol. Discuss the power requirements and

Design an EIX cell (or a series of cells) that can compete with the ELTECH system
(see 6b). Discuss the power requirements. Also, consider elution of the resin to
recycle make-up stream to a plating operation that uses a concentration of
200 g/1 CuSO4.

Winter 2001

The important reactions that the students must con-
sider are the electroplating of copper and hydrogen evo-
lution at the cathode and the corrosion of copper at the
anode. An explanation of the relationship between free
energy change of reaction (AG) to electrochemical po-
tential through the Nernst equation and then use of a
Pourbaix diagram[41 (potential vs. pH) give a direct use
of thermodynamics. The students readily determined
and demonstrated the minimum potential required for
electroplating of copper (and also in order to avoid
hydrogen evolution).
By connecting the electrodes in a solution in a beaker
(which is typically done first to understand the basics of
anodic and cathodic reactions) and then in the actual
plating apparatus, the student stepped (or ramped) the
potential difference, E, and measured current, i (to attain
a polarization curve).
For this experiment, the applied potential was de-
creased in steps beyond the mass transfer limiting cur-
rent plateau until a sharp increase in current was ob-
served, indicating the dominance of H2 evolution, as
shown in Figure 2. The current efficiency of copper
electrodeposition was determined by measuring the
weight gain of a sample at limiting current over a speci-
fied period of time, using Faraday's law. Therefore, the
kinetics of copper plating and at very cathodic poten-
tials, hydrogen evolution, were illustrated.
Due to the reciprocating motion of the paddle, the
current transients at a set applied potential were peri-
odic. Therefore, an average of the maximum and
minimum current values at each applied potential
difference associated with the reciprocating paddle

List of Recommended References

For the Novice
Electrochemical Engineering Principles, G. Prentice, Prentice-Hall (1991)
Industrial Electrochemistry, 2nd ed., D. Pletcher, F.C. Walsh; Chapman &
Hall (1990)
Electrochemistry: Principles, Methods, and Applications, C.M.A. Brett,
A.M. Oliveira-Brett; Oxford Science (1993)

For the More Advanced
Electrochemical Science, J. O'M. Bockris, D.M. Drazic; Taylor & Francis
Modem Electrochemistry, Vols I & 2, J. O'M. Bockris, A.K.N. Reddy;
Plenum Press (1977)
Surface Electrochemistry, J.O'M. Bockris, S.U.M. Khan; Plenum Press
Electrochemical Methods: Fundamentals and Applications, A.J. Bard, L.F.
Faulkner; John Wiley & Sons (1980)
Electrochemical Systems, 2nd ed., J.S. Newman; Prentice-Hall (1991)
Electrode Processes and Electrochemical Engineering, F. Hine; Plenum
Press (1985)
Electrochemical Reactor Design, D.J. Pickett; Elsevier (1979)
Environmental Electrochemistry, K. Rajeshwar, J. Ibanez; Academic Press

motion was used in Figure 2.
The mass transfer limiting current plateau for copper electrodepo-
sition is easily identifiable in Figure 2. The limiting current densi-
ties for the binary electrolyte were higher than the supporting
electrolyte, as expected-indicating the influence of electrical mi-
gration on the transport of copper ions for the binary electrolytic
solution. The effect of forced convection is also readily observable

1 _________________

Figure 3. Mass transfer correlation.
Chemical Engineering Education

10 -

1.0 -

1,000 10,000

Figure 2. Polarization curve from copper deposition in
0.5 M CuSO, and 1.7M H2SO, in a reciprocating paddle

Figure 4. Topography of the copper deposit from a binary
electrolyte at a high velocity with a paddle height of
1 cm above the cathode.

by the increase in the limiting current density. Since the flux
of an ionic species is related to the current density, the
average mass transfer coefficient, k (cm/s), can be deter-
mined from the limiting current density, i* (A/cm2)

k= i* (1)
where n is the number of electrons transferred, F is Faraday's
constant (9.6485 x 104 C/equiv), e is the current efficiency,
and Cb is the bulk reactant concentration (mol/cm3). For our
system, we anticipated that the characteristic parameters
were the linear paddle velocity (U) and paddle height above
the cathode surface (L), to be related by
Sh = aReb Scc (2)

where Sh = kL/D is the Sherwood number, Re = UL/ v is the
Reynolds number, Sc = v / D is the Schmidt number, and v
is kinematic viscosity of the fluid. The power of the Schmidt
number, c, was estimated as 1/3 based on correlations in the
literature. A log-log plot of Sh/Sc"1/ versus Re for all the
experiments in both supporting and binary electrolytes is
given in Figure 3. The coefficients of the correlation in Eq.
(2) were determined to be a = 0.039 and b = 0.80. This
correlation was then used for the scale-up design problem.
One of the difficulties in electrochemical engineering is
understanding the potential and current distributions inher-
ent in such systems. To explore these concepts, the students
measured the topography of the plated copper layer by a
stylus profilometer, as shown in Figure 4. The thickness of
the films was observed to be greatest at the cathode edges, as
shown in the topographical maps in Figure 4, and decreased
as the paddle was raised up to about 1 cm.121 Comparison of
deposition uniformity between a free convective and a
highly agitated plating bath, as determined by the topo-

Winter 2001

To optimize deposition rate and uniformity, the highest
paddle velocity (43 cm/s) and a paddle height of 1.0 cm
above the cathode was recommended. Comparison of the
deposition surfaces with and without agitation showed no
loss of uniformity with agitation, while increasing the depo-
sition rate by at least five times.
The other experiments listed in Table 1 followed a similar
procedure, with varying degrees of success. The principles
of chemical engineering and design were readily demon-
strated in each of the experiments. A few students became
interested in the specialization of electrochemical engineer-
ing and later they even took my graduate course on electro-
chemical engineering.

The author acknowledges the contributions of colleagues,
Professors Pao Chau and Richard Herz, who have also taught
and improved this course over the years. The laboratory has
been skillfully managed by Victor Gruol.

1. Rochefort, S., S. Middleman, and P.C. Chau, "An Innovative
ChE Process Laboratory," Chem. Eng. Ed., 19(3), 150 (1985)
2. Rice, D.A., D. Sundstrom, M.F. McEachern, L.A. Klumb,
and J.B. Talbot, "Copper Electrodeposition Studies with a
Reciprocating Paddle," J. ofElectrochem. Soc., 135(11), 2777
3. Powers, J.V., and L.T. Romankiw, U.S. Patent 3,652,442
4. Pourbaix, M., Atlas of Electrochemical Equilibria in Aque-
ous Solutions, Pergamon Press, New York, NY (1966) O



logical survey of the copper deposit, showed
that the paddle motion seemed to improve the
deposition uniformity at the cathode edges par-
allel to the paddle motion.
The effect of thiourea on copper deposition was
also studied. Thiourea (SC(NH,),) is often added
to plating solutions to refine the grain size and
brighten the deposit. The addition of thiourea to
the plating solutions did not influence the limit-
ing current density for copper deposition, but
did affect the surface kinetics and properties of
the deposit.

During the two quarters of the design lab (10
weeks each), two groups of chemical engineering
seniors designed, built, and operated a reciprocat-
ing-paddle electrochemical system. A correlation
was determined to describe the mass transfer lim-
ed operation of an electroplating bath using this device in
whichh binary and supporting electrolyte solutions exhibited
ie same power law correlation. This correlation would al-
)w scaling the system for industrial applications.

E" classroom



In ChE Undergraduate Courses

Indian Institute of Technology Bombay, Powai, Mumbai 400 076 India

T teaching methods involving only a lecture (or its vari- exercise philo
ants) and a problem set format have several short- with limited d
comings.1ll For example, they neither promote the ing (to non-cl
creativity1-3" nor develop the independent thought pro- Principles, ov
cesses that are desirable for a students' future endeavors CFA exercise
in the "real world." cises that this
One of the strategies that has been widely discussed to ASSIGNME
partially offset the constraints of the lecture-problem set
teaching method, is the use of open-ended problems. Such The follow
problems develop the important skill of divergent produc- of the course
tion121 in students. Most open-ended problems, however, are students on th
limited by the instructor; for example, the creativity aspect is Problem Ana,
limited to finding various solutions to a particular, instruc- to industry or
tor-assigned (and thereby, instructor-limited) problem. rial and ener
fJbr tehe therm
There are usually several students who are inherently more endedproblei
creative than the experience-honed instructor, and such an and analysis
exercise does not fulfill the academic passions of those stu- Origi
dents. Also, for many students the concept of rational choice, Focus
which is crucial to success in the real world of industry or Quali
research, is not well developed. For example, based on ques- Origi
tions posed to first-year students taking material and energy Prese,
balances, we found that only 4 to 5 students in an 80'- A concise rep
student class are clear as to why they chose chemical engi- cate your wor
be evaluated
neering as their major. This probably happens because most if the problem
students come from well-protected family environments into
a reasonably well-protected campus environment for their
undergraduate studies, and they accept established hierar-
chies or trends without considering their own individual
strengths or affinities.
To address these issues, a Choose/Focus/Analyze (CFA)
exercise was conceived and given to students taking the
first-year Material and Energy Balances course and the sec-
ond-year Engineering Thermodynamics course for the past
four years. A similar exercise (with minor variations in
Copyright ChE Division of ASEE 2001

sophy) was also given to students taking courses
lepth, such as Elements of Chemical Engineer-
hemical engineering students) and Bioprocess
'er the same period. This article discusses the
along with summaries of some student exer-
instructor found interesting.

ng assignment was made during the discussion
-information material that was handed out to
e first day of classes:
lysis: Students have to choose a problem of relevance
Sany human endeavor and analyze it using the mate-
gy balance principles (or thennodynanics principles,
dynamics course) learned in class. This is an open-
n that has been designed to improve the choice, focus,
skills in students. The evaluation will be based on
nality in approach 15%
level 15%
of analysis 20%
ty of work 20%
zal contribution 20%
station (mainly communication) 10%
ort (in the format that you think would best communi-
k) submitted a week before the last day of classes will
strictly based on the criteria given above. It will help
n is chosen well in advance (within the first four

G.K. Sureshkumar is an Associate Professor
in the Chemical Engineering Department and
an Associate faculty member in the Biotech-
nology Center at the Indian Institute of Tech-
nology, Bombay. He received his BTech from
the Indian Institute of Technology, Madras,
and his PhD from Drexel University. His cur-
rent research interests are mainly in free radi-
S cal mediated optimal operation of bioreactors.
Professor Sureshkumar can be contacted at

Chemical Engineering Education

weeks) and sufficient time, distributed throughout the course's
duration, is devoted to it.
Each student had to perform the exercise
individually, and it carried either a 15% or a
20% weightage toward the final grade. ThiS
To make students self-reliant, the instruc- diSCi
tor offered help only when
The student decided to visit an
industry. An introductory letter was Chl
provided in these cases, but the Fo
students were clearly informed that a Al
letter alone would not guarantee their
admission to the industry. The fact exerTci
that more than 75 students over the
past four years have managed to visit
an industry for this exercise indicates SUHa Ut
their relevant abilities, either native SOm e.
or developed for this exercise. eXesit
The student wanted to know if the
problem was "too small" or "too thiS Il
large." The instructor would give his foi
opinion on that aspect alone. .i
The instructor was readily available, how-
ever, to clarify the other aspects of the course,
such as class material, concepts, and prob-
lem sets.
Students who thought of novel aspects to analyze were
awarded high marks under the "originality in approach"
heading, whereas the students who made good contribu-
tions, irrespective of whether the aspect was novel or not,
were awarded high marks under the "original contribution"
heading. Also, students who focused clearly on their task
received high marks under the "focus level" heading, and
students whose analysis had good depth scored high under
the "depth of analysis" heading.

Summaries of selected student exercises are presented in
the following paragraphs. The report titles were those given
by the students, with the student's name and the course title
denoted below.

Fighting Alcoholism: A Chemical Aspect!
(Gaurav Tayal) (Material and Energy Balances)

This report demonstrated the use of material and energy
balances to analyze a situation that is socially relevant. In his
statement of motivation, Gaurav Tayal noted, "I have quite a
few friends in my hostel [dorm] who can be classified as
casual drinkers. What inspired me to undertake this exercise
was the withdrawal symptoms that my friends suffer the
next morning after drinking." He also conducted a short
survey that he did among his friends in the hostel who drink,
Winter 2001



V .



which showed that almost all the participants wanted control
over their state the morning after. He clearly identified (fo-
cused on) the following aims for this exercise:
"What should be the maximum rate of
icle intake of an alcoholic beverage of a given
IRS alcohol (C,HSOH) concentration so that it
does not cause intoxication (hangovers occur
only when intoxication sets in; otherwise the
e/ alcohol is easily metabolized by the body)?"
J "What should the relationship be
between the rate of alcohol intake, the time
R_ period of drinking, the strength of the liquor,
*Ifuf and the time taken by the body to revive?"
To achieve the above aims, Gaurav represented
the stomach, small intestine, and blood as con-
IS Of trol volumes, made suitable assumptions, con-
Senlt suited several books (including encyclopedias),
performed material balances on ethanol, wa-
_-ta ter, and total mass, and concluded
H CtOr "To avoid intoxication, a normal
d person should drink beer at the rate of less
than 790 ml h'."
ing. "To be in a position to drive home
safely after a party, a normal person should
drink less than 90 ml of whiskey over a two-
hour period. "
While the actual numbers above may be subject to debate,
the beauty in application of the material balance principles
and the social relevance is clear.

Methyl Isocyanate Poisoning from
Union Carbide Factory at Bhopal
(Sagnik Basuray) (Material and Energy Balances)
The motivation for this student was to see whether or not
the biggest chemical factory disaster in India could have
been avoided, using concepts that he had learned in his first
chemical engineering course. It is well known that there
were at least five levels of safety measures at the factory,
including a flame tower to burn vented gases, all of which
either failed or were not operational when the gas leak oc-
curred. Sagnik performed material balance calculations on
the relevant sections of the plant, which included tank No.
610 that leaked, and calculated the composition of the vari-
ous streams for different inputs. Different inputs were con-
sidered to determine which inputs would have still averted
the disaster. He also carried out energy balance calculations
to conclude that even if the flame tower had been operational
for burning away the released gas, it would have collapsed
because it was structurally incapable of handling such a high
rate of energy inflow as 1.05 MJ min-'; therefore it was
designed badly. Sagnik used the known civil engineering
data on the tower to draw this conclusion.

From an instructor's viewpoint, a more refined analysis is
needed before drawing such a strong conclusion, but the
rudiments of good application are evident.

Thermodynamics of Breathing
(Narendra Dixit) (Engineering Thermodynamics)

Narendra Dixit wrote, "the process of breathing has al-
ways been a marvel to human intellect. It is something that
goes on and on, sustaining life under the most critical of
circumstances, stopping only when there is no more life.... It
is an enlightening exercise to identify how much energy one
spends (or consumes) to sustain one's own life."
His well-identified objectives were
To formulate a thermodynamic model of the respira-
tory apparatus
To analyze the thermodynamics involved in the
mechanism of breathing
After referring to several medical and other books, he
divided the process of breathing into two segments: a nasal
process involving processes at the nose, and a post-nasal
process involving processes in the lungs. He modeled the
post-nasal breathing process as a cylinder with a frictionless
piston with three openings for air, CO2, and 02. With suit-
able other assumptions, he determined that the total work
done during each breathing cycle is a low 0.03 J, the entropy
change is only 3.42 x 10-4 J K' per cycle, and the irrevers-
ibility is 0.1026 J per cycle. He concluded by marvelling at
the superiority of natural mechanisms from a thermody-
namic viewpoint.

Analysis of a Spirit Lamp Using Material Balances
(Gaurav Misra) (Material and Energy Balances)
The motivation for Gaurav Misra was his fascination with
the simple spirit lamp. Based on material balance principles
and certain assumptions (the limitations of which he was
acutely aware of), he not only derived expressions to achieve
the following objectives, but also suggested simple experi-
ments to obtain numerical values of the relevant quantities.
His objectives included
Obtaining an expression for the rate offlow of atmo-
spheric air to the lamp
Obtaining an expression for the rate of rise offuel in
the wick
Obtaining an expression for the fuel efficiency of the
Relating the parameters of the wick with the efficiency
of the lamp
It is worthwhile to point out that the student had no expo-
sure to fluid mechanics at the time and had to acquire some
fluid mechanics principles from senior students and books to
be able to attempt analysis of the relevant parts of his project.

Thermodynamic Analysis of Glass-Fiber Production
(Manoj Kumar Pandey) (Engineering Thermodynamics)

Manoj decided to apply thermodynamic principles to the
glass-fiber production process. He focused on
Estimation of the power input required
The effect of varying power input on viscosity (a
narrow range in viscosities is required for good-
quality fiber)
Analysis of the cooling system and estimation of
temperatures of the fiber exiting the nozzle
He chose suitable control volumes for his analysis and
estimated a power of 15 KW to produce about 10 Kg h-1 of
fiber. He also found that the power input should be con-
trolled to within less than 1% variation for good quality
fiber and that the temperature of the fiber exiting the
nozzle is 175'C.

Generation of Electrical Power Using Automobile
(Prateek Jain) (Material and Energy Balances)

Prateek Jain considered a small dynamo at the end of a
tailpipe to find out whether automobile exhaust can be used
to generate electricity. He gathered background information
on engines that use compressed natural gas as fuel, exhaust
compositions, temperatures at exhaust, and other relevant
information. Then he performed material and energy bal-
ances on the engine and tailpipe and found that the velocity
of exhaust gases from the tailpipe could be used to drive a
dynamo that could light a small bulb.

There have been many exercises that this instructor found
interesting, although only a few are described above. The
more common exercises included industrial data consistency
checks using real data obtained from industry, material and
energy balances, and exergy/irreversibility analysis in engi-
neering thermodynamics.

The above examples are good exercises, and it is obvious
that the students thoroughly enjoyed the work. Over the past
four years, an estimated 65% of the students expressed ap-
preciation to the instructor for assigning this exercise be-
cause they felt they learned something useful. Two espe-
cially perceptive students thanked the instructor for not help-
ing them at all, realizing that no help was the best path to
learning. A few students who are currently graduate students
pursuing their doctoral degrees in the U.S. have said that the
CFA exercise is the only thing they still remember from the
course after four years.
Except for five students (over the past four years), the rest
of the students had no comments. Two of the five respond-

Chemical Engineering Education

ing students said that the exercise was not useful to them,
and the remaining three wanted the instructor to assign project
titles. Among these five responses, three were received the
first time the exercise was assigned and the other two were
received the second time it was assigned. No adverse com-
ments have been received since then. This indicates that the
instructor perhaps became better at making the students ap-
preciate the purpose of the exercise through mentioning it
more often at appropriate junctures during the semester.
The class average in this exercise is usually around 65%,
save for the first time it was given when it was 54.1%. This
is probably because in subsequent years, copies of good
reports from the previous years were made available as
reference material.

The CFA exercise, in the form mentioned above, may not
be suitable for courses with limited depth since they cover so
many different principles in just a superficial manner. For
such courses, the following exercise was assigned:
This exercise expects students to "adopt" a chemical (bio-
chemical) industry by the third week of classes. Then, stu-
dents should relate the principles taught in class to the
actual processes taking place in the "adopted" industry and
analyze, preferably, one aspect in detail. A concise report
submitted a week before the last day of classes will be
evaluated strictly on the following aspects:
Link between fundamentals and actual processes 40%
Analysis of the actual processes) 30%
"Reality "factor 20%
Presentation (mainly communication) 10%
If students visited the industry, they received the 20% "real-
ity" factor. If they decided to perform a library exercise, then
the closeness of their report to actuality formed the basis for
marks on that aspect.

Although the instructor did not face any real difficulty
with the evaluation except for the time needed to grade
reports for large classes (five to six full days), some col-
leagues raised questions about certain evaluation aspects. It
is worthwhile mentioning some of those apprehensions:
How can undergraduate students do independent work?
The student exercises show that even first-year undergradu-
ate students are capable of visiting an industry, indepen-
dently, and choosing novel aspects for analysis when en-
couraged to do so.
How can you find out if a problem was taken from some
unseen source? How do you judge originality? It is known
that as long as grades are important, some students will cheat
to get the highest possible grade.I41 However, the instructor
usually knows the level of students' knowledge (at least on
Winter 2001

the imparted subject) in a class taught by him, and therefore
he has a notional expectation. On that basis, if a report is
suspicious, the instructor can have a one-on-one interview
with the student. On the average, this instructor needed to
interview approximately 20 students out of a 80'-student
class. By asking the correct questions, most of which are
technical, it was easy to determine if the student had cheated.
If just the problem was picked from some source and the
analysis was done by the student, zeroes were given for
the relevant aspects in the evaluation (such as originality
in approach, focus level, and original contribution) so
out of a possible 100 points, the student received marks
in the thirties. This prevents cheating to a large extent in
subsequent years because of the way the word spreads
among the students.
The fact that no complaint/comments that someone got
away with submitting "lifted material" have been received
from an acutely grade-conscious set of undergraduates over
the past four years, coupled with verbal comments that the
instructor has been very fair, shows that it is possible to
effectively weed out the cheaters.

Course Fundamentals It is the instructor's opinion
that the students who scored above 50% (recall that the
class average is usually around 65%) had picked up the
course fundamentals to a desirable degree because they
were able to view an aspect of their choice from, say, a
material-and-energy-balances point of view.
Self-Reliance Since the instructor denied any help at
any stage of the project, students were forced to be-
come self-reliant. It was initially difficult not to be
nice to the students when they asked for help, but in
the greater interest of the students, this instructor be-
came accustomed to it.
Rational Choice The students were asked to make a
rational choice among the innumerable ways they could
have approached the project and to be responsible for
it. For example, if they did not obtain proper data from
an industry and therefore had to change their project
midway through it, they realized that they were com-
pletely responsible for choosing that particular indus-
try in the first place.
Creativity/Lateral Thinking A wide scope exists for
exhibiting one's creativity in such an exercise because
it invites the student to see all aspects from the point of
view of the fundamental principles. Even when the
student is not inherently creative, but desires to be
creative, a significant amount of time (about half the
semester) is provided to the student to apply himself
toward that goal. In the instructor's judgment, about
20% of the students were creative through effort. Also,

creativity could be manifest in many aspects of the
exercise, such as creative choosing, creative focusing,
creative analysis, and creative presentation. Students
who were creative were richly rewarded in the "origi-
nality in approach" and/or the "original contribu-
tion" categories of the evaluation. The exercise also
provided good scope to exhibit lateral thinking, as
demonstrated in some of the samples discussed in
the earlier section. Some students also showed their
"synthesis" abilities in their choice of the problem
for analysis.
- Focus The focus level in almost 90% of the reports
was acceptable, and in an estimated 60% of the reports
it was good.
I Communication and Professional Appearance of
Reports This exercise (deliberately) did not have a
pre-determined format for reports in order to help the
students think about how to present their work in an
effective fashion. When a certain format is provided, it
tends to curtail the freedom of organization, and in
many cases, organization different from the traditional
one is more effective in communication. The instructor
estimates that about 80% of the class communicated
their work reasonably well, and 50% of the class did it
well (one reading was sufficient for understanding)-a
fact that was surprising to the instructor in the first two
years. Also, an estimated 60% of the reports had a very
professional appearance. This indicates that when some-
thing is assigned as their responsibility, the students do
a good job in ways not expected of them.
Helping Others The students at I.I.T. Bombay are so
highly competitive that on many occasions their "cut-
throat" competition has saddened this instructor. In
this particular exercise, however, it was a pleasant
surprise to find classmates helping each other-be it in
discussing possible ideas, sharing instructive web-site
addresses, or teaching word-processing skills-while
at the same time protecting their novel ideas. In their
reports, many students acknowledged the help they
received from their friends.
Teamwork It is true that the CFA exercise does not
promote teamwork, but the importance and value of
teamwork is emphasized in tutorial sessions through-
out the course. During the sessions (on the average,
one hour per week, per course), the students are ex-
pected to work out problems given in the problem sets.
Normally, teaching assistants grade the performance
in the problem sets, which carries a 5% to 10%
weightage toward the final grade. In the courses taught
by this instructor, the class is divided into 10 to 15
groups of 5 to 6 students each. The problem sets are
distributed about a week prior to the tutorial session,
and the students are given complete freedom to discuss

the problems with anyone. The only requirement is
that they learn how to solve the problem. During the
tutorial session, one student from a group, chosen by
drawing lots, works out the problem on the board and
is graded by the instructor for correctness in approach
and answers to questions by the class/instructor (90%),
and communication to the class (10%). Whatever marks
the student earns are given to the entire group, thereby
making the student responsible for the groups marks-
or, in other words, the group is made responsible for
each member knowing the solution. Ten percent
weightage toward the final grade, coupled with the
ignominy before their classmates if they do not pre-
pare, are significant motivating factors for the majority
of students to take the tutorials seriously. Thus, the
importance of teamwork is emphasized.

From a broader perspective, Prausnitz[51 opined that chemi-
cal engineering is one of the humanities that has a deep
human intent and that the role of context should be inte-
grated into the chemical engineering curriculum rather than
being delegated to a course in humanities. Felder[21 has said
that if we are to produce engineers who can solve society's
most pressing problems, we must somehow convey that
problems in life are open-ended and convergent production
(generation of the right answer to a well-defined problem),
which is synonymous with academic excellence in engineer-
ing, is only one of the skills required. This instructor be-
lieves that the CFA exercise takes us a step closer toward
realizing their suggestions.

I thank the students of Material and Energy Balances
(Introduction to Chemical Engineering), Thermodynamics,
Elements of Chemical Engineering, and Bioprocess Prin-
ciples classes that I have taught over the past four years for
their enthusiastic participation in the CFA exercise. I also
thank my colleague, Professor Raghunathan Rengasamy, for
his suggestion to communicate this exercise as a paper to
Chemical Engineering Education. In addition, I thank my
colleagues, Professors Kartic Khilar and Hemant Nanavati
for their input.

1. Carlson, E.D., and A.P. Gast, "Animal Guts as Ideal Reac-
tors: An Open-Ended Project for a Course in Kinetics and
Reactor Design," Chem. Eng. Ed., 32, 24 (1998)
2. Felder, R.M., "Creativity in Engineering Education," Chem.
Eng. Ed., 22, 120 (1988)
3. Prausnitz, J.M., "Towards Encouraging Creativity in Stu-
dents," Chem. Eng. Ed., 19, 22 (1985)
4. Felder, R.M., "Cheating-An Ounce of Prevention,...Or the
Tragic Tale of the Dying Grandmother," Chem. Eng. Ed.,
19, 12 (1985)
5. Prausnitz, J.M., "Chemical Engineering and the Other Hu-
manities," Chem. Eng. Ed., 32, 14 (1998) 0
Chemical Engineering Education


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

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

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

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

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

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

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


FALL 2001


Deadline is June 1, 2001

Full Text

xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd