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

IRI *-jj

Special Laboratory Section

Feantre articles ...

and Chemical Engineering (it ...
Puerto Rico Mayagfiez

on the


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

T. J. Anderson

Phillip C. Wankat

Carole Yocum

James 0. Wilkes and Mark A. Burns
University of Michigan
William J. Koros
University of Texas, Austin

E. Dendy Sloan, Jr.
Colorado School of Mines

Gary Poehlein
Georgia Institute of Technology
Klaus Timmerhaus
University of Colorado

Dianne Dorland
University of Minnesota, Duluth
Thomas F. Edgar
University of Texas at Austin
Richard M. Felder
North Carolina State University
Bruce A. Finlayson
University of Washington
H. Scott Fogler
University of Michigan
David F. Ollis
North Carolina State University
Angelo J. Perna
New Jersey Institute of Technology
Ronald W. Rousseau
Georgia Institute of Technology
Stanley I. Sandler
University of Delaware
Richard C. Seagrave
Iowa State University
M. Sami Selim
Colorado School of Mines
James E. Stice
University of Texas at Austin
Donald R. Woods
McMaster University

Chemical Engineering Education

Volume 32 Number 3 Summer 1998

162 Minnesota's Matt Tirrell, Written by his Colleagues

168 Univerity of Puerto Rico, Mayagiiez Campus

174 An Experiment in Applied Optics: Determination of the Kinetics of the
Oxidation of an Organic Dye,
P. Delgado, A. Kasko, J. Nappi, R. Barat
178 Incorporating Safety into a Unit Operations Laboratory Course, Julia A. King
184 Getting the Most out of a Laboratory Course, Aziz M. Abu-Khalaf
190 A Simple Method for Determining the Specific Heat of Solids,
K. Hellgardt, G. Shama
194 Using Peer Review in the Undergraduate Laboratory, James A. Newell
198 Low-Cost Experiments in Mass Transfer: Part 4. Measuring Axial Dispersion in
a Bubble Column, M.H.I.Baird, K. Nirdosh
202 Unit Operations Lab: Mass Transfer and Axial Dispersion in a Reciprocating-
Plate Liquid Extraction Column,
June Luke, N. Lawrence Ricker

206 The New Faculty Member, Rebecca Brent, Richard Felder

208 The Practical Side of Chemical Engineering at the University of New
Brunswick, Guido Bendrich, Todd S. Pugsley
234 A Seminar Course on Professional Development, Edmond I Ko

214 Case Study Projects in an Undergraduate Process Control Course,
B. Wayne Bequette, Kevin D. Schott, Vinay Prasad, Venkatesh Natarajan,
Ramesh R. Rao
220 Teaching Antiwindup, Bumpless Transfer, and Split-Range Control,
Serena H. Chung, Richard D. Braatz
238 Quantifying the "Curve," Jude T. Sommerfeld

224 An Introduction to Process Flexibility: Part 2. Recycle Loop with Reactor,
W.E. Jones, J.A. Wilson

230 Human Societies: A Curious Application of Thermodynamics,
Erich A. Miller

> 189, 223 Book Reviews

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 1998 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 responsibilityfor them. Defective copies replaced
if notified within 120 days of publication. Write for information on subscription costs and for back copy costs and
availability. POSTMASTER: Send address changes to CEE, Chemical Engineering Department., University of Florida,
Gainesville, FL 32611-6005. Periodicals Postage Paid at Gainesville, Florida.

Summer 1998

W educator



Matt and teaching assistant Rafael Galvan demonstrating nylon polymerization during a poly-
mer chemistry lecture. Galvan, now at Dow Chemical, coauthored Polymerization Process Model-
ing with Matt, Neil Dotson (Eastman Chemical), and Bob Lawrence (University of Massachusetts).

University of Minnesota Minneapolis, MN 55455

In a department that has been home to such chemical
engineering legends as Neil Amundson, Rutherford Aris,
and Skip Scriven, Matt Tirrell has carved out a niche for
himself as a modern-day Renaissance Man. Applying a no-
tably broad knowledge of chemical engineering and materi-
als science to a mulligan stew of engineering research, he is
a marked asset to a department emphasizing collegiality and
a team-teaching philosophy.
Abundant energy and a gift for organization allow Matt to
keep multiple and varied projects moving forward at the
same time. After 18 years of making a name for himself,
Matt took the reins of the Chemical Engineering and Materi-
als Science Department at the University of Minnesota in

1995, and in so doing he added one more huge responsibility
to his shoulders, but he did it without neglecting his other
duties as scholar, teacher, mentor, editor, and researcher. His
colleagues are justifiably amazed and proud of his diversity
and leadership.
Matt was born September 5, 1950, in Phillipsburg, New
Jersey, on the Delaware River border with Pennsylvania,
and is the oldest of three children. His younger brother,
Dave, also developed a strong interest in polymer chemistry
and is currently Professor of Chemistry and Chemical Engi-
neering at Caltech. The youngest sibling, Mary, is raising
three children in Bangor, Pennsylvania, close to Phillipsburg,
and helps in her husband's construction business in addition

Copyright ChE Division of ASEE 1998

Chemical Engineering Education

4 Matthew V. Tirrell,
Matthew V. Tirrell, Jr.,
and Matthew V. Tirrell,
III, (our Matt), at 9
months, on the family
porch in Phillipsburg,
New Jersey.

Matt, proud
graduate of
High School
in 1968. >

A Matt (3) and
Dave (9 months).
Dave had Matt's
support even before
he could sit up by
himself. Dave
followed a similar
career path and is
now a professor at

to working hard on the local school board.
Matt's parents were divorced when he was still quite young,
and he was largely raised by his mother. His summers were
spent at a swimming club, playing golf, and working as a caddie.
One of his childhood friends was Jeff Brinker, who is now a
ceramist at Sandia and a chemical engineering professor at the
University of New Mexico.
Matt entered college at Northwestern University in 1968.
While trying to decide the direction of studies he would take, he
was swept up in the anti-war protests of the early 70s, with the
result that for a short time he considered a life in politics. It's to
our benefit that he decided on a more concrete, if no less conten-
tious, field.
While at Northwestern, Matt became interested in polymers
during a co-op term spent doing PVC compounding at Cincin-
nati Milacron. When he returned to campus, he sought out Bill
Graessley for a polymer undergraduate research project. (Matt
spoke at Bill's retirement from Princeton in May.) After gradu-



Joe McHale (Chemistry, Belvidere High School)
excited Matt's interest in chemistry (and his brother
Dave's too!) and encouraged him to enter the field.

Josh Dranoff (Chemical Engineering, Northwest-
ern) taught in Northwestern's high school summer
program between Matt's junior and senior years.
He helped Matt see possibilities in chemical engi-
neering as a career. Matt's dad, who was an ME,
also encouraged him to consider chemical engi-
neering over chemistry.

Bill Graessley (Chemical Engineering, Northwest-
ern) On returning from his Milacron co-op to North-
western, Matt sought out Graessley to learn more
about polymers. Bill asked Matt to set up the first
anionic polymerization experiments. He did-in a
coke bottle. Matt continued working in Bill's lab
when he wasn't co-oping. The position had perks
such as a desk-not bad for an undergraduate.
Graessley encouraged Matt to go on to graduate

Stan Middleman (University of Massachusetts)
Matt picked polymer science and engineering at
University of Massachusetts because of its polymer
reputation and because it brought him closer to his
chemistry interests. A Northwestern University
friend, Bob Weiss was already there. Matt picked
Stan as his advisor for the interesting and imagina-
tive research he was doing. Stan was an excellent
teacher, a model Matt emulated when he went to

Skip Scriven and Chris Macosko (University of
Minnesota) Jim Douglas recommended that Matt
"go for the best," and he did. Among many things
he learned from faculty in his early years at Minne-
sota include Chris Macosko setting the tone of
collaboration and sharing all major equipment
and Skip Scriven modeling organization of large
quantities of information from seemingly diverse

Summer 1998

Minnesota's five

"living Heads":

Tirrell (1995 present)

Rutherford Aris (1975 78)

Ken Keller (1978 80)

Neal Amundson (1950 75)

Ted Davis (1980 95)

ating from Northwestern, Matt chose the University of Mas-
sachusetts for graduate studies in the Polymer Science and
Engineering Department.
In 1977, Tirrell accepted an offer to join the Chemical
Engineering and Materials Science Department at the Uni-
versity of Minnesota, and he has been there ever since.
When the Department Head, Ted Davis, left to become Dean
of the Institute of Technology, Matt became the new Depart-
ment Head at Minnesota. Along with the usual daily chal-
lenges of running a large department (not the least of which
is communicating to 30-plus busy, outstanding, and very
active faculty), Matt's tenure has the added responsibilities
of overseeing a transition to the semester system and plan-
ning a building addition.
Matt's leadership style clearly leans toward the inclusive.
It is important to him that the faculty be kept fully informed
at each developmental stage of major projects. In a tradition
started by Neal Amundson, many of the faculty share a large
round table during lunchtime, facilitating the exchange of
ideas and information. By involving as many faculty as
possible in all facets of departmental matters, he helps them
focus on the common goal of "how best can we project our
unified department to the several external constituencies,
i.e., the chemical engineering, materials science, biomedi-
cal, polymer chemistry communities, etc."
When prompted, Matt is more than willing to talk about
the strengths of his department. He believes that an excellent
and caring staff in the office, the shop, the lab, in computing,
and in accounting is the backbone of the department. Self-
less colleagues, notably Associate Head William Gerberich,
the Directors of Undergraduate Studies, David Shores and
Wei-Shou Hu (Materials Science and Chemical Engineer-
ing, respectively) and of Graduate Studies, John Weaver and
Lanny Schmidt, (Materials Science and Chemical Engineer-

ing, respectively) make his job a lot easier. He is grateful for
their willingness to manage the huge burden of the semester
transition, a project in which "no one wants to invest time,
but the outcome of which concerns everyone."
One of the difficult challenges faced by a department
head is maintaining connections to his own research com-
munity, but Matt has managed quite well in that respect. His
research focuses on huge molecules in small spaces, and his
work with block copolymers has led to innovations that have
allowed oil companies to extract more oil from porous rocks
and to create smoother-flowing solutions used in paints,
solvents, and toner for copying machines. Later, while lead-
ing the polymer program at the Center for Interfacial Engi-
neering (CIE, an NSF-funded academic/industry/government
collaboration), several companies with specific research prob-
lems expressed an interest in pursuing biomedical applica-
tions of polymer research. Having done his PhD thesis on
the effect of flow on proteins, Matt jumped at the chance to
expand the program. He developed the notion that some of
the same ideas used to prevent fine particles in inks and
toner from forming clumps could be used in biomedical
engineering problems. As a result, part of his research group
began working on treating surfaces to improve biological
interaction. A new Biomedical Interfacial Engineering arm
was added to the CIE, which later developed close con-
tacts with the Biomedical Engineering Center (BMEC) at
the Medical School.
Matt is valued as a conduit for interactions between the
physical sciences and the medical school at Minnesota. As
Prof. Leo Furcht, Head of the Department of Laboratory
Medicine and Pathology has put it, "Matt has proved his
skill at building consensus among people from many depart-
ments who have mutual and conflicting interests. Added to
that, he is internationally regarded as a leader and innovator
in polymer science, which is critical to the future of bio-
Chemical Engineering Education

His research focuses on huge molecules in small spaces, and his work with block
copolymers has led to innovations that have allowed oil companies to extract
more oil from porous rocks and to create smoother-flowing solutions used in
paints, solvents, and tonerfor copying machines.

medical engineering."
Tirrell has a reputation
among his graduate students, (38 For
PhD students and 13 MS stu- and
dents) as a resourceful and car- Now
ing mentor. His dedication to
their education is a result of his N B
pride in his students. He relishes
their successes. Matt is known loans Btsas *
to show remarkable restraint in Arup Chakraborty UI
letting his prot6g6s find their Nily Dan University
own direction and to think for Angela Dillow Unive
themselves, guiding them with Steve Granick Unive.
a gentle hand when they seem Nino Grizzuti Univer
to be heading off course. Rasti Mi l K -
Michael Kilbey Clem
Levicky, soon to be teaching at
Columbia, echoes the thoughts Rasti Levicky Colum
of many of Tirrell's former stu- Jaye Magda Univers
dents when he says, "I learned Guangzhao Mao Wa)
independence from Matt's be- James Schneider Car
ing open to what I thought were John Torkelson Nortl
interesting and important things
to do." Matt has an informal
style that encourages students to discuss issues with him,
and believing it important to their education, he fosters inter-
actions among the students themselves. He gets them to talk
as a way to learn. He also stresses the importance of commu-
nication in writing and oral presentations. "If you cannot
make your information come across to your audience, you
have to question your own understanding of the material."
It is clear that Matt enjoys his work. In 1992, during a trip
to NIST in Washington, DC, Matt, Frank Bates, and gradu-
ate student Kurt Koppi (now at Dow Chemical) were trying
out for the first time an instrument Koppi had built. They sat
down and almost immediately discovered something they
hadn't expected-block copolymer orientation perpendicu-
lar to the flow. They had a great time collecting valuable
data on a process that had never been seen before, reveling in
the excitement of first discovery. The data was shortly there-
after written up for Physical Review Letters, and has be-
come a frequently cited paper.
Matt's goes to great lengths to help his students find
interesting jobs after they have completed their studies. They
must see something they like about the way he manages his
group since a large proportion of them choose academic


of De
ssity o
bia U
ity oJ
ne S

Summer 1998

careers for themselves.
rell's They see him working ex-
Students tremely hard, juggling
st-Docs many projects, and doing it
Academia well, and it obviously in-
spires them to do the same.
As Kurt Koppi says, "Matt
ic University is a terrific mentor. His di-
y of Florida verse background spans the
sity of California, Berkeley whole spectrum of chemi-
laware cal engineering. The depth
of Massachusetts and breadth of his knowl-
of Illinois edge helps his students
when they are learning to
Naples do research, learning to
University focus their thoughts and
University to see the overall picture.
Utah Matt is able to help them
tate University sort out the key features
Mellon University and organize their thought
ern University processes to get to the
main issues." In 1981,
thanks to the nomination
of his students, Matt's
teaching was recognized with a Gordon Starr Outstand-
ing Contribution Award.
Matt has been similarly dedicated to the health of his
profession. He has served on the editorial boards of Journal
of Polymer Science, Polymer Physics Edition, Macromol-
ecules, Journal of Chemical Physics, Journal of Rheology,
and other idea forums. His most demanding service position
currently is as editor of the AIChE Journal.
In the course of his career Matt has also earned the recog-
nition of his peers. He has received the NSF Presidential
Young Investigator Award, a Guggenhein Fellowship, the
AIChE's Allan P. Colburn and Charles M.A. Stine Awards,
the John H. Dillon Medal from the American Physical
Society, the Alumni Merit Award from Northwestern
University, and last year was elected a member of the
National Academy of Engineering.
Tirrell's accomplishments have not gone unnoticed by
the University of Minnesota. When the University began a
capital campaign to increase the school's endowment, part
of its program was to increase the number of endowed chairs
to help retain accomplished professors. At a youthful 36,

A gourmet,
Matt has
developed a
flair for
meals. With a
for French
Proven al
cuisines, he
has built a
collection of
some 200
with recipes
from around
the world...
In addition to
[Matt enjoys]
[and] golf.

Matt was singled out for the Shell Chair, which Matt has used to enhance the polymer
facilities at Minnesota and to nurture his research group.
Attracting and keeping Matt was a shrewd move for Minnesota. Matt has served the
University community far beyond the call of duty, an indication of how deeply he
believes in the value of education. He has acted on numerous committees, including
(currently) the key Faculty Consultative Committee, and has led faculty searches,
chaired two separate departments, been a faculty representative to the Senate, and
chairman of the faculty club.
Among his many attributes, Matt is a fund-raiser par excellence. Both chemical
engineering and materials science, and the Biomedical Engineering Center are vastly
better off as a result of his efforts. What makes him successful at it? He works hard to

lht of Matt's reputation as a cook, an invitation to dinner at
"Chez Tirrell" is a coveted prize!

identify possible sources, he addresses himself to links of common purpose between the
University department and the funding source, he draws in and ties together disparately
skilled colleagues to present cogent possibilities, and he pays attention to the little
details that give proposals their air of imminent success. In other words, Matt is
comfortable thinking both big and small.
Outside the University, Matt has done volunteer work for the American Civil
Liberties Union and has been on the board of directors of a prominent Minneapolis
theatre, Theatre de la Jeune Lune. He likes being involved, feeling that he's contributing
something to his community.
As hard as he works at his public duties, in private, Matt really cooks. A gourmet,
Matt has developed a flair for preparing exquisite meals. With a special appreciation for
French Provenqal and Mediterranean cuisines, he has built a collection of some 200
cookbooks with recipes from around the world. In addition to cooking, as an assistant
professor, rugby was Matt's game, but as an associate professor he turned to more
genteel pursuits--distance running. He completed two Twin Cities Marathons and for
five years belonged to an afternoon running club. Now, while he has again taken up
distance running, he keeps up his health principally through membership in a
fitness club and by golfing (in 1995, his best-ball foursome won the department's
Chemical Engineering Education

4 Matt and his wife, Pam, enjoy traveling

together. Here, they have just landed

in Beijing, China.

The Tirrellfamily-Matt with

his mother, Lorraine,

his sister Mary,
and his brother Dave. V

coveted "Amundson Open" title).
Matt also enjoys other sports, such as skiing and biking, but
acknowledges that a love of sports can get out of hand. He
credits his wife Pam for helping him "balance my conventional
male tendency to antisocial activities such as sports" with
movies and plays. Nevertheless, Matt holds Minnesota
Timberwolves season tickets.
Pam and Matt met when she was a writer and editor for
University Relations at Minnesota. Matt had won an NSF Presi-
dential Young Investigator Award and Pam was assigned to
interview him. We can only speculate that she liked what she
saw. Pam grew up in International Falls, Minnesota ("America's
Icebox") and remembers separate changing rooms for boys and
girls on the city's many ice rinks. She says, "It was tough to
play hockey in a skirt, but we did." Pam has since moved on to
work in guest relations at the historic St. Paul Hotel. She and
Matt share a restored townhouse near downtown St. Paul, not
far from where F. Scott Fitzgerald, and later Garrison Keilor,
lived. Pam and Matt have no children, but enjoy visiting with
their twelve nieces and nephews.
That Matt maintains continuity in his many interests is all
the more remarkable when one realizes how much time he
spends in the air. Traveling has played a significant role in
Matt's career, and brings together the two strands of his public
and private life. He has been a visiting professor in cities like
Bahia Blanca, Guadalajara, Canberra, and Paris. He recently
received notification that he'd surpassed one million actual
miles traveled (to say nothing of frequent-flier bonus miles),
and even this understates the extent of his travels. He often
accepts speaking invitations, and his various collaborations re-
quire him to take quick trip here and there to keep up with
projects. For example, his consulting work with the state and
Summer 1998

government consortium, French Petroleum Institute in
Paris, takes him back to France at least once a year.
For his effort, there have been numerous returns for
Matt. He has seen academic programs started, students
graduating, projects being well funded, professional hon-
ors and awards coming his way. but what higher honor
could there be than to have a former graduate student say,
as many of Matt's students do, "When I meet people in the
field, I'm proud to tell them I worked with Matt Tirrell."
As you converse with Matt Tirrell you realize that you
are with a unique individual; one with the rare ability to
convey ideas in well-crafted complete sentences. Every-
thing Matt says is well considered and easy to follow. This
characteristic may be the key to his success in both re-
search and in teaching. "To help myself pay attention in
lectures during my sophomore year at Northwestern, I
started to think about how I would explain the subject. I
feel that this carries over directly to research. Explain-
ing things clearly is intimately connected to doing good
research." As the end of your appointment approaches,
you notice Matt cocks his head to check the time on
your watch. This busy man does not wear one. With a
subtle and not unfriendly gesture, Matt suggests he
must be going. 0


Chemical Engineering at the University of

Puerto Rico
Mayagiiez Campus

Chemical Engineering facilities at the University of Puerto Rico-Mayagiiez.

During the last four years, the Department of Chemical Engineering at the Univer-
sity of Puerto Rico-Mayagiez Campus has undergone a major transformation. We
developed and implemented a strategic plan that provided specific goals and
guidelines for the transformation, with the main goal of the Department being to
become one of the top fifty US chemical engineering departments by the year 2002.
Our mission is to
Satisfy the technological needs of Puerto Rico related to chemical engineering.
This mission will be accomplished by means of teaching, research, and services to students
coming from all socioeconomic levels. These students, in turn, will become competitive
professionals with a global perspective and with a clear understanding of their social
The University of Puerto Rico-Mayagtiez is located on the west coast of the island of

[The] main
goal of the
[is] to become
one of the top
fifty US
by the year
2002.... This
mission will
by means of
research, and
services to
coming from

Copyright ChE Division of ASEE 1998

Chemical Engineering Education

Puerto Rico, about ninety miles from San Juan. The location is relatively close
to beautiful beaches as well as other island attractions, such as a tropical rain
forest called "El Yunque," a phosphorescent bay located at the town of Guanica,
the Arecibo Observatory, and national monuments such as Porta Coeli, which is
located at San German.
The Department of Chemical Engineering at the University of Puerto Rico
saw its first four students graduate in 1930 with degrees in chemical engineer-
ing with emphasis on sugar cane refining. In 1948, the University approved the
Department's first Chemical Engineering Program with emphasis in other
areas, and in 1978, the Department moved to its present location, which is
part of the engineering complex at the University. The Department pres-
ently has a faculty consisting of 24 professors, of which 18 have PhD
degrees and 6 have MS degrees.
The 36,600 ft2 facility consists of administrative and faculty offices, eight
modern classrooms, an amphitheater, ten research laboratories, a pharmaceuti-
cal process laboratory, a unit operations laboratory, and two computer centers.
There is also an office for the Student Chapter of the American Institute of
Chemical Engineers (AIChE) and for Puerto Rico's Institute of Chemical
Engineering (IIQPR). As part of the services available to the students, the
department has a Student Aid Center (SAC) facility, a non-profit corporation
that provides services such as photocopying and supplies.


Undergraduate Program The undergraduate progr
has an average enrollment of 170 students per year, of wh
55% are female. This includes approximately thirty to fo
internal and external transfer students per year. The progr
is ranked among the top ten in the United States based
number of degrees awarded (see Table 1).
As part of its undergraduate program, the Department
fers a Bachelors in Science Degree in Chemical EngineerJ
(BSChE) with several elective courses as options. In the 1
68 years, the chemical engineering program at the Univers
of Puerto Rico has graduated more than 2,400 engineers, I
largest source of Hispanic chemical engineers in the Unii
States (see Figure 1).
The basic undergraduate curriculum, accred-
ited by the Engineering Accreditation Com-
mission of the Accreditation Board for Engi- 140
neering and Technology (ABET), requires 172 | 120
credits/hours for completion in five years. It > 1oo
includes courses in basic sciences and other
branches of engineering to provide the stu- S
dent with a sound, fundamental, scientific, 6
technical, and sociohumanistic education. The 40
students are required to take courses in ad- 20-
vanced mathematics, physics, and chemistry 0
during the first two years. At the beginning of
the third year of studies, the students start a
three-year program dedicated mainly to chemi-
cal engineering courses (see Table 2, Section
Summer 1998

Students in the well-equipped com-
puter center at UPR-Mayagiiez.

Degrees Awarded and Ranking
(Based on degrees Awarded) from 1990 to 1996

# of Degrees Ranking'
Awarded (# of schools)
66 8(126)
68 8(120)
88 4(113)
142 1(105)
97 7(124)
132 7(134)
105 9(141)

# of Degrees Ranking'
Awarded (# of schools)
11 26(122)
7 53(115)
1 99(108)
4 67 (97)
8 43(117)
5 83 (127)
15 15 (133)

SData obtainedfrom the Annual Report of the American Chemical Society (ACS)
Committee on Professional Training (Chenmical and Engineering News).

Figure 1. BSChE degrees granted from UPR-Mayagiiez

The Department has eight classrooms and one amphi-
theater to accommodate the 450+ enrolled students.

1). Also, as part of the BSChE curriculum, the Department
offers a number of specialized courses (see Table 2, Section 2).
Undergraduates can also take graduate courses as electives (see
the section on the graduate program).

The undergraduate research courses offer students an opportu-
nity to obtain first-hand experience in the latest developments in
areas such as electrochemistry, photocatalysis, surface catalysis,
process control optimization, biochemical engineering, biomedi-
cal engineering, and supercritical fluids. Our program also of-
fers several elective course options, in collaboration with other
engineering departments, such as environmental engineer-
ing, manufacturing engineering, and (in the near future) a
biotechnology option.
The manufacturing engineering option is part of the Manufac-
turing Engineering Partnership (MEEP) created in 1994 by three
engineering schools: Pennsylvania State University, the Univer-
sity of Washington, and the University of Puerto Rico-Mayagiiez.
It is cosponsored by the Procter & Gamble Foundation and aims
at providing a proper balance between science and engineering
practice in such a way that the students will develop the skills
employers value. The courses are supported by a learning fac-
tory and laboratory facilities for hands-on activities integrated
into the courses and field trips.
The BSChE program has an excellent reputation both on the
island and on the mainland. Many prestigious local and US
companies, such as Amoco, Kodak, Xerox, Union Carbide,
Champion, Pfizer, Abbot, Johnson & Johnson, Pharmacia-
Upjohn, ARCO Chemical Co., DuPont, Procter & Gamble, and
Merck Sharp & Dohme recruit on our campus (see Table 3). In
addition, state government agencies such as the Environmental
Quality Board, Aqueduct and Sewer Authority, Solid Waste
Management Authority, and the Electric Power Authority re-
cruit our graduates. An average of twenty students per year
continue their education with graduate studies.
It should be mentioned that several foundations, such as Procter
& Gamble, AMOCO, DuPont, and Sloan provide financial as-
sistance to develop and promote research activities at the under-
graduate level.
Graduate Program The graduate program was established

Alumni Profile and Employability

# of Graduated
Students Working
in the US'

Total %2

8 21.6
3 6.8
15 18.3
16 16.8
11 12.6
3 2.3
1 1.2
6 5.0
12 11.9
13 9.3

'Data obtained from the Placement Department, Univ. rn,., ,.. ....
2 Percent taken from the total amount of students graduated that academic year.

Chemical Engineering Education



# of Graduated
Students Working
in Puerto Rico'

Total %2

6 16.2
14 31.8
23 28.0
31 32.6
17 19.5
37 28.2
32 39.0
69 58.0
30 29.7
30 21.4

# of Students
Working Toward a
Graduate Degree'

Total %2

7 18.7
10 22.7
23 28.0
27 28.4
18 20.7
32 24.4
16 19.5
17 14.3
21 20.8
25 17.9

Courses Offered

Standard Courses
* Material and energy balances
* Momentum transfer operations
* Chemical engineering thermodynamics (2 semesters)
* Heat transfer operations
* Kinetics and catalysis
* Mass transfer operations
* Unit operations laboratory (2 semesters)
* Analysis and control of processes
* Process design (2 semesters)
* Mathematical analysis in chemical engineering
* Chemical Engineering electives

Specialized Courses
* Advanced process control
* Air pollution control
* Computer simulation of processes and units
* Equilibrium stage processes
* Microclimate and dispersion of air pollutants
* Industrial waste control
* Introduction to biochemical engineering
* Introduction to biomedical engineering
* Particulate systems
* Pharmaceutical process design
* Plastics technology
* Transport phenomena
* Undergraduate research
* Unit operations in food processing

in 1972 and offers programs leading to Master in Science
(MS) or Master in Engineering (ME) degrees and, in the
near future, a program leading to a PhD in chemical
engineering. The MS degree (Option 1) requires comple-
tion of advanced courses and research in chemical engi-
neering and requires a thesis report plus a final oral ex-
amination. Option 2, the ME degree, differs from Option
1 in that the students develop an advanced project; it also
requires a final oral examination. The courses offered for
these options (see Table 4) cover most of today's top
chemical engineering areas.
The Department's graduate program has generated more
than three million dollars in research proposals during the
last five years and has obtained funding from agencies
and institutions such as the National Science Foundation
(NSF), National Institutes of Health (NIH), De-
partment of Energy, Department of Defense,
and Sandia National Laboratories. As a result
of the research activity, the Departments' fac- G,
ulty gave more than 60 presentations in techni-
cal conferences and other activities and sub- Advanc
mitted almost 40 articles for publication during Advanc
the 1993-97 period. The Department's research
and development areas include biochemical,
biomedical, catalysis, reactors, colloids, inter- Advanci
faces, materials, expert systems, control, poly- Advanc
mers, composites, thermodynamics, transport, Catalysi
separations, environmental ChE, and energy. Electric
Figure 2 shows the number of students gradu- Mathem
ated from the Masters program in the years
1974 to 1997.
One of the key priorities of the University of Plant de
Puerto Rico is the development of graduate Selected
programs and research activity. As a result of Separati
that initiative, the Chemical Engineering De-
partment has developed a strategic plan with
the PhD program as one of its primary compo-
nents. The implementation of the PhD pro-
gram is critical to reaching the Department's j 16
main goal of being among the top 50 US de- 14 -
partments. The program has the following ob- 12
jectives: o10

To educate the students in how to master and 8 8
apply scientific methods as a fundamental tool 6
in research. 4
To develop the students' capacity to make 2
original contributions to the field of chemical
To develop in the students a high sense of
social and ethical responsibility, knowing not
only the technical and economical aspects of

Summer 1998

The Department's graduate program has
generated more than three million dollars in
research proposals during the last five years and
has obtained funding from [numerous]
agencies and institutions ...

their work, but also safety, health, and environmental protection
The program will require a minimum of 58 credits/hours
for completion. The new courses designed for the PhD that
will be added to the graduate courses already developed are
listed in Table 5.

4- 70
{ 60



Figure 2. Master's degrees granted by UPR-Mayagiiez.

graduate Courses Offered

ed heat transfer
ed process control
ed reactor design
ed thermodynamics
ed transport phenomena
chemical engineering
atical methods in chemical engineering
al methods in chemical engineering
al process optimization
al process simulation
topics in biochemical engineering
on process analysis

PhD Courses to be Added
to Curriculum

* Models for flow systems in
chemical reactors
* Atmospheric transport
* Special topics in heteroge-
neous catalysis
* Food fermentation and
* Special problems at doctoral
* Finite elements in transport
* Doctoral
* Doctoral

The photograph at the left shows

the east section of the 3,500ft2

unit operations laboratory,

and the west section is

shown in the photograph below.

^*^Tw 1 7- ,_

The Department has two facilities dedicated exclusively to unit
operations, process control, reactor design, and pharmaceutical
operations. The laboratory facilities reflect the requirements of
the undergraduate program and are fully equipped for their
effective use.
The 3,500 ft2 Laboratory of Unit Opertions is equipped with heat
exchangers, hydraulic benches, flow-measuring devices, a cooling
tower, an absorption tower, distillation columns, chemical reac-
tors, and equipment for digital process control. The facility also
has a wet chemistry laboratory equipped with analytical instru-
mentation such as gas chromatography, high-performance liquid
chromatography, UV-Vis spectroscopy, and atomic absorption.
These facilities are undergoing major changes to offer the stu-
dents a state-of-the-art laboratory. Some of the changes include
the addition of high-tech chemical analysis instruments, updat-
ing and validation of existing experiments, and the creation of
new experiments.
As part of the addition of new experiments, the department is
planning to develop and install a laboratory module dedicated
exclusively to process and manufacturing control using program-
mable logic controllers and a virtual control room. In addition, the
department is evaluating the implementation of experiments in the
areas of molecular simulation, micro-chemical systems, and field
applications. These developments will give the students first-hand
experience in the latest technologies and opportunities available to
chemical engineers.

The Department's research facilities include ten modern labora-
tories, an instrumental analysis laboratory, and a computer center.
The graduate students also have access to the Central Research
Instrumentation Laboratory (managed by the Department of
Chemistry), which provides quantitative analysis services for
research projects.
The research laboratories are equipped with a total carbon ana-
lyzer, a glucose analyzer, gas chromatographs, Fourier transform

infrared equipment, high-performance liquid chromato-
graphs, a GC-MS, polarographs, an X-ray fluorescence
analyzer, spectrometers, a dissolved oxygen analyzer,
humidity analyzers, and gel permeation chromatographs.
The Central Research Instrumentation Laboratory includes
gas chromatography/FT-IR equipment and a nuclear mag-
netic resonance apparatus.
The Department is designing a 25,000 ft2 facility that
will host the Environmental & Biotechnology Research
& Incubator Opportunities Laboratory. This five-mil-
lion-dollar facility will be used by entrepreneurs and
faculty involved in research and development activities,
Also, the Department is finishing an area of 2,000 ft2 that
will be used for a research laboratory and an instrumental
analysis laboratory. All these initiatives are part of the
implementation of the PhD program.

Merck Sharp & Dohme Lecture Series The Merck
Sharp & Dohme Lecture Series was established by a
grant given from the Merck Foundation in 1972. The
purpose of the grant is to exchange new technical and
scientific developments among those who wish to ex-
plore in depth a particular research area in chemical
engineering. The grant has allowed our department to
invite distinguished scientists and recognized authorities
to lecture and has offered our graduate and undergradu-
ate students an unparalleled opportunity to know first-
hand these leading scientists and their projects. Most of
the authors of textbooks used in our Department have
been invited as lecturers in this series. It is one of the
longest uninterrupted lecture series in America (25 years).
Chemical Engineering Education

Above is Dr. Nelson Cardona's micro-
calorimetry laboratory, and below is
Dr. Jose Colucci's environmental
research projects laboratory

Environmental Symposia As part of the activities celebrated
annually in the Department, the Environmental Symposia are an
essential part of the strategic plan initiatives to exchange informa-
tion between the Department, the industrial sector, and the com-
munity. Merck Sharp & Dohme cosponsors these seminar ses-
sions from industry and academia on topics related to environ-
mental issues. A poster presentation is given by undergraduate
and graduate students.
Process Design Course The Department has established a
partnership with industry to assure that our students complement
their academic experience with applied projects. In 1996, the De-
partment modified the Process Design II course, incorporating
business experiences into the projects as part of the initiative
funded by Procter & Gamble Foundation. The business approach
includes assigning real projects or problems from the industrial
and governmental sector with emphasis on the development of
communication and leadership skills. The final work is presented
to representatives of industry and government agencies for their
consideration and evaluation.
Honor Student Activity The Department, in collaboration
with the student chapters of AIChE and IIQPR, celebrates the
Summer 1998

Honor Students Activity each year. It is dedicated to
undergraduate students with a GPA higher than 3.00 (on
a 4.00 scale), recognizing their hard work and dedica-
tion. The activity consists of an open house, an official
ceremony, and a lunch. More than 300 guests come to
the event each year.

The Department is developing several initiatives that
will support its intention to serve Puerto Rico. These
initiatives include the construction of an Environmen-
tal & Biotechnology Research & Incubator Opportu-
nities Laboratory, implementation of the PhD pro-
gram, a new distance learning program, and develop-
ment and implementation of a new millennium unit
operations laboratory.
Also, the Department is updating the existing com-
puter facilities with the latest software and hardware.
One of the initiatives of the Department is to create a
multipurpose computer facility in order to integrate the
use of computers at all levels in the chemical engineer-
ing curriculum. In addition, the facility will support the
development of the ME program via videoconferencing.
We have been working to improve our program in
order to continue graduating the finest engineers and to
solve present and future technical challenges. As the
major source of Hispanic engineers in the nation, we
are committed to continue these trends for the benefit
of our society. We are looking forward to being one
of the top fifty U.S. chemical engineering depart-
ments by the year 2002!
Additional information about the chemical engineer-
ing program can be obtained by contacting

The University of Puerto Rico Mayagiiez Campus
Department of Chemical Engineering
PO Box 9046
Mayagilez, Puerto Rico 00681-9046
Phone: (787) 832-4040, ext. 2568
FAX: (787) 265-3818

We thank Prof. Arturo Hernfndez-Maldonado, Dr.
Marla P6rez-Davis, Dr. Jos6 Colucci-Rios, and Prof.
Lueny Morell for their contributions to this article.
We also acknowledge the help of Dr. Jack Allison, Dr.
Nelson Cardona, Prof. Federico Padr6n, Dr. Lorenzo
Saliceti, the Placement Department of the University,
and the University's Department of Publications. Photo-
graphs are courtesy of Mr. Carlos Diaz and the
University's Department of Publications. 0



Determination of the Kinetics

of the Oxidation of an Organic Dye

New Jersey Institute of Technology Newark, NJ 07102

Applied optics is gaining in importance for science
and engineering students. A National Science Foun-
dation (NSF) workshop'" recently recommended an
emphasis on optics research and education because
Optical Science and Engineering is an enabling technol-
ogy-that is, a technology with applications to many
scientific disciplines and with the potential to contribute in
significant ways to those disciplines.
This workshop recommended a strong optical component in
the undergraduate curriculum of science and engineering
students. An interdisciplinary group at New Jersey Institute
of Technology (NJIT) has created such an applied optics
curriculum.121 It is supported through the Combined Re-
search-Curriculum Development program of NSF and has
been very successful.13'4
Optics should no longer be viewed as a tool of only the
physicist. Optical principles, especially as applied to spec-
troscopy, are obviously important to chemistry students. How-
ever chemical engineering students are not usually exposed
to optics beyond general undergraduate physics courses.
There is a definite need to expose these students to more
optics, especially since optical-based sensors for process
monitoring are rapidly expanding into the traditional chemi-
cal industries.
Optical diagnostics applied to a chemical reactor offer the
chance to demonstrate two principles: 1) the determination
of reaction kinetics without direct sample analysis of reac-

Paglo Delgado, Jr., is currently employed by Merck & Company as an
analyst. He earned his BS in chemical engineering from the New Jersey
Institute of Technology in 1998.
Anna Kasko is currently employed by FMC Corporation as a process
engineer. She earned her BS in Chemical Engineering from the New
Jersey Institute of Technology in 1997.
Jarod Nappi is currently employed by Merck & Company as a process
engineer. He earned his BS in chemical engineering from the New
Jersey Institute of Technology in 1997.
Robert Barat is currently an associate professor of chemical engineer-
ing at the New Jersey Institute of Technology. He received his PhD in
chemical engineering from the Massachusetts Institute of Technology in
Copyright ChE Division of ASEE 1998

tants or products, and 2) rapid monitoring for process con-
In this paper we will describe an effective, simple, and
inexpensive experiment in applied optics for monitoring the
oxidation of an organic dye. Absorption of a laser beam
passing through a batch reactor produces data from which
overall kinetics are determined. The use of an in-situ laser
beam diagnostic avoids the cumbersome, potentially inaccu-
rate, and costly method of direct sampling for optical ab-
sorption in an ultraviolet-visible spectrophotometer.

An organic food color is a complex organic dye with a
broad-banded absorption spectrum. In the case of a helium-
neon (He-Ne) laser at 632 nanometers (nm), the dye of
choice is one with a spectrum with a reasonably strong
absorption at the red 632-line, i.e., blue or green. Transmis-
sion of the beam through a volume of absorbing dye solution
is governed by the Beer-Lambert law151
It = I exp(-oCL) (1)
I, transmitted intensity
I, incident intensity
a absorption cross section
C dye concentration
L optical path length (distance laser beam passes through dye

Typically, Eq. (1) is rewritten as

A= tnI =o CL (2)

where is absorbance. Since there is negligible absorption
by water at 632 nm, Io can be taken to be the intensity of the
laser beam exiting the reactor.
The oxidation reaction can be written as
A + bB -> products (3)
Chemical Engineering Education

[ Laboratory

The [NSF] workshop recommended a strong optical component in the undergraduate
curriculum of science and engineering students ... In this paper we will describe
an effective, simple, and inexpensive experiment in applied optics
for monitoring the oxidation of an organic dye.

A dye
B oxidant
b overall stoichiometric coefficient
indicating the number of moles
of B consumed for each mole of
A oxidized.

The assumption is that Eq. (3) prob-
ably represents an overall stoichi-
ometry, i.e., the dye oxidation oc-
curs via a mechanism of several el-
ementary reactions. Overall reaction
kinetics, however, can be deter-
The rate of reaction (-rA) can be
written as
dCA n~ m
-rA dCA -kCnC B (4)
k overall rate constant
CA,CB concentrations of dye and oxi-
dant, respectively (mole/cm3)
n,m overall reaction orders

If the reaction is performed when Figure 1. Exp
the concentration of oxidant is in
considerable excess over that of the dye (i.e., CA< C, is effectively constant. Equation (4) can then be rewritten
-rA dA k' CA (5a)
k' =kCC (5b)
Combining Eqs. (5a) and (2) results in the first working
relation needed to model the experimental data:

dA k* An (6a)

k =k'(L)' (6b)
and is a constant for a given experiment with a specific
amount of excess bleach. Taking the logarithm of Eq. (6a)

en -a)= (k*)+n en(A) (6c)

Summer 1998

'erimental layout.

Using a differential approach,
curves of A. vs. time are generated
from a signal trace. Slopes of these
curves, representing the time rate of
change (-dA/dt), are determined and
then plotted vs. absorbance A ac-
cording to Eq. (6c). The "best fit"
t slope is the order n, with intercept
giving k*.
Alternatively, using an integral
approach, two cases can be derived
from Eq. (6a) depending on the value
of n. If n=l, then Eq. (6a) integrates

/n. k*t (7)

S where A. is the absorbance measured
initially with water and dye, but be-
18 fore the oxidant is added. If n is not
equal to 1, then Eq. (6a) integrates
ter a'-n =(n -l)k* t + (8)

Equations (5b) and (6b) are com-
bined to give

k* = k C ( L) (9)

Absorbance data are correlated with either Eq. (7) or (8) to
determine the "best fit" reaction order n and rate constant k*.
Whether a differential or integral approach is taken, the
series of k" values from runs with different volumes of
excess bleach are then correlated with bleach concentra-
tions, using Eq. (9) to determine the "best fit" reaction order
m. Taking the logarithm of Eq. (9) yields

(n(k*)= In[k(a L)-n ]+m en(CB) (10)

A plot of k* vs. C, according to Eq. (10) should yield a
slope of m and the rate constant k to within a constant
(oL)' n.

The schematic for this experiment is shown in Figure 1.
The batch reactor can be any transparent container with two
parallel flat sides that can serve as windows for the He-Ne
laser beam. The container should be appropriate in overall


size for whatever magnetic stirrer is used. We recommend a
Teflon-coated stirring bar. The intensity of stirring should be
sufficient to obtain good mixing, yet not so great that it
causes a vortex to form and extend into the laser beam path.
The incidence of the laser beam onto the front and back
sides of the reactor will generate two back reflections. The
reactor should be slightly skewed off-normal to the beam to
prevent the back reflections from reentering the laser. Once
it is set, care should be taken to avoid moving the reactor
relative to the beam. An outline of masking tape around the
base of the reactor on the magnetic stirring surface will
ensure that the reactor is always in the correct position.
The exiting laser beam is directed into a photodiode or
other optical detector. The detector should have a response
time much shorter than typical reaction times (see Figure 3).
A photodiode is definitely fast enough, in addition to being
inexpensive and readily available. Care should be taken to
avoid saturating the detector. Placement of neutral density
filters (NDFs) or other attentuators in the path of the laser
beam before the reactor will reduce the level of light hitting
the detector. Detector linearity (i.e., non-saturation) can be
checked by ensuring that a 50% drop in detector signal
occurs when a 50% NDF is placed before the detector.
The detector signal voltage can be monitored with a chart
recorder or other signal detection apparatus (e.g., analog-to-
digital data acquisition-equipped personal computer). One
configuration currently in use in our interdisciplinary
applied optics laboratory course connects the photodiode
detector to a lock-in amplifier. The laser beam is modu-
lated with a chopper, and the lock-in amplifier is inter-
faced to a laboratory computer via a general-purpose
interface bus (GPIB).
The bleach and dye reagents can be volumetrically
added to the reactor with graduated burettes suspended
over the reactor. Water should also be volumetrically
added to the reactor.


In the experiment described in this paper, the reactor is
a clear acrylic topless box with a 4x4-in square base and
a 5-in height. Durkee Liquid Food Color (blue or green)
and household laundry bleach (5.25 weight % NaOCI
active ingredient with the balance assumed to be water)
are the reagents. The bleach is delivered through a 500-ml
burette, which has a sufficiently large bore such that the
time to add the reagent was small relative to the reaction
time scale. The dye is delivered through a 5-ml burette.
A Spectra-Physics 1-milliwatt helium-neon laser is
used in conjunction with an amplified Thorlabs silicon
photodetector (Model PDA150). The voltage signal is
observed with a chart recorder. A typical simulated trace

Figure 2. Sample detector signal trace.

is shown in Figure 2. It should be noted that the "zero line"
properly corresponds to the detector signal with the laser beam
blocked so as to account for any detector voltage offset.
In a typical experiment, 350 ml of water is added to the
reactor box (about one-half full) prior to its placement on the
magnetic stirrer. Optical alignment of the laser beam, the
box, and the detector should be done with the first batch of
stirred water. Care should be taken with laser light reflecting

10 12 14 16

time (min)

Figure 3. Absorbance vs. time for run #2.
Chemical Engineering Education

- Vt --

Bleach Added
Time t = 0

Laser Through
Dye + Water

-_ ____ -- V- -Vo- -

Laser Through Zero
Stirred Water Line

[Laboratory 1

back from various surfaces. Masking tape should be placed
around the base of the reactor box to ensure that it would
always be returned to the same position so that the optical
alignment would not be disturbed.
Detector linearity is then tested. We determined that a
total optical density of 1.5 (approximately at 97% inten-
sity reduction) was needed in the path of the laser beam
in front of the reactor box.
With detector linearity established, a trace representing Io
can be recorded. One drop of dye (approximately 0.03 ml) is
added to the stirring water. A short delay should ensue to
establish a new constant signal trace, which indicates the
maximum absorbance. Then, at time t=0, a fixed volume of
bleach (at least 10 ml) is added to the stirred solution. The
signal trace is allowed to ascend over time until it reaches an
effectively constant value-typically the same value indica-
tive of a clear solution, which indicates a complete reaction.
The solution is visibly clear at this point. It should be noted

-09 08 -07 06 -05 04 03 -02 -01 0
In(A )

Figure 4. Rate vs. absorbance; differential
approach, Eq. 6c.

2 3 4 5 6 7 (
time (min)

that the laser light is not absorbed by the bleach, as evi-
denced by a test made with bleach added to plain water.
It is assumed that a given detector signal voltage, v, is
directly proportional to a given beam intensity, I, as in

v=aI (11)
where a is the constant conversion factor. Equation (2) then

A = (L _- = rn vj (12)
it) "tv
where vo is the signal corresponding to the laser beam pass-
ing through clear, stirred water, and v, is the time variant
signal corresponding to the attenuated beam as the reac-
tion proceeds. These values are represented on the sample
trace in Figure 2.


The data and results presented in this paper were taken
directly from a student group experiment in our senior chemi-
cal engineering laboratory course.
Using a differential approach, Figure 3 illustrates a plot of
absorbance A vs. elapsed time for one drop of blue dye, 20
ml of bleach, and 600 ml of water. Slopes (-dA/dt) were
determined graphically. Logarithms of the slopes are plotted
vs. (n A (as in Eq. 6c) in Figure 4. The "best fit" slope of
1.03 suggests first-order kinetics in dye concentration.
Using an integral approach for the same run, Figure 5
shows a first-order plot using Eq. (7). The correlation coeffi-
cient (R'=0.9948) is sufficiently good to declare the reaction
kinetics to be first order in dye concentration.
Table 1 summarizes the results from all the runs. Several
other runs were made with different bleach volumes, but
always with 1 drop of blue dye and the same starting volume
of water. In spite of some fluctuation in the "best fit" order,
there is no reason to suspect anything other than first order
with respect to dye.
For the case of n= the "lumped" rate constant k" reduces

"Best Fit" Reaction
Order with Respect to
Dye Concentration

Volume Order
Run Bleach n
1 10 1.15
2 20 1.03
3 30 1.03
4 40 1.27

to k', which is plotted vs. initial
bleach concentration in Figure
6 according to Eq. 9. The "best
fit" slope of 1.00 shows that the
reaction is first order in bleach
concentration. The overall rate
constant k is found to be 16.4
liter/mole-minute. An "error"
propagation analysis has esti-
mated the uncertainty in tf rate
constant at 12%.

--- Continued on page 197.

Figure 5. Absorbance vs. time; integral approach, Eq. 7.
Summer 1998

9 10 11 12 13 14

I Laboratory




Michigan Technological University Houghton, MI 49931

Chemical process safety is taught at Michigan Tech-
nological University (MTU) via two methods: a re-
quired junior-level lecture course and integrated into
the existing senior-level unit operations course.11[ The re-
quired junior-level course typically covers industrial hy-
giene and toxicology, flammability, relief systems, hazard
identification, risk assessment, accident investigation, and
case histories. After completing this class, the students are
ready to apply these topics in the unit operations laboratory,
which is the major chemical engineering laboratory experi-
ence they encounter.
The unit operations course consists of thirteen different
pilot-scale standard experiments (batch filtration, continu-
ous filtration, continuous-stirred tank reactors, cooling tower,
flow measurement, fluidization, liquid-liquid extraction, poly-
mer processing, pumping a, pumping b, single-pass heat
transfer, shell-and-tube heat transfer, and vacuum drying)
and two industrial-scale pilot plants (PSCC, Process Simula-
tion and Control Center, which uses the Honeywell Total
Plant Solutions control system), the first being an industrial-
scale pilot plant for fractionation of a water-ethanol system,
and the second being a PDMS (polydimethylsiloxane) jack-
eted-reactor pilot plant.
The students are divided into groups of four. The members
of each unit operations group remains the same for an entire
academic year, and during the year, each group conducts six
experiments (or cycles), two of which are the PSCC pilot
plants. For each cycle, each group spends one week pre-
paring for their assigned experiment, two weeks in the
laboratory conducting their experiment, and one week
writing their report.
For each laboratory cycle, one group of students is ran-
domly assigned as the safety committee.121 Safety is the sole
duty of this group, and there is a different safety committee
for each cycle. This procedure began in 1983 as a method of
incorporating safety into the unit operations course. We
regard safety as an integral part of the unit operations course,

and student involvement in the program prevents the need
for faculty/staff-mandated safety rules. Safety procedures
for the unit operations laboratory course are reviewed and
approved by the unit operations students, as well as by
faculty/staff. This paper focuses on further integration of
safety, including using Job Safety Assessments and the
Internet, into the existing unit operations course.

Most of our chemical engineering faculty are involved in
the unit operations course. I was assigned as the faculty
member in charge of the safety committee for each labora-
tory cycle. Typically, the safety committee had been respon-
sible for accident prevention and safety education, including
Conducting safety audits of the unit operations labo-
ratory before, during, and after the laboratory pe-
Distributing "Prevent Accidents with Safety"
(PAWS) forms
Ensuring that each group is familiar with the emer-
gency shutdown procedure for its experiment
Assisting other groups with safety-related matters
Conducting the safety meeting for that cycle
Conducting other safety-related objectives that
change from cycle to cycle
Basically, the committee is responsible for the safety of all
unit operations students in the laboratory and is assigned to
answer (or find the answer) to any safety-related questions

Copyright ChE Division ofASEE 1998

Chemical Engineering Education

Julia A. King is Assistant Professor of Chemi-
cal Engineering at Michigan Technological
University. She received a BS from Purdue
University (1982) and her MS (1987) and PhD
(1989) from the University of Wyoming. She
has industrial experience from employment at
Exxon Baytown Refinery and DuPont/Conoco.
Current research interests include polymer and
composite materials.

Laboratory ]

We regard safety as an integral part of the unit operations course. .. For each laboratory cycle, one
group of students is randomly assigned as the safety committee. Safety is the sole
duty of this group, and there is a different safety committee for each cycle.

that the students may have. The safety committee members
are assigned a grade based on their performance of these
duties. Due to MTU's strong safety culture and required
junior-level safety course, the safety committee generally
performs its assigned tasks well.
Based on my industrial experience, I led the development
of the following enhancements to MTU's already strong
safety involvement in the unit operations course:
Creation of a unit operations laboratory Internet homepage
with an emphasis on safety
Development of a "Safety Inspection Checklist" for the
unit operations laboratory
Revision of the PAWS form to include safety suggestions
Creation of a PAWS Tracking System that is available on
the Internet
Creation of Job Safety Assessment (JSA) forms for each of
the thirteen traditional experiments.

Everyone is encouraged to visit the Unit Operations Labo-
ratory homepage, located at
From this page, it is possible to obtain Material Data Safety
Sheets (MSDS) via a link to
or to obtain more safety information specific to the unit
operations course (Safety Inspection Checklist, PAWS form,
PAWS Tracking System, and Job Safety Assessment form).
In addition there is a link to "Tips for Safe Lifting Practices"
and to "Hygienic Practices to Keep in Mind." These materi-
als have been presented by past safety committees and the
class decided they should be placed on the Internet as a
reminder to help prevent accidents and to promote good
hygiene practices.

In the past, the safety committee was responsible for con-
ducting safety audits of the unit operations laboratory. But a
formal checklist had not been developed to conduct and
record the findings of the audit. Table 1 describes how to
complete the SIC, and Table 2 (next page) shows the items
listed on the checklist. The actual checklist is designed with
a check box for each item for each day of the month.
In order to construct the SIC form, the safety committees
studied similar forms from various sources.135" The form is
divided into two major sections: the first covers general

Summer 1998

laboratory items and the second covers safety items related
to each individual experiment. The safety committee uses
this form at the beginning, during, and at the end of each
laboratory period in which experiments are being conducted.
Typically, during a month there will be students in the labo-
ratory on six different days.
One SIC form is used for an entire month and is posted by
the unit operations laboratory office, and at the end of the
month it is filed in the office. If a safety problem is noticed
while they are conducting the inspection, the committee
attempts to take immediate action to remedy the problem.

Safety Inspection Checklist (SIC)

The Safety Inspection Checklist was developed to provide the Safety
Committee, the faculty, and staff with an organized and effective means
for inspecting the Unit Operations Laboratory. It also provides a record
of what has been inspected in the past. The inspection is done to ensure
that the lab operates in the safest manner possible, that all equipment in
the UO Lab is in proper working condition, and to detect and correct
any deficiencies in the lab.

How to Complete the SIC
The SIC is divided into two categories: Unit Operations Area and
Individual Experiments. The Unit Operations Area of the check list
covers items that are not assigned to an individual experiment, such
as fire extinguishers, safety showers, ladders, etc, while the
Individual Experiments section involves equipment and hazards
associated with one particular experiment and includes such things
as guards over pumps, gloves worn for changing a die, etc.

The Safety Committee will fill out the SIC at the beginning and end
of the UO Lab day. The checklist is a month-long document with the
days of the month indicated at the top. The month and year that the
inspection was performed must be filled out if that item is blank on
the document. A box associated with each item on the checklist is
split in half so that each item can be checked twice--once at the
beginning of the day and again at the end of the day. After
completing the inspection, the inspector must initial the bottom of
the document under the proper day. Once the two inspections have
been completed, the SIC is then posted on the chalk board for future
inspections. When the month has been completed, the document
should be turned over to Tim Gasperich in the UO Lab office for
storage and a new SIC can be obtained on disk from Dr. King.

If any problems arise while performing the inspection, the group
must take immediate action to remedy them. If additional assistance
is required, the group must inform the appropriate personnel. If there
is any equipment missing, a PAWS form should be filled out, and if
a safety hazard is noted during the inspection, a PAWS form should
be completed and the safety hazard should be corrected as soon as

Go to SICform.


Safety Inspection Checklist

Emergency Procedures
Evacuation routes posted
All exits and fire doors clearly marked and unobstructed
Telephones accessible and labeled with emergency numbers
Eyewash and safety showers clearly marked and unobstructed
Eyewash inspection up to date with tag
Safety shower inspection up to date with tag
Water continued to flow when handle was released
Eyewash flushed out
Safety shower flushed out
First Aid
Adequate supplies stocked
Clearly marked and unobstructed
Fire Extinguishers
Clearly marked and unobstructed
Inspection up to date with tag
Correct extinguisher for hazards present
Personal Protective Equipment
Ankle-high boots worn with proper material
Long pants worn: No loose clothing, hair, or jewelry
Appropriate eye wear and properly marked
Safety goggles worn when handling hazardous chemicals
Hard hat worn at all times; Earplugs worn in designated areas
Appropriate gloves worn and available
Dust masks and respirators in UO lab office
Left-hand rule used
Power off to make electrical connections
Extension cords away from traffic and water
3-pronged plugs on cords with ground
Cords without frays or splices
Make sure the overhead crane is locked and tagged
Stored in the proper cabinet; Storage cabinets labeled clearly
Clearly and properly labeled
Transported properly
Counters and floors clean and uncluttered
Ladders in good condition and chained when not in use
Cylinders labeled, upright, and secure
Waste containers provided and labeled
Make sure drain plugs are present
Drain is accessible
Batch filtration
Agitator locked out when adding slurry to the tank
Mercury manometer is used
All valves in proper position when flushing out lines
Continuous Filtration
All guards are securely in place
Use left-hand rule to turn on power
Power supply locked out when manually stirring tank
Continuous Stirred Tank Reactors
Cart wheels are kept blocked
Agitator immersed in solution when starting and stopping
Valves closed when starting reaction
Cooling Tower
Guard on blower securely in place
Ear plugs worn for noise level
Liquid water flowing to heater before starting steam flow

Steam valve closed before liquid water valve
Flow Measurement
Mercury manometer is used
Pump guards properly secured
Vent and drain lines before changing orifice
Dust mask worn when screening sand
Ear plugs worn when using Ro-Tap
Open valve on air supply slowly
Heat Transfer
Insulated gloves worn when operating steam valves
Opened steam valves slowly
Electric timers are away from water
Stayed clear of steam traps and heater
Liquid-Liquid Extraction
Protective gloves, apron, and goggles worn to handle acetic acid
Kerosene samples poured back into kerosene barrel after use
Kerosene pump guard in place
Sampling valves closed after samples taken
PDMS Bench-Scale Reactor
Goggles, rubber apron, and rubber gloves used to handle KOH
Red safety can used to weigh and transfer endblock A, 245 fluid
Pipes and hoses are in good condition and connections are tight
System inerted with nitrogen at all times
Experiment run in the hood
Main gas cylinder valve fully open
Glassware and thermometer transferred in proper container
Solvent Recovery
Gloves to be worn during sampling
Take precautions during sampling (i.e., dripping after sampling)
Check position of valves
PDMS Jacketed Reactor
Goggles, rubber apron, and rubber gloves used to handle KOH
Red safety can used to weigh and transfer endblock A, 245 fluid
Pipes and hoses are in good condition and connections are tight
System inerted with nitrogen at all times
Main gas cylinder valve fully open
Glassware and thermometer transferred in proper container
Face shield worn when adding chemicals to reactor
Polymer Flow
Guards in place and properly installed
Capillary viscometer pressure no greater than 20 psig
Tanks not overfilled
Polymer Processing
Insulated gloves worn when changing dies;
Gloves to be worn during sampling
Dies fastened correctly
Heating wires properly attached
Proper tools used for scraping polymer
Pumping A
Pump guards are in place and properly secured
Shut down procedure executed properly
Pumping B
Pump guards are in place and properly secured
Shut down procedure executed properly
Distillation-Solvent Recovery
System inerted with nitrogen at all times
Vacuum Drying
Vacuum pump guard properly secured
Insulated gloves used to operate steam valves
Dust mask worn when screening sand
Ear plugs worn when using Ro-Tap

Chemical Engineering Education

1 w m

Chemical Engineering Education

__ :Laboratory 1

When it is not possible to remedy the problem immedi-
ately, the committee fills out a PAWS form so there will
be a record.

The Prevent Accidents with Safety (PAWS) program, ini-
tiated at MTU in 1989, is designed to actively and positively
involve the students in the safety program. The unit opera-
tions students are responsible for their own safety as well as
the safety of others working in the area, and any student
observing an unsafe act is expected to correct the action
before an accident occurs. After correcting the unsafe act,
the student completes a PAWS form, which mentions the
group that corrected the unsafe act. PAWS points are awarded
(positive for safe acts, negative for unsafe acts), and the

Filling Out a PAWS Form

Part I (Situation Observed)
1. A situation is observed in the lab that needs to be reported. If you
are making a safety suggestion, skip to Part 2.
2. Obtain a PAWS form from the safety committee, Unit Operations
Lab Homepage, or the folder located in the Unit Operations Lab
3. Check the line on the PAWS form that describes the situation
observed. If none of the lines apply, check "Other."
4. Make additional comments about the situation in the "Comments"
section of the PAWS form. Be sure to include what happened,
where it happened, how it happened, and if reporting equipment
problems, tell what equipment needs to be fixed. Do not include the
name of the person who violated safety procedures; this is not what
PAWS forms are for.
Part 2 (Action Taken/Safety Suggestion)
1. Report what action, if any, was taken to remedy the safety situation
or to record safety suggestions in the "Action Taken/Safety
Suggestion" section of the PAWS form. Make sure to record which
experiment the suggestion applies to.
2. Sign and date the PAWS form. Include your group number on the
3. Place the PAWS form in the box labeled "PAWS Forms" located in
the Unit Operations Lab office.
Safety Committee Duties
1. Make sure that there are enough PAWS forms for the lab day.
2. Supply each lab group with PAWS forms at the beginning of the
lab day. Have more ready in case they are needed.
3. Collect PAWS forms from the box at the end of the lab day.
4. Review PAWS forms.
5. Input PAWS forms into PAWS tracking system.
6. Once the lab cycle is completed and the PAWS forms have been
entered into the PAWS Tracking system, give the forms to Dr.
Go to PAWS form.

Summer 1998

group with the most PAWS points each quarter is rewarded
with a dinner hosted by the developer of the PAWS pro-
gram, Dr. A. J. Pintar. Submissions of a "nitpicking" nature
are discouraged by not assigning them any points. In a
typical cycle, approximately twenty PAWS forms are sub-
mitted. Most of them concern personal protective equip-
ment, improper handling of chemicals, and equipment prob-
lems. Most PAWS forms are submitted by groups other than
the safety committee.
My industrial experience has taught me that people often
have good ideas for improving safety, so the PAWS form
was modified to encourage students to develop ways for
improving safety in the lab as well as to report unsafe acts. In
addition, students developed a checklist for common safety
concerns. Table 3 describes how to fill out the PAWS form,
and Table 4 displays a blank PAWS form.
I also encouraged the safety committees to develop a
PAWS Tracking System that could be used to track the
forms that were submitted. This system was placed on the
Internet so students could view the action that had been
taken on their item, could prepare for their next experiment
by studying past safety mistakes or concerns observed by


Prevent Accidents With Safety
Unsafe Situation Report Form

Situation Observed
Improper protection equipment (hard hat, safety glasses, boots)
SDidn't use goggles while transporting chemicals
One person moving ladder over 6 feet tall
Transporting glassware without bucket
Equipment left unattended
Equipment problem

Comments (where, how, what, experiment name)

Action Taken /Safety Suggestion

Lab Group Date

[ Laboratory

previous groups, etc.
Table 5 shows the PAWS Tracking System items submit-
ted (first page only) for cycle 3 of the 1996-97 academic
year. These items include unsafe acts, safety suggestions,
equipment problems, etc., related to each experiment. The
form also lists who is responsible for remedying the problem
and if the item is open or closed ("open" meaning action is
still needed and "closed" meaning action has been taken and
the item resolved). Open items are discussed with the stu-
dents, faculty, and staff at the required safety meeting for each
laboratory cycle. Each safety committee is responsible for
making progress toward "closing" the remaining "open" items.

JSA forms are often used in industry and are typically
completed for each "job" conducted in an industrial labora-
tory. They list each step in the procedure, the potential
hazards of each step, the recommended safe procedure to use
for each step, and the required personal-protection equip-
ment. Using them as a model, I encouraged the safety com-
mittees to develop a JSA form to use for our thirteen tradi-
tional unit operations experiments. Table 6 describes how
to complete the form and Table 7 shows a blank JSA
form, which also includes a safety awareness section
(nearest fire extinguisher, eye wash, etc.) and the emer-
gency shutdown procedure.

In the 1996-97 academic year, the safety committees and
the groups conducting the experiment believed that the JSA
form was useful to complete, could prevent accidents, and
should be implemented as a required procedure. Thus, in the
1997-98 academic year, each unit operations group conduct-
ing one of the thirteen traditional unit operations experi-
ments had to independently complete a JSA form for their
experiment prior to running the equipment. The students do
not have access to the JSA forms completed for the experi-
ments in the previous year, but the faculty advisor for each
experiment uses a completed JSA form as a guide for evalu-
ating the group's JSA form.

Student involvement in safety procedures prevents the
need for faculty/staff-mandated safety rules. The students
enjoyed working on these safety enhancements to the unit
operations course. They were enthusiastic about the projects,
they produced high-quality products, safety in our unit
operations course has been improved, and our students
are better prepared to work in industry as a result of these

I would like to thank the entire 1996-97 unit operations
class for their participation in the project. I would also like to

PAWS Tracking System

Cycle 3

Item Group Experiment Date PAWS

Unsafe Act/Situation/

Person Open/

# # Name form filled out Suggestion Action Taken Responsible Closed Comments

69 1A Safety 10/1/96 Open electrical conduit Tim Gasperich was TPG Closed
connection on electrical informed and he fixed the
box for heating oven. electrical box.
70 8B Heat Transfer 10/3/96 Leaky drum for heat Labeled and set aside TPG Closed
exchanger red gate valve drum, cleaned up floor.
on drum rusted on Tim Gasperich was
bottom near valve informed and he
created very wet floor, fixed the drum.
71 Dr. Safety 12/3/96 Ladders left Put chain around AJP Closed
Pintar unchained overnight, ladders.
72 11A Liquid-Liquid 12/5/96 Found broken, dirty Picked up and put in Closed
Extraction glassware on floor by broken-glass
liquid-liquid container.

73 11A Liquid-Liquid


Worn sticker on
liquid-liquid extractor.
H.F.& R. of
raffinate stream

Sticker replaced and
numbers filled in
according to
kerosene specs.
(Traced pipes to
kerosene tanks.)

TPG Closed

82 Chemical Engineering Education

Laboratory I

How to Fill Out a JSA Form

Job Safety Assessments (JSA)

Introduction to Job Safe)t
The Job Safety Assessment (JSA) Form provides
a faster way of identifying potential hazards in the
Unit Operations Laboratory. It acts as a useful
reference displaying possible hazards that can oc-
cur in each step in an experiment.

Format of Finalized JSA
The first section is used for identifying each ex-
periment, including the name of the process. Space
is provided for the person responsible for filling
out the original JSA form. Experimental proce-
dure for a process may change and this could
cause a change in the safety hazards associated
with that experiment, making revisions necessary.
There is a space provided for the person making
these revisions, as well as a revision number, revi-
sion date, and revision approval by the appropri-
ate faculty. Either Dr. Ellis or David Caspary are
the designated persons to approve the revised JSA.
Also, the first section contains a portion that shows
the necessary personal protective equipment re-
quired for the lab. In most experiments, all equip-
ment listed on this form (hard hat, safety glasses,
ankle-high books, long pants) are required, but
there are exceptions for some experiments.

The second section is divided into four columns.
The first column contains a detailed sequence of
the steps in an experiment, including startup, run
time, and shutdown. Potential hazards that can
occur at each step of the procedure are listed in the
second column. The recommended procedure to
prevent these hazards is listed in the third column,
and personal protective equipment required for a
step is listed in the fourth column.

The last section contains the location of safety
devices nearest the experiment as well as an emer-
gency shutdown procedure. The safety devices
include the nearest fire extinguisher, emergency
eye wash and shower, emergency exit, first-aid
kit, drain, and telephone.

Go to JSA Form

Blank Job Safety Assessment Form

Job Safety Assessment Form Unit Operations Laboratory
Department of Chemical Engineering
Michigan Technological University
Process Name: Hazard Level (high Written by:
medium, low)
Revised by: Revision #:

Process Location Revision date: Revision approved by:

Required Items for Lab:
Hard Hat Safety Glasses Ankle-High Boots Long Pants

Sequence of Steps Potential Hazards Recommended Safety Required Personal
Procedure Protective Equipment

Run Time


Safety Awareness:
Nearest Fire Extinguisher
Nearest Eye Wash & Shower
Nearest Emergency Exit
Nearest First-Aid Kit
Nearest Drain Plug
Nearest Telephone
Emergency Shutdown Procedure

thank Dr. Anton J. Pintar, Dr. Daniel A. Crowl, Dr. Thomas
G. Ellis, Mr. David W. Caspary, and Mr. Tim P. Gasperich
for their advice.

1. Pintar, A.J., D.W. Hubbard, and D.A. Crowl, "Teaching
Process Safety to Undergraduate Chemical Engineering Stu-
dents," AIChE 1993 Annual Meeting, Symposium on Inte-
gration of Safety and Process Design in Chemical Engineer-
ing Education, St. Louis, Missouri, November 9, 1993
2. Pintar, A.J., "Safety Manual for Use in the Chemical Engi-
Sumnmer 1998

neering Unit Operations Laboratory," Department of Chemi-
cal Engineering, Michigan Technological University,
Houghton, MI 49931, August 1996
3. Laboratory Safety Evaluation Checklist, University of Texas
at Austin, Office of Environmental Health and Safety, Haz-
ardous Materials Division (1996)
4. Maertens, J., Safety in Chemical Engineering Research and
Development, VCH Publishers Inc., New York, NY (1991)
5. Guidelines for Incorporating Safety and Health into Engi-
neering Curricula: Vol. 1. Laboratory Safety, Joint Council
for Health, Safety, and Environmental Education of Profes-
sionals, Savoy, Illinois (1994) J




King Saud University Riyadh 11421, Saudi Arabia

he laboratory is the proper place for students to apply
the theories and principles of chemical processes and
unit operations that they have learned in the class-
room. It can also be used to simulate industrial practices
where the ability to operate plants, to perform original de-
signs, and to modify existing processes is required. Operat-
ing a chemical plant requires that the chemical engineer be
proficient in problem solving and troubleshooting, and over
the years considerable efforts have been exerted to strengthen
the links between industry and academia in order to attain
that proficiency. Examples include laboratory experiments,
practice schools,m' research projects, summer internships in
industry,'2' co-op projects,'3 and others.
This paper will review the current goals of a laboratory
course and present the author's experiences in using labora-
tory time to cover several important topics related to both
industry and academia. The subjects that will be discussed
are safety procedures, startup and shutdown, troubleshoot-
ing, calibration with statistical applications, maintenance,
and mathematical modeling and simulation. Different avail-
able experiments will be used to achieve these goals.

Traditionally, a laboratory course emphasizes, through prac-
tical sessions, the understanding and application of theories
and principles taught in the classroom and presented in text-
books. During the course, the students become familiar with
the available process equipment, with instrumental analysis,
and with various measurement techniques. They also prac-
tice report writing, deal with experimental errors, and learn

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

S Copyright ChE Division ofASEE 1998

Copyright ChE Division of ASEE 1998

to recognize the discrepancy between theory and practice.
A new look141 with new objectives in the chemical engi-
neering laboratory has been suggested, with some of the
following objectives:
U Using the laboratory to develop engineering aware-
ness.11 Engineering awareness is developed through
several applications. Students gain practical experi-
ence, acquire skills, and get an idea of the technical
difficulties encountered in the design and construc-
tion of processing units.
E Introducing statistical concepts in the experiments.'61
Statistical ideas are incorporated into existing
experiments, which are modified through the analysis
of data, to introduce certain statistical concepts.
Mathematical modeling and simulation.17 10 A step-
by-step method is followed to develop steady-state
and dynamical models representing experimental
systems. Students are asked to perform analytical and
numerical solutions, using available simulation
B Troubleshooting.1'" Troubleshooting experiments are
described to develop the student's ability to diagnose
and correct unacceptable process performance.
E Simulation of industrial work.'21 The procedures used
in a typical unit operations lab course are modified to
simulate industrial practice. Applied problems and
instructions are included.
Performing economic evaluations."131 Estimations of
capital and operating costs are performed on typical
experiments. Scale-up and economic optimization of
an existing plant are considered.
While some of these goals pertain to industrial practice,
others reinforce mathematical, statistical, economical, and
process concepts. New and/or modified experiments are de-
veloped as necessary to meet the goals, and other objectives
can be included, such as familiarizing the students with
safety regulations, maintenance, and calibration. It is not
difficult to achieve all of these goals. The following sections
describe our method for meeting the goals within the time

Chemical Engineering Education


allotted to a typical lab course.

Students usually underestimate the impor- reviews]
tance of safety measures in the lab and occa- goa
sionally argue about strict safety regulations, laborati
particularly when they feel the running ex- and pre
periment is safe. But experiments are not the author's e
only source of danger in the lab. For ex- using labd
ample, danger could come from a pipe carry- to co
ing hot water or steam to other parts of the topics
lab. To counter this, we emphasize the fol- both inc
lowing items in such a way that they become ac
part of the student's daily practice, both in The su
and out of the lab:
are safety
We familiarize the students with startup an
hazard symbols, terms, and abbrevia- trouble
tions. Symbols include those on caliber
personal protective equipment, statistical
dangerous materials, and workplace
labels. Terms such as threshold limit mainten
value (TLV), hazard rating (HR), math
lower and upper flammability limits mode.
(LFL and UFL), etc., are covered, and simu
the most common abbreviations, such
as ACGIH (American Conference of
Governmental Industrial Hygienists) and MSDS
(Material Safety Data Sheet) are defined. The concepts
of fire triangle and tetrahedron are fully explained.
We teach the students to always be ready for an
emergency. We point out the location of emergency
outlets, fire extinguishers, in-line showers, and
emergency eyewash stations.
We ask that students wear a lab coat (or apron) during
each lab session, and that other personal protective
items be used when necessary. For example, the
students should wear, as needed, safety eyewear
(splash goggles), a face shield, and gloves during
distillation and reaction experiments, etc. We ask
them to put on dust masks during a solid-handling
experiment and anti-noise ear muffs during noisy runs
such as a cooling tower tutor.
We also show the students a video tape on safety or refer
them to related books.114'151 We give them specific assign-
ments on safety topics and discuss the topics later in a
seminar session. In particular, we ask them to compare the
safety equipment that is available in the lab with the recom-
mendations in a standard reference. The students can also be
asked to prepare data sheets on the specific materials and
hazard symbols that will be involved in the experiment they
plan to do (references 14 and 16 are very useful in this regard).
Finally, a sheet of safety procedures specific to a certain
Summer 1998

the current
Is of a
ory course
sent[s] the
cperiences in
oratory time
ver ..
related to
iustry and
objects .
d shutdown,
rtion with
lance, and
ling and

on the process, the materials, and the type of equipment
being used. The following is a list of common rules used in
industry-based on my own experience as a co-op student in
a refinery and as a production engineer in other plants. The
rules should be emphasized during the lab session.
Never introduce a cold stream suddenly into hot
equipment (such as a heat exchanger). Similarly, a hot
stream should not be suddenly introduced into cold
Introduce the cold stream first, and stop the hot stream
Any change in operating conditions should be done
gradually and should not exceed the listed operating
conditions of the specific equipment.
Make sure that the equipment is gradually, com-
pletely, and safely drained and purged.
An effective teaching method is to listen to the students'
suggestions and ideas about startup and shutdown and then
to discuss with them the significance and implications of the
general rules above. This helps them to realize that a sudden
change in temperature might cause a thermal or a mechani-
cal stress and result in a bad effect on the equipment. For
example, high pressure surges can destroy pressure gauges
or other measuring devices or dislocate trays from a distilla-
tion column. They will also learn that rapid venting could
result in a cooling effect that might freeze the remaining

lab is distributed to the students. It contains
general procedures (disposal, proper use of
service equipment, gas cylinders, etc.), evacu-
ation instructions in case of fire or chemical
spillage, etc. It should also include symbols
and abbreviations of equipment and pipelines
in the lab.
The lab instructor might also include other
topics such as hazard-assessment techniques.
The most common is the hazard and operabil-
ity study (HAZOP), which is a systematic tech-
nique for identifying all plant or equipment
hazards and operability problems.[17]

In industry, startup (putting the process and
its related equipment into proper operation)
and shutdown (when the operation is com-
pletely stopped) are related to integrated equip-
ment and interdependent processes. The equip-
ment is often outside the plant and susceptible
to weather effects, whereas the lab usually
contains small-scale individual equipment with
relatively constant ambient conditions.
Startup and shutdown procedures depend


liquid, and that complete drainage and venting is neces-
sary with flammable materials to avoid the formation of
explosive mixtures.
Table 1 is a general guide of the steps that should be taken
for startup and shutdown. Since startup and shutdown are
transient in nature, they can be modeled mathematically and
their behavior can be theoretically predicted.

Troubleshooting is the ability to characterize or diagnose a
problem and to present corrective action to solve it. This
ability is essential for people involved with the operation of
chemical plants-in particular shift engineers. Abnormal
operation can be harmful to operators, to equipment, and to
product quality. Corrective action might be constrained by
the time and safety of both the employees and the equip-
ment, so a systematic approach should be followed in trouble-
shooting. This requires a good theoretical background in the
process, experience, and engineering sense. A systematic
approach requires the presence of enough and reliable (rela-
tive to operators and instruments) data and information about
the situation, recognizing the problem, choosing the correc-
tive action from the different alternatives, and later feed-
back. Knowledge-based systems1181 are currently used to im-
prove plant operations by using automated diagnosis.
Trouble in industry can be caused by[19] such things as
misoperation, false alarms, equipment or chemical failure,
inadequate equipment design, and process failure. Lab ex-
periments can be effectively used to familiarize students
with troubleshooting procedures by designing special ex-
periments for this purposep11I or by using available experi-
ments with problems that occur either naturally or that are
created intentionally.
At the conclusion of the course, students should be asked
to summarize their experience in the form of a list that other
students can benefit from, revise, and extend if possible. In
this way, they learn to keep records of all failure and break-
downs as is done in industry. There should be a thorough
discussion of the cause and effect of the problem and how to
solve it in light of the systematic approach mentioned above.
Students should perform the required modifications when
necessary. Table 2 gives some examples of typical prob-
lems, their cause, and possible corrective actions.

Maintenance is all important. It keeps plants running, pre-
vents troubles, and identifies the cause of inadequate perfor-
mance.1191 The availability of maintenance software can help
an operation to run smoothly.1201 When students locate a
trouble source, they should participate in the repair and
maintenance process. For example, they can dismantle (with
the help of technicians) a malfunctioning pump for general

checkup or repair; they can clean a dirty reactor or a packed
bed; they can check valves for leakage; and when it is
necessary to modify a design, they can do it themselves.
It is important to familiarize students with the important
and related terminology of maintenance. This includes
planned maintenance, corrective maintenance, routine main-
tenance, servicing maintenance, running maintenance, and
shutdown maintenance. Other suggested projects include
Cleaning equipment such as heat exchangers, cooling
towers, coolers in distillation columns, etc. This
requires understanding of fouling and scales and the
methods of cleaning. The severity of deposits can
determine the method to be used: in running mainte-
nance, chemical treatment is applied without the need
to dismantle the equipment; for heavier deposits,
fluids under high pressure and/or temperature (such as
steam) are used. When these methods fail, the equip-
ment must be completely dismantled to remove the
adhering deposits. This is known as shutdown
Water treatment. This can be accomplished by
preventing the formation of deposits by using
antifoulants, preventing the reactions that cause
deposits by using inhibitors, and using dispersants to
prevent coagulation of suspended solids.
Noisy equipment. Fans and other equipment such as
mixers should be frequently inspected since friction is
a possible source of ignition.

Steps for Startup and Shutdown

Discuss with the students the objectives of the experiment, the
measurement techniques and instruments, and the materials and the
form of energy to be used. Relate the last with the safety section
regarding precautions and handling.
I Ask the students to familiarize themselves with the system, to
locate and check all the relevant equipment and valves (all types,
including relief valves), and draw a simple but detailed flowchart,
with full identification of all relevant parts. On a separate sheet, list
all input and output valves and designate them as "open" or
"closed" for startup.
With the help of the above list, check which valves to open and
which to keep closed. Refer to the guidelines given before,
particularly if heating/cooling is involved. Make sure that the
measuring equipment is ready. While other steps might be specific,
when dealing with a reacting system you generally introduce both
reactants simultaneously, or add one and then the other, or fill the
tank with one reactant and then add the other stream.
D Follow up the experiment, take the measurements, and keep
watching the input and output streams (through flowmeters), the
heat source, and the workability of other equipment.
0 Prepare yourself for shutdown. Again, refer to the valve list
generated at the beginning of the preparation and to the general
guide list.

Chemical Engineering Education

_ _Laboratory ]

Replacing steel pipes in old equipment with plastic

The objective of calibration is to prepare calibration curves
and to apply useful statistical techniques. Examples include
constructing calibration curves of conductivity against con-
centration and recording liquid flow rate with a rotameter. In
the case of conductivity, standard solutions are prepared and
their conductivity measured by a suitable conductivity meter.
The concentration range should cover the actual expected
measurements. With a rotameter, an adjustable flow from a
suitable pump is passed through the meter and its reading is
recorded; then the effluent is measured at certain time inter-
vals. A smooth curve can thus be generated.
Statistical analysis can reveal the presence of errors in the
measurements and the propagation of these errors later in the
calculation of an unknown sample concentration.1211 I recom-
mend that the students perform the following calculations:
Use linear regression to get the best fit. Calculate the
slope and intercept of the calibration line and the
percentage fit, which is 100 (correlation coefficient)2.
Calculate the residual standard deviation, which is
equal to the square root of residual sum of squares
divided by residual degrees of freedom.
Decide whether there is a fixed bias or a relative bias,
or both or neither. Fixed bias results in an intercept that
is not equal to zero, and a relative bias results in a
slope that is not equal to unity.
Quantify the precision of the prediction of the true
concentration of unknown samples.
Several calibration lines can be collected from differ-

ent groups, and a mean value can be calculated. A t-
test can be used by each group to compare its value
with this mean. Finally, a repeatability test can be used
to check the precision of the test method. Groups will
test similar unknown samples and quantify the
repeatability as r=t 2s, where r is the repeatability
measure, s is the standard deviation of the readings,
and t is the t-test at a certain confidence limit.

Statistical analysis could also be applied to the experi-
ments themselves, e.g., determining the order of the reaction
and the reaction rate constant from a batch reactor run.
These values are then used in the continuous reactor calcula-
tions. It is appropriate to analyze propagation of errors
throughout these calculations and to do some other signifi-
cant tests on the value of the reaction rate constant.
Other suggested'15 calibration projects are a mercury-in-glass
thermometer, a thermistor, and type-T and -K thermocouples.

Mathematical modeling is the process of describing and
approximating actual physical systems using mathematical
tools. A real process is mathematically abstracted for pur-
poses of understanding and predicting its behavior. Reduc-
ing the experimental effort required to design or optimize
the process is another motivation for developing a math-
ematical model. The model can be checked against experi-
mental data and then reconsidered in order to be more effec-
tive and more useful in achieving the required objectives.
The first step in model formulation[22'231 usually involves
drawing a picture of the system under investigation and
selecting the important dependent (responding) and inde-
pendent (changing) variables, along with the parameters that

Typical Problems, Causes, Remedies

Possible Cause (Remedy)

Cooling tower Wet-bulb temperature equal to or greater than
dry-bulb temperature of outlet air

Temperature gradient of water at the bottom
section is very small

Distillation Conductivity readings from different trays are
not consistent


Fluctuated flow, with air bubbles in flow lines

* False reading (Check measuring devices)
* Wick is not wetted (Check water and wick)
* Air is blocked (Check air flow)
* False reading (Check measuring devices)
* Lack of water (Check water flow)
* Air is blocked (Check air flow)

* False readings (Select the correct range at the calibrated temperature)
* Incorrect sampling (Cover drawn samples with aluminum foil, cool them to
calibration temperature using a water bath, and make sure probe is well
immersed in the sample.)

* Direct pumping within short distance (Use head tanks. If bubbles are still
present, increase the tank height.)
* Low level in feed/head tanks (Increase the level of solution in the tanks.)
* Back pressure (Check for any resistance to flow.)

Summer 1998 18

Experiment Problem

I Laboratory 1

are expected to be important (physical constants, physical
size, and shape). The second step is bringing together all
applicable physical and chemical information, conservation
laws, and rate expressions. The third step requires setting down
of finite or differential volume elements, followed by writing
the conservation laws. Then an appropriate mathematical solu-
tion method is sought with the proper choice of the boundary
value of the dependent variables, which finally relates depen-
dent variables to one or more independent variables.
Mathematical modeling seems to be a difficult subject to
many students as well as to people working in industry. Deal-
ing with differential equations in industry is sometimes simply
avoided and the steady-state simulators are "tricked" into do-
ing the work.[241 But, modeling can be made simple and inter-
esting when doing lab experiments, particularly when a team
effort is practiced. In industry the team might include special-
ists such as chemists and statisticians.125 In the lab, students
should perform the tasks with the help of the instructor.
I use a CSTR setup17-9] to study startup and shutdown
processes. It can be versatile. Students can study second-
order or first-order reactions under isothermal or
nonisothermal conditions; they can model the dynamics and
steady-state behavior of the system; they might consider
different approaches to startup and then model them. Other
experiments are used to build simple models. For example,
in a cross-flow heat exchanger, a model is built to describe
the transient heat-transfer process between a heated element
of copper and air.
The developed models could be solved analytically and/or
numerically; using both approaches is preferable. I encour-
age the students to use available packages in order to empha-
size principles rather than programming. For example, they
write a main program and call available subroutines from
IMSL, or use MATLAB. Sometimes I ask them to use
spreadsheets, which are preferred by industrial people and
can be used in a solid-handling experiment to simulate the
breakage of a known sample of a solid material.[261 I ask the
students to solve problems related to the subject of the ex-
periment from their textbooks in an attempt to form a link
between lectures and lab work. For example, I ask them to
solve problems on transient balances related to crushing and
grinding, and to build the model.[27]
Data acquisition software can serve as a convenient tool
for quickly developing computer simulations of chemical
engineering unit operations for use in classroom demonstra-
tions.[281 These packages can be used to create virtual unit
operations. Simulation is used to avoid some of the disad-
vantages associated with certain experiments.129" This is use-
ful, for example, when a great amount of time is needed to
perform the experiments or when a complex phenomenon
such as ion exchange or adsorption is considered.

The objective of a lab course is no longer a matter of data
collection and the preparation of a full and lengthy report.
Many useful things can be extracted from a lab session when
the new objectives are properly invested. The list might look
lengthy, but this is not a problem as long as the objectives
are achieved within the time limit without affecting the
scheduled experiments. Safety procedures require one ses-
sion and should be stressed in every other session. Mainte-
nance and troubleshooting are performed using available
experiments. Startup and shutdown are related to every ex-
perimental run. Mathematical modeling and simulation should
be done in connection with the specified experiment.
Several important points can be drawn from this discus-
U Industrial work can be effectively simulated in the unit
operations lab without affecting the academic approach.
This is achieved by stressing subjects such as startup,
safety, troubleshooting, report writing, statistical
analysis of errors, and modeling.
U The ability to solve problems and troubleshoot is
developed by following systematic procedures of safety
and of startup and shutdown, and by allowing the
students to tackle practical problems and search for
corrective solutions.
B Available experiments can be used to achieve the
required goals, and when necessary new experiments
can be introduced or existing ones can be modified.
Students can participate in all of these activities.
B Students realize that simple subjects look difficult when
they are not understood or practiced. This is obvious in
practicing startup and shutdown procedures. They
should be taught to think of applications in order to
understand and memorize more easily.
U A worker who understands hazard and safety precau-
tions improves work practice and becomes aware of
protection and handling procedures.
E The team effort should not be ignored. It can be
practiced by considering troubleshooting, modeling,
and analysis.
E Subjects are interrelated. For example, mathematical
modeling (in particular, the dynamic type) can help
troubleshooting that is usually transient in nature.
B The typical unit operations lab is a fruitful area where
many applied subjects can be practiced effectively.
The instructor should make sure that the students under-
stand and grasp the above topics. This could be achieved by
discussions during the lab sessions, by oral and written ex-
ams, and by asking the students to write short reports as part
of a final exam. An experience like the one we have pre-

Chemical Engineering Education


sented here will, hopefully, give the instructor courage to
teach such a course without hesitation.

Thanks to the editors and to the reviewers of this paper for
their helpful remarks.

1. Johnston, B.S., T.A. Meadowcroft, A.J. Franz, and T.A. Hatton,
"The MIT Practice School," Chem. Eng. Ed., 28(1), 38 (1994)
2. Riggs, J.B., and R.R. Rhinehart, "Introducing Graduate Stu-
dents to the Industrial Perspective," Chem. Eng. Ed., 31(2),
3. Chung, J., "Co-op Student Contribution to Chemical Process
Development at DuPont Merck," Chem. Eng. Ed., 31(1), 68
4. Douglas, J.M., "ChE Lab: A New Look," Chem. Eng. Ed., 9(1),
5. England, R., and R. Field, "Using the Laboratory to Develop
Engineering Awareness," Chem. Eng. Ed., 23(3), 144 (1989)
6. Burke, A., A. Phatak, B. Hudgins, and P. Reilly, "Introducing
Statistical Concepts in the Undergraduate Laboratory," Chem.
Eng. Ed., 27(2), 130 (1993)
7. Abu-Khalaf, A.M., "Mathematical Modeling of an Experi-
mental Reaction System," Chem. Eng. Ed., 28(1), 48 (1994)
8. Abu-Khalaf, A.M., "Dynamic and Steady-State Behavior of a
CSTR," Chem. Eng. Ed., 30(2), 132 (1996)
9. Abu-Khalaf, A.M., "Start-Up of a Non-Isothermal CSTR:
Mathematical Modeling," Chem. Eng. Ed., 31(4), 250 (1997)
10. Kofke, D., M.R. Grosso, S. Gollapudi, and C. Lund, "CESL in
the Chemical Engineering Simulation Laboratory," Chem.
Eng. Ed., 30(2), 114 (1996)
11. Myers, K.J., "Troubleshooting in the Unit Operations Labo-
ratory," Chem. Eng. Ed., 28(2), 120 (1994)
12. Marrero, T.R., and W.J. Burkett, "Introducing Industrial
Practice in the Unit Operations Lab," Chem. Eng. Ed., 29(2),
13. Langrish, T.A., and W. Davies, "Putting Commercial Rel-
evance into the Unit Operations Laboratory," Chem. Eng.
Ed., 28(1), 40 (1995)
14. Lewis, R., Rapid Guide to Hazardous Chemicals in the Work-
place, 3rd ed., Van Nostrand Reinhold (1994)
15. Stricoff, R., and D. Walters, Laboratory Health and Safety
Handbook, Wiley (1990)
16. Weiss, G. (editor), Hazardous Chemicals Data Book, Noyes
Data Corporation (1980)
17. Peters, M.S., and K.D. Timmerhaus, Plant Design and Eco-
nomics for Chemical Engineers, 4th ed., McGraw Hill (1990)
18. Ramanathan, P., S. Kannan, and J.F. Davis, "Use Knowl-
edge Based Systems Programming Toolkits to Improve
Troubleshooting," Chem. Eng. Prog., 89(6), 75 (1993)
19. Gans, M., D. Kohan, and B. Palmer, "Systemize Trouble-
shooting Techniques," Chem. Eng. Prog., 87(4), 25 (1991)
20. Kmetz, C.J., "An In-Plant Container-Labeling Program,"
Chem. Eng. Prog., 86(3), 33 (1990)
21 Caulcutt, R., and R. Boddy, Statistics for Analytical Chem-
ists, Chapman and Hall (1983)
22. Rice, R.G., and D.D. Do, Applied Mathematics and Modeling
for Chemical Engineers, John Wiley (1995)
23. Barton G.W., "Model Development and Validation: An Itera-
tive Approach," Chem. Eng. Ed., 26(2), 72 (1992)
24. Horwitz, B.A., "Avoid Nausea when Solving Dynamic Prob-
lems," Chem. Eng. Prog., 92(3), 44 (1996)
25. Semple, T.C., and J.H. Worstell, "Rethink Your Model for
Summer 1998

Improving Processes," Chem. Eng. Prog., 92(3), 56 (1996)
26. McCabe, W.L., J.C. Smith, and P. Harriott, Unit Operations
of Chemical Engineering, 4th ed., McGraw-Hill (1985)
27. Felder, R.M., and R.W. Rousseau, Elementary Principles of
Chemical Processes, 2nd ed., John Wiley (1986)
28. Davis, R.A. "Create Virtual Unit Operations with Your Data
Acquisition Software," Chem. Eng. Ed., 29(4), 270 (1995)
29. Anklam, M.R., R.K. Prudhomme, and B.A. Finlayson, "Ion
Exchange Chromatography Laboratory: Experimentation and
Numerical Modeling," Chem. Eng. Ed., 31(1), 26 (1997) J

V m book review

Mathematical Methods
in Chemical Engineering
by A. Varma and M. Morbidelli
Oxford University Press, New York, NY; 690 pgs; $80 (1997)

Reviewed by
E. James Davis
University of Washington

This text follows the Minnesota tradition of applied mathematics
in chemical engineering established by Professors Amundson and
Aris in that the treatment of the numerous topics is rigorous and
vigorous. There is an attempt to complement the mathematical
fundamentals with examples arising in chemical engineering, but
the balance between theory and application is not uniform through-
out the book. The authors naturally lean toward examples from
their own research and experience.
The nine chapters cover a wide range of subject matter starting
with matrix theory and proceeding to a particularly long chapter
dealing with first-order linear ordinary differential equations and
stability theory. With respect to Chapter 1, the authors acknowl-
edge their debt to Professor Amundson; that chapter summarizes
Amundson's book Mathematical Methods in Chemical Engineer-
ing: Matrices and Their Application. The first two chapters account
for almost one-third of the 690-page book, but Chapter 2 addresses
subjects such as Liapunov's direct method and the Hopf bifurcation
theorem not covered in typical texts on advanced engineering math-
ematics. The 135 pages of Chapter 2 include interesting applica-
tions such as the analysis of the Belousov-Zhabotinskii oscillatory
Chapters 3 and 4 are more conventional in their coverage of
linear ordinary differential equations and special functions, respec-
tively. A clear presentation of the Green's function for solving
nonhomogeneous equations is a plus. The applications included in
these chapters are rather lean, and the presentation of orthogonal
polynomials such as the Chebyshev and Laguerre polynomials and
other special functions is left to problems at the end of Chapter 4
without relevant applications.
The classification of partial differential equations in Chapter 5
Continued on page 219

[ Laboratory



Loughborough University Loughborough, Leicestershire, England LE11 3TU

Specific heat is one of those properties that all under-
graduate students of chemical engineering would claim
at least some measure of familiarity with, often from
close and bloody encounters in the performance of energy
balances. Our experience has been that this familiarity rarely
extends to methods of determining this important intensive
property. Many of the sophisticated methods that have been
developed over the last century or so are inappropriate to the
repertoire of the first-year laboratory, either because they
require elaborate and expensive apparatus and/or because
they are time consuming and not easily scheduled into what
appears to be that universal period of time (3 hours) devoted
to laboratory classes. The interested reader should consult
any of the texts on classical calorimetry[e' 1.21 for a full
account of the available methods.
In this paper we describe a simple and inexpensive under-
graduate experiment for the determination of the specific
heat of solids based on the evaporation calorimeter devised
by Dewar in 1904.[3] An added advantage of the experiment
is that it introduces some important thermodynamic prin-
ciples in the development of the theory underlying the evalu-
ation of specific heat. To deflect the charge that we may be

Klaus Hellgardt has been a lecturer in the De-
partment of Chemical Engineering at
Loughborough University since 1995. His main
research interests lie in the field of heteroge-
neous catalysis. He obtained his Dipl.-Ing (TH)
from the University of Karlsruhe. In 1991 he was
awarded an EC Research Training Fellowship
during which he completed his PhD and DIC at
Imperial College in London.

Gilbert Shama has been in the Department of
Chemical Engineering at Loughborough Uni-
versity since 1989 and is closely associated
with the operation of first-year undergraduate
laboratories. He obtained his BSc degree from
UMIST, his MSc from Birmingham, and his PhD
from Imperial College, London.

over-virtuous in our endeavors, we confine ourselves to a
method that will allow the determination of the average
specific heat of a solid in the range of 80K to 298K.

The terms "specific heat" and "heat capacity" are often
used synonymously. In fact, the term heat capacity is a
legacy of the "caloric" theory of heat that Count Rumford
essentially laid to rest at the end of the 18th Century. This
theory held that heat was a substance that could literally be
"poured" from one body to another. We now know that heat
cannot be contained within a system, but rather is a manifes-
tation of the interaction of a particular system with its sur-
roundings. The modern concept of specific heat is that it is a
measure of the energy changes of a system when energy, in
the form of heat, is transferred across its boundaries. Trans-
fers of energy can be conducted under a number of different
conditions (e.g., constant pressure or constant volume), each
of which can be used to define a specific heat; the concept is
further elaborated upon below.
The general change in enthalpy of a system that is in
mechanical equilibrium with its surroundings can be ob-
tained with
dH=dQ+dWe +VdP (1)
The term We is used to denote all forms of work that may be
referred to as "non-P,V work." i.e., electrical, gravitational,
etc. Now, if we assume that there is no non-P,V work done
on the system, and heating or cooling occurs at constant
pressure, this equation reduces to
dH= dQ (2)
Using the relationship
dQ=CdT (3)
we can then write

C=T (4)

Copyright ChE Division of ASEE 1998

Chemical Engineering Education


which is the definition of the heat capacity or specific heat
(at constant pressure). A similar deduction starting from the
description of the internal energy of the system U leads to

Cv = (5)

It can be further shown that the relationship between Cp and
Cv is

Cp-Cv=( VT (6)

Following on from Eq. (4), however, if, at constant pressure,
the temperature of a compound, S, is raised by a small
increment AT, a finite amount of heat needs to be trans-
ferred, and
Q = AH = nsCmAT (7)

Here, cm is the molar specific heat and ns is the number of
mols of compound S. The specific heat is usually a function
of the temperature and approaches zero for very low tem-
peratures (a few degrees Kelvin above absolute zero). In this
temperature range, the function Cp(T), or Cv(T) for that
matter, is sufficiently accurately described by the theories of
Einstein91] and Debye."lo At higher temperatures, the rule of
Dulong and Petitl'" applies. They observed as early as 1819
that the molar heat capacities of elementary solids have
values close to 25 J/mol K. The theoretical justification
of this law can be drawn from Boltzmann's equipartition
theorem. Based on the energy of motion of each oscillat-
ing atom in a solid, the internal energy of the system can
be described as

U=6NA 2kT =3 RT (8)

Differentiation then leads to the constant value 3 R

Cv = 24.9 J / mol K

Quantization of the vibrational energy of the atoms by
Einstein led to an expression for Cv,

Cy = 3 Rf2 (9)

that could describe the low tempera-
ture behavior of the heat capacity. The
application of Debye's continuum
theory to the vibrational modes of at-
oms in a solid predicts a cubic law
dependency for the specific heat of el-
emental solids and is a definite im-
provement over Einstein's work:

Cv = 12 4R ] T (10)
5 (TD)

Figure 1. Set-up of eva

This experiment was designed as a first-year
laboratory experiment for undergraduate
students in order to familiarize them with the
concepts of thermodynamics. Once it is set up,
results for a number of different elements can
be conveniently obtained and compared.

In our experiment, various solids are cooled from room
temperature, T,, to the temperature of liquid nitrogen, T,.
The average molar specific heat between T2 and T, is de-
noted here as C The enthalpy change for a solid immersed
into the liquid nitrogen can then be described as

( T
AH = H(T)H(T)-H(T,) = ns CdT = nsC(T2 -T1)
fp CP M (T2 TI)

This change results in the boiling off of nN, mols of nitro-
gen. If the molar heat of evaporation of liquid N2 is known
( AHN, ), then Cm simply follows from
-- nN,AH(12)
S-ns(T TI) (1

A simple evaporation calorimeter was fabricated from two
Dewar flasks and additional components mostly comprising
standard laboratory items. The entire calorimeter is housed
within a PVC container made from a 0.6-m section of 0.2-m
diameter PVC pipe, to which is fitted a base plate of 0.4-m
diameter. This provides adequate stability for the entire as-
sembly and minimizes the risk of it being knocked over
during the experiment. For additional security, we recom-
mend that the PVC container be fastened to the laboratory
bench with screws. The large, outer Dewar ("Dewar 2" in
Figure 1) has a capacity of 0.5 litre. This Dewar is placed
inside the PVC container, and insulating material is poured
into the gap between the inside wall of the container and the
Dewar. Expanded polystyrene chips of the sort often used in
packaging proved to be ideal for
this purpose. The smaller Dewar
("Dewar 1" in the figure) has a
ground-glass socket at its opening
and a capacity of 0.2 liters. The
cone closure was modified in our
SSample workshop to provide inlet and out-
let tubes as shown in the figure. A
Dewar 1 small glass flask of 50-ml capac-
Dewar 2 ity, into which samples are placed,
Insulation is connected to one branch of the
modified cone by clear flexible tub-
ing. A longer length of identical
portion calorimeter, tubing is attached to the other

Summer 1998


I Laboratory

branch and is directed to a "beehive shelf" in a large plastic
tray filled with water, which serves to collect the nitrogen
gas evaporated during operation. A 2-liter measuring cylin-
der is used for this purpose.
Operation of the calorimeter requires partially filling Dewar
2 with liquid nitrogen. Dewar 1 is then carefully lowered
into the large Dewar. (We recommend that a rubber bung be
placed at the bottom of Dewar 2 in order to prevent break-
age.) Liquid nitrogen is then added to Dewar 1 and care is
taken to ensure that the level of liquid nitrogen in Dewar 1 is
lower than that in Dewar 2. Approximately 1 g of solid
sample is charged into the small glass flask and connected
to the cone. This will result in evolution of nitrogen gas,
which must be recorded (using a stopwatch) in order to
provide a baseline. Once the baseline has been estab-
lished, the contents of the flask are steadily discharged
into the liquid nitrogen, and the volume of gaseous nitro-
gen evolved is recorded. Once a steady state has been
reestablished, the constant evolution of nitrogen is again
recorded (second baseline).

Prior to conducting the experiment, students are briefed on
the particular hazards associated with liquid nitrogen and are
given eye protectors and insulated gloves. We recommend
that students' handling of liquid nitrogen be minimized and
that "bulk" transfers from storage Dewars be carried out by
suitably qualified laboratory technicians.
Normally, it will not be necessary to replenish the volume
of liquid nitrogen during the course of the experiment once
the Dewars have been charged. We employed glass Dewar
vessels and encountered no difficulties with their use. (It
may be of interest to point out that metal Dewar flasks are
also available, and that anyone having particular concerns
about breakage may wish to consider using them instead.)
One less obvious hazard associated with the use of liquid
nitrogen is the condensation of liquid oxygen from the atmo-
sphere during the experiment. The volume contained in the
Dewars of the capacity quoted above are sufficiently small

to minimize this occurrence. As an additional precaution, we
fabricated a simple check valve from a truncated plastic
filter funnel and rubber "O" ring (see Figure 1) by which the
smaller of the two Dewars is suspended inside the outer one.

By extrapolating the volumetric flow rates before and after
the experiment, the actual gas volume AV evolved can be
determined (see Figure 2). The value of AV is the distance
between the two baselines at the point where the two areas,
Al and A2, shown in the figure are equal. Using the ideal
gas law, the number of mols of evaporated N, can be calcu-
lated. The partial pressure of nitrogen follows from

PAir = PN2 + PHO + Phydrostatic (13)
The atmospheric pressure in the laboratory has to be deter-
mined very accurately, and the partial pressure of water at
room temperature is taken from the appropriate tables. The
correction for the hydrostatic pressure is given by

Phydrostatic = PH20gh (14)
When the heat of evaporation of liquid nitrogen
(AHv = 5.56- )r12.'31 and the molecular weight of the sub-

Figure 2. N, evolution; determination of the
volume of evaporated gas.

Chemical Engineering Education

Comparison of Specific Heat Values for Gold, Tin, Silver, Zinc, and Silicon14-221

Experimental Values
Element [J/mol K] Data from Literature [J/mol K]
Gold 25.95 [77K 293K] 25.35 [290K]; 25.42 [298K]; 25.38 [298K]; 19.54 [80K]; 19.63 [80K]
Tin 25.55 [77K 293K] 26.98 [298K]; 25.32 [298K]; 20.71 [80K]
Silver 23.45 [77K 293K] 25.28 [290K]; 25.35 [298K]; 25.54 [298K]; 17.91 [80K]; 17.87 [80K]
Zinc 22.85 [77K 293K] 25.4 [298k]; 25.38 [298K]; 16.86 [80K]
Silicon 13.55 [77K- 293K] 19.99 [298K]; 19.74 [298K]; 5.28 [80K]


stance investigated are known, the heat capacity can be
calculated according to Eq. (12).

We have used the above apparatus to measure the specific
heats of gold, silver, silicon, tin, and zinc. The experimental
results that we determined are compared with data obtained
from various sources in Table 1.
It is clear from Table 1 that the average C, values found
from our experiments lie somewhere between the values
quoted in the literature for the distinct temperature limits of
the experiment. The specific heats of all elements consid-
ered (with the exception of silicon) seem to obey the rule of
Dulong and Petit and are close to 3R as mentioned above.
The somewhat lower value obtained for silicon is due to
the fact that the Cp value for Si approaches lower values at
much higher temperatures than is observed for the other
metals. This is presumably because the highest frequency
for the 3N normal modes of vibration for a silicon crystal
is larger than for the other elements due to the intrinsic
bond strength.l58-231

This experiment was designed as a first-year laboratory
experiment for undergraduate students in order to familiar-
ize them with the concepts of thermodynamics. Once it is set
up, results for a number of different elements can be conve-
niently obtained and compared. In order to further develop
the concepts introduced by this experiment, students could
be asked to compare their results with the rule of Dulong and
Petit as well as the predictions provided by the theories of
Einstein and Debye. The definition of degrees of freedom
and their contribution to the specific heats Cp and Cv of
gases could also be addressed. For especially keen students,
the process of achieving very low temperatures by adiabatic
demagnetization could be elaborated on.

a thermal expansivity (K-')
Cp specific heat at constant pressure (J mol 'K-')
Cv specific heat at constant volume (J mol' K ')

C' average molar specific heat (J mol-' K-')
F frequency factor
g gravitational acceleration (ms-2)
h height (m)
H enthalpy of a system (J mol-')
AHv heat of evaporation (J mol-')
K compressibility (bar')
k Boltzmann's constant (J K-')
n number of mols (mol)
NA Avogadro's constant
Summer 1998

P pressure (bar)
Q heat (J)
p density (kg m3)
R universal gas constant (J mol-' K ')
T temperature (K)
TD Debye temperature (K)
U internal energy of a system (J mol')
V volume (m3)
W non P,V work on system (J)

1. McCullough, J.P., and D.W. Scott, eds., Experimental Ther-
modynamics: Vol. 1. Calorimetry of Non-Reacting Systems,
Butterworths, London, England (1968)
2. Hemminger, W., and G. Hohne, Calorimetry: Fundamentals
and Practice, VCH Publishers, New York, NY (1984)
3. Dewar, J. Proc. Roy. Inst., 17, 581 (1904)
4. Atkins, P.W., Physical Chemistry, 3rd ed., Oxford Univer-
sity Press, Oxford (1986)
5. Denbigh, K.G., The Principles of Chemical Equilibrium,
2nd ed., Cambridge University Press, Cambridge (1966)
6. Toulikian, Y.S., and E.H. Boyco, Specific Heat of Metallic
Elements and Alloys, IFI/Plenum (1978)
7. Daubert. T.E., Chemical Engineering Thermodynamics,
McGraw-Hill Int. Ed., New York (1985)
8. Partington, J.R., in Advanced Treatise on Physical Chemis-
try, Vol. 3, Longmans, Green and Co., London, New York
9. Einstein, A., Ann. Phys., 22, 800 (1907)
10. Debye, P., Arch. Sci. Phys. Nat., 33, 256 (1912)
11. Dulong, P.L., Ann. Chim., 10, 395 (1819)
12. Yaws, C.L., Physical Properties, McGraw-Hill Publishing
Co., New York, NY, (1977)
13. Ullmann's Encyclopedia of Industrial Chemistry, 5th ed.,
Vol. A17, VCH Weinheim, 457 (1991)
14. Dewar, J. Proc. Roy. Soc., 89, 158 (1913)
15. Bolz, R.E., and G.L. Tuve, "Thermal Properties of Pure
Metals," CRC Handbook of Tables for Applied Engineering
Science, 2nd ed. (1973)
16. Kelly, K.K., Bureau of Mines Bulletin (1961); in CRCHand-
book of Chemistry and Physics, 67th ed. (1987)
17. Hultgren, "Selected Values of Thermodynamic Properties of
Metals and Alloys," in CRC Handbook of Chemistry and
Physics, 67th ed. (1987)
18. Wagman, D.D., W.H. Evans, V.B. Parker, R.H. Schumm,
S.M. Bailey, I. Halow, K.L. Churney, and R.L. Nuttall, "The
NBS Tables of Chemical Thermodynamic Properties," J. of
Phys. and Chem. Refer. Data, 11(2) (1982)
19. Wagman, D.D., W.H. Evans, V.B. Parker, S.M. Bailey, I.
Halow, and R.H. Schumm, "Selected Values of Chemical
Thermodynamic Properties," National Bureau of Standards
Technical Notes 270-3, 270-4, 270-5, 270-6, 270-7, and 270-
8, Institute for Basic Standards, Washington DC 20234
(1968, 1969)
20. National Bureau of Standards, Circular No. 500, Part 1
(1952); in CRC Handbook of Chemistry and Physics, 67th
ed. (1987)
21 Furukawa, G.T., W.G. Saba, and M.L. Reilly, "Thermody-
namic Functions of Copper, Silver, and Gold," NSRDS-NBS,
in CRC Handbook of Chemistry and Physics, 67th ed. (1987)
22. Corruccini, R.J., and J.J. Gniewek, National Bureau of Stan-
dards Monograph 21 (1960); in CRC Handbook of Chemistry
and Physics, 67th ed. (1987)
23. Stock, Z., angew Chem, 35, 342 (1922) O





University of North Dakota Grand Forks, ND 58202-7101

he value of peer review in developing both critical
thinking and student writing skills has been well
documented['14] and is best demonstrated by better
laboratory reports. First drafts are often improved because
students realize that their peers will be reading their writ-
ing.J5 Additionally, students are given an opportunity to
revise their original report in response to the reviews.
Reviewers benefit by being forced to consider the various
elements that result in an effective report. They must be
given some guidelines, however. The instructor should pro-
vide sufficient structure and guidance to prevent students
from giving entirely negative or hierarchal evaluations.151
Providing a structured report sheet for the students to
use, similar to a referee report, is advantageous for this
purpose (see Table 2).
At the University of North Dakota, peer review has been
incorporated into the undergraduate research lab as part of
an ongoing effort to develop the oral and written communi-
cation skills of our chemical engineering undergraduates.[6'71
A student is required to submit a technical journal "article,"
similar in scope to an extended abstract, based on his or her
lab experiment. A copy of the report is then given to some
student in the class who has not run the experiment. This
"reviewer" must learn the details of the experiment, evaluate
the technical report, make specific suggestions for revision,

James A. Newell became an Assistant Profes-
sor at the University of North Dakota in 1995.
He received his BS from Carnegie-Mellon Uni-
versity, his MS from Pennsylvania State Uni-
versity, and his PhD from Clemson University.
His research focuses on high-performance poly-
mers and composites. He was recently named
Sthe Dow Outstanding New Faculty Member by
the North Midwest section of ASEE.

At the University of North Dakota, peer
review has been incorporated into the
undergraduate research lab as part
of an ongoing effort to develop
the oral and written
communication skills of our
chemical engineering undergraduates.

and identify both the strengths and the weaknesses of the
report. While the peer review does not affect the grade of the
original journal author, the review itself is graded.
Finally, the original report writer receives the peer review
along with a faculty review and is given an opportunity to
return to the lab to gather any additional data that is
required before submitting a revised report in response to
the reviews. The revised report is graded separately from
the original report.
Each student writes one technical journal article, one peer
review, one operations manual, one oral presentation, and
one revised final report during the course. Tables 1 and 2
show the handouts given to students and the referee report
they are asked to use with their review.

Peer review was used for the first time in the undergradu-
ate laboratory in the fall semester of 1996, and a noticeable
increase in the quality of writing was immediately apparent.
The original technical journal reports were better than they
had been in previous years.
Although it is difficult to quantify this improvement, it

Copyright ChE Division of ASEE 1998

Chemical Engineering Education


Peer-Review Handout

Chemical Engineering Lab II
The Peer Review
Almost no journal articles are published in their original form. External readers often can offer new
insights and perspectives, detect areas of weak or faulty reasoning, and address ambiguous or incorrect
writing issues. Thus, when an editor of a technical journal receives a submission from a researcher, the editor
sends the article to another expert in the researcher's field. This peer reviewer identifies strengths and
weaknesses of a paper, locates inconsistencies in reasoning or argument, evaluates the technical merit of the
paper, and makes two specific recommendations that change the paper. First, the reviewer makes a general
assessment of the paper and recommends one of four courses of action. Specifically.
Publish the paper as is. (This seldom happens.)
Publish the paper after minor revisions are made.
Have the author make the suggested major revisions and I'll review it again.
Do not publish this paper.
Next, the reviewer makes a detailed series of recommendations for improving the paper. These may include,
but are not limited to, suggesting additional experiments, requesting additional explanation or analysis,
challenging conclusions or premises, and providing proofreading and flow suggestions. The peer reviewer is
the guardian of quality for technical journals and his or her role is every bit as important as that of the article

The peer reviewer will submit three copies of the "Reviewer Report" form along with the original journal
article and a letter to the journal editor (the professor). Grammatical and typographical errors should be
marked directly on the original manuscript. The letter to the editor should include a brief greeting, a statement
of purpose (why you are writing this letter), a short summary of your publication recommendation (publish or
don't publish), and a brief justification of your recommendation. Regardless of which recommendation you
make, you will not be asked to perform a second review of the paper.
The first page of the review provides an area for overall evaluation and specific criticisms and sugges-
tions. Direct questions are asked and explanations for your answers should be included in the comment
section, which comprises the rest of the review. The comments should be specific and informative with direct
questions, observations, or recommendations being made. Your grade will be based on the following:
Depth of analysis, including recommendations (50%)
Demonstration of technical understanding (25%)
Clarity of expression (25%)
Note: Your peer review will not affect the grade of the technical journal article author.

Miscellaneous Observations
1 Criticisms of articles should be constructive in nature. Comments such as "This is awful" will not lead to
a better paper (or a better grade).
El In addition to criticizing the article, your review should point out what was good about it.
El It is not enough to say what is wrong; you must also suggest what can be done about it.
LE Look for areas that are unclear. Often the author will present useful information, but it will be lost in
rhetoric or hyperbole.
3 Your suggestions must be reasonable. You could recommend running more trials, but you cannot tell the
writer to run 30 or more or to buy more sophisticated equipment.
I1 While critiquing the paper, consider the things that make a journal article stronger or weaker. Look for
these strengths and weaknesses in your own writing.

was unmistakable, and in fact,
the truly atrocious papers dis-
appeared altogether. It appears
that students are reluctant to
give inferior work to their
Students took the reviewing
task seriously. They avoided
simple hierarchial judgments
and focused on what made the
paper either effective or inef-
fective. They did an outstand-
ing job of identifying gram-
matical and mechanical
problems in the reports,
while still identifying
strengths and weaknesses in
the data analysis.
The final revised reports
were substantially better than
technical journals from previ-
ous years. Grammatical errors
were essentially eliminated.
More impressively, both the
level of analysis and demon-
strated technical understanding
were much greater.
The forced revision provided
important feedback that helped
the student improve both the
writing and the analysis. The
students also spoke of their in-
creased confidence in writing
the revised paper.
Additionally, student feed-
back concerning the peer re-
views has been uniformly posi-
tive. The students indicate that
writing a review led them to
recognize weaknesses in their
own writing. This improve-
ment in writing skills has been
evident in subsequent labora-
tory classes.
Since these results were pre-
sented at the 1997 American
Society for Engineering Edu-
cation conference,181 chemical

Summer 1998



Review Report Form

Reviewer Name (1 copy only)



1. Does this article warrant publication in this journal?
Acceptable in present form
Acceptable with minor revision, no further review necessary
Major revision and a second review is required
Not acceptable (provide detailed explanation under comments)

2. Is the title satisfactory?

3. Does the abstract adequately summarize the paper?
Could it be more complete or concise?

Indicate suggested revision on the manuscript or under comments.

4. Are sufficient references provided?
Are they appropriate and free from obvious omissions?

If not, explain.

5. Does the paper present material efficiently? Indicate suggested changes on the manuscript
or under comments.
(a) Could the clarity or efficiency be improved by changes in the order of the
(b) Should the language or grammar be improved?
(c) Are there portions of the text that could be omitted?

6. Are there errors in factual information, logic, statistical analysis, or


Address these issues in detail in the comments. Suggest improvements.

7. Mechanical errors (address on manuscript)
Figures or tables improperly or incompletely labeled or titled or not cited
Misuse of references (failure to cite, reference needed and not provided)


Overall, the use of peer reviews ap-
pears to be successful in the undergradu-
ate laboratory.


Student peer reviews seem to be an
effective and comparatively simple means
of enhancing student writing and data-
analysis skills. Key factors in operating
an effective peer-review system in the
lab include.

> Providing the student with a
template to help focus the

> Grading the reviewer.

> Making sure the original author
revises the paper to address the
reviewer's concerns.

> Keeping the reviewer anony-

1. Grimm, N., "Improving Students' Re-
sponses to Their Peers' Essays," Col-
lege Composition and Comm., 37, 91
2. Elbow, P., Writing Without Teachers,
Oxford University Press, New York,
NY (1973)
3. Herrington, A., and D. Cadman, "Peer
Review and Revising in an Anthropol-
ogy Course," College Composition and
Comm., 42, 184 (1991)
4. Holt, M., "The Value of Written Criti-
cism," College Composition and Comm.,
5. Howard, R., and S. Jamieson, The
Bedford Guide to Teaching Writing in
the Disciplines, Bedford Books of St.
Martin's Press, Boston, MA (1995)
6. Newell, J., D. Ludlow, and S.
Sternberg, "Progressive Development
of Oral and Written Communication
Skills Through an Integrated Labora-
tory Sequence," Chem. Eng. Ed., 31(2),
7. Ludlow, D., and K. Schulz, "Writing
Across the Curriculum at the Univer-
sity of North Dakota," J. of Eng. Ed.,
83(2), 161 (1994)
8. Newell, J.A., "The Use of Peer Review
in the Undergraduate Laboratory,"
ASEE National Meeting, Milwaukee,
June (1997)
9. Marr, J., personal communication
(1997) 0
Chemical Engineering Education


Experiment in Applied Optics
Continued from page 177.

An alternative approach to the unsteady-state batch reac-
tor is the steady-state continuous stirred tank reactor (CSTR).
The current apparatus could be easily modified by replacing
the burettes with delivery reservoirs. Flow meters and pumps
(or elevation) would be required. Inlet and withdrawal ports
should be fitted onto the transparent stirred vessel. The ex-
periment would still be run with excess bleach.
The theoretical model of the CSTR would begin with a
steady-state species balance[61

CA -CA -k' Ct= 0 (13)
CA dye concentration in the vessel
CAo dye concentration in the feed
r residence time.

Using Eq. (2), the species balance is

Ao A k'(L)1 -n n = 0 (14)
where A, is the absorbance of the reactor feed in the absence
of reaction (i.e., a blank run of dye and water without bleach
For first-order kinetics (n=l), Eq. (14) reduces to

= 1+k'T (15)

where a plot of VA vs residence time T would yield the
apparent rate constant k'. The advantage of the CSTR con-
figuration is the application of the laser diagnostic to a
steady-state problem. The disadvantage is that large vol-
umes of solutions would be required.

There are some interesting implications in this study. Since
the reaction proved to be first order in dye concentration, the
absorption cross section a was not used (see Eq. 9). This
means that an optical calibration to determine a is not
The reader may have noticed that this entire analysis was
done without knowing the reaction stoichiometry (see Eq.
3). Also, since the reaction is first order in dye, the initial dye
concentration is not needed. This is especially convenient
since the actual concentration of the dye liquid in the Durkee
containers is not known. The label indicates that the dye is
manufactured overseas, making inquiries difficult. It has
been identified as FD&C Blue No. 1, a complex organic
molecule of non-condensed aromatics.
Since the stoichiometry is not known, there is no a priori
Summer 1998

Figure 6. "Lumped" rate constant vs. bleach
concentration; Eq. 9.

way to determine how much bleach constitutes an "excess"
condition. If insufficient bleach is added, the variation in
time of both dye and bleach concentrations must be consid-
ered in the kinetic analysis of the absorbance data. If the
stoichiometric coefficient b in Eq. (3) is incorporated into
the analysis, its value could be determined as a "best fit"
Finally, this experiment is effective and relatively inex-
pensive. An old chart recorder, easily found in any surplus
equipment collection, helps to keep the cost down, and glass-
ware should be available in any chemistry stockroom. A
small He-Ne laser can be obtained for less than $100, al-
though the photodiode detector will cost more. The entire
experiment can be set up for about $700.

The authors would like to thank the NSF Combined Re-
search/Curriculum Development Program for funding a ma-
jor interdisciplinary optics initiative at NJIT under which
this experiment was developed (grant # EEC-9527491). The
principal author (RB) would like to thank senior technician
Benedict Barat for carrying out the very first test experi-
ments to establish viability.

1. "Optical Science and Engineering," National Science Foun-
dation Workshop, Arlington, VA, May 23-24 (1994)
2. Johnson, A.M., J.F. Federici. H. Grebel, T. Chang, and R.
Barat, "Optical Science and Engineering at NJIT: Multi-
Disciplinary Combined Research/Curriculum Development
Program," presented at the Annual Meeting ofASEE (1997)
3. Optics and Photonics News, p. 12, March (1997)
4. Optics and Photonics News, p. 10, July (1997)
5. Hecht, Optics, 3rd ed., John Wiley & Sons, New York, NY
6. Fogler, H.S., Elements of Chemical Reaction Engineering,
2nd ed., Prentice Hall, Englewood Cliffs, NJ (1992) 0




Part 4. Measuring Axial Dispersion in a Bubble Column

McMaster University Hamilton, Ontario, Canada L8S 4L7

he bubble column is one of the most efficient meth-
ods of gas-liquid contacting, and it is simple, com-
prising merely a vertical column and a means of
distributing gas into the liquid at the base of the column.
Deckwert1 has given a detailed account of bubble columns
and their usefulness as gas-liquid reactors as well as mass
transfer devices. Bubble columns are often used in organic
gas-liquid reactions such as hydrogenation, chlorination, and
alkylation. An important application is in the "liquefaction"
(hydrogenation) of suspended coal particles at high tempera-
ture and pressure.
Bubble columns are often operated countercurrently, with
the liquid fed to the top and the gas distributed to the base of
the column. The full benefits of countercurrent operation
can only be realized if both phases move in plug flow,[21 but
there is evidence of considerable axial mixing in the liquid
phase of bubble columns.11 With axial mixing, the mean
driving force for mass transfer is less than it would be in
plug flow.121
The effect of liquid-phase axial mixing on countercurrent
mass transfer can be expressed in terms of the Peclet number
Pe = UL/E (1)
where U is the superficial velocity of the liquid phase, L is
the active height of the gas-liquid dispersion, and E is the
axial dispersion coefficient. If the Peclet number is greater
than 20, plug flow of the liquid can be assumed for engineer-
ing design purposes."3' A Peclet number less than about 0.05
will correspond to well-mixed liquid behavior; that is to say
the solute concentration is essentially uniform throughout
the column. The terms U and L are well defined, but the

* Address: Lakehead University, Thunder Bay, Ontario, Canada
P7B 5E1

axial dispersion coefficient E must often be measured before
Pe can be estimated. Therefore, the measurement or predic-
tion of E is an important step in the design of a bubble
column contactor or reactor.
The axial dispersion coefficient E is defined in terms of
the axial dispersive flux (N) of a solute:

N -Ec (2)
Although in this equation E is analogous to the molecular
diffusion coefficient, it is determined by bulk liquid motion
such as turbulence, circulation, or motion of bubble wakes.
Its value is therefore independent of the molecular structure
of the solute. Because of its dependence on bulk fluid mo-
tion, E can be many orders of magnitude greater than mo-
lecular diffusion coefficients. Typically, E is in the range of
1 to 100 cm2/s, compared with about 10-5 cm2/s for molecu-
lar diffusion coefficients in liquids at ambient temperatures.

Malcolm Baird received his PhD in chemical
engineering from Cambridge University in 1960.
After some industrial experience and a post-doc-
toral fellowship at the University of Edinburgh, he
joined the McMaster University faculty in 1967.
His research interests are liquid-liquid extraction,
oscillatory fluid flows, and hydrodynamic model-
ing of metallurgical processes.

Inder Nirdosh received his BSc and MSc in
chemical engineering from Panjab University (In-
dia) and his PhD from Birmingham University
(United Kingdom). He joined Lakehead Univer-
sity in 1981, and his research interests are in the
fields of mineral processing and electrochemical

Copyright ChE Division of ASEE 1998

Chemical Engineering Education

Laboratory y

The objective of this experiment is to provide quantitative
data on unsteady axial dispersion as well as a visual observa-
tion of the turbulent mixing phenomenon in a small labora-
tory bubble column. The experiment uses an acid-base-indi-
cator system that adds some complexity to the data analysis
but eliminates the need for sampling and analysis of liquid at
different times and locations.

Figure 1 shows the experimental bubble column schemati-
cally. Axial mixing can be measured by the well-known
pulse tracer addition method1[3 in which the spread of the
initial pulse of tracer is determined by the unsteady diffusion
equation, assuming that no bulk flow is present:

ac E (3c
I- E (3)
at 7Z2
The conventional tracer method requires sampling the liquid
at one point (preferably more than one point) in the column,
followed by analysis of the samples. This is too time con-
suming for most undergraduate experiments. In this experi-

Acid addition

Figure 1. Schematic diagram of bubble column showing
coordinate system for equations. The baffle stack can be
removed for "open column" experiments.

ment, the tracer added is an acid that reacts very rapidly with
a base, which is initially uniformly distributed through the
liquid in the presence of a suitable indicator. The change in
color of the aqueous solution as the pH passes through the
neutral point is easily observable, and the movement of the
color change through the column can be timed and related to
the solution of Eq. (3) so as to give calculated values of E at
various conditions.

The column in Figure 1 is initially filled with a dilute
alkaline solution at a molar concentration Cb We recom-
mend diluting exactly 50 mL of 1.0 mol/L NaOH into dis-
tilled water in the column (volume about 40 L), with a few
mL of phenolphthalein solution to indicate alkalinity. The
concentrated "tracer," added to the top of the column at
t=0, is a measured volume of 1.0 mol/L hydrochloric
acid, which is more than sufficient to neutralize the so-
dium hydroxide. Observation of the column after the
addition of acid shows that the red color of the alkaline
phenolphthalein starts to disappear at the top of the col-
umn; the reaction zone (color change) then moves down-
ward until the last trace of red finally disappears at the
base of the column.
The free acid and base mix according to Eq. (3), which is
written for each solute, is
Where free acid is present

E 2Ca Ca C(4
(1-E) az2 at

Where free base is present

E aCb b (5)
(1- E) z2 at

The term in holdup (e) is a correction factor for the volume
of gas in the dispersion.
Note that E is the same for acid and base. The acid-base
reaction is extremely fast and takes place on a 1:1 mole ratio
at a zone where Ca and Cb are both very small (pH about 7).
The reaction (color change) zone is assumed to be a very
thin boundary at which

Z= Zr
Ca, Cb- 0

Na =-Nb or -a
az az

The chemical reaction is confined to a very thin zone (z z z,
and ca o0_ cb). At other parts of the column there is no
chemical reaction, and therefore Eqs. (4) and (5) do not
contain a reaction term. This type of assumption is well

Summer 1998


known in the analysis of gas absorption with instantaneous
irreversible reaction.[41
Equations (4) and (5) can be combined in a way consistent
with Eqs. (6-8) by creating a concentration variable c' as

For ca > 0 c' = a +cb (9)
For cb >0 c'=cb, -Cb (10)

In this way, an equation of the same form as Eq. (3) can be
obtained, with c'=0 at t=0 and c' = (ca. + ) at t = o.

The standard solution to Eq. (3) in terms of c' can be
shown to be

c = c +2 e-"an2T2 n2zj (11)
where, in this case
Sa=--- (12)
(1- )L2

If the elapsed time is long enough for ca > 0.1, all terms but
the first in the series can be neglected, and Eq. (11) becomes

=c',c 1+ 2e 2 cos T (13)

Making the transformation from c_ by Eq. (9), we find

c'= +(c b ) i 1+2e-2 Ocos ZL (14)

where c, is the final concentration of acid remaining in the
column after mixing has been completed (Ca >o).
From the point of view of quantitative measurement, the
most accurate quantity is the time t* it takes for the last trace
of red (alkalinity) to vanish at the base of the column. The
neutral zone at z, = L corresponds to c' = cb and Eq. (14)

cb, = (ca +Cb )(- 2 e-"2

e -a2 = Ca-
2(ca- +cb1)

This equation is more usefully expressed in terms of R, the
overall acid/base ratio,


Ca- + Cbl Ca-
ca+cbi cbl

e-an2 = R-1

Taking logarithms and rearranging gives

1 (2R
H2 R -)
2 Hence

(1-E)L2 (2R )
t E fYnR (20)
x2E R-1'
If t* and e are measured and R and L are known, E can be
calculated in a given case. Better still, several experiments
can be done under the same conditions of gas flow rate (i.e.,
same E) but with different values of R. Then the measured
values of t* can be plotted against [n(2R /(R 1)) to check for
linearity. The slope of the plot will provide an average value
of E.

The bubble column used in this project (see Figure 1) is
vertical, cylindrical, open at the top, and closed at the bot-
tom. Internal diameter is 8.9 cm and height is 65 cm. The
material of construction is clear (methacrylate) plastic. Air is
supplied from the 20 psig (230 kPa) line via a rotameter and
needle valve. The metered flow of air enters the base of the
column via a bubble distributor. A full list of dimensions and
operating conditions is given in Table 1. Experiments can be
carried out in the absence of column internals, or a stack of
baffles (Figure 1) can be placed in the column.

Before experiments are begun, a check should be made to
ensure that the column is mounted vertically, using a spirit
level. Mixing can be affected if the column is tilted away
from the vertical.
At the beginning of each experiment, the column is filled
to within about 10 cm of the top with distilled water. Then,
exactly 50 mL of 1.0 mol/L sodium hydroxide is added. The

Summary of Experimental Conditions

Column diameter (d) ........................................ 8.9 cm
Column height ........................................... 65 cm
Height of gassed dispersion (L)....................... 60 cm (typical)
Number of centrally mounted disk baffles ........9
Baffle thickness ............................................... 0.75 cm
Baffle diameter .................... ........................ 5.08 cm
Spacing between baffles .................................. 5.08 cm
Values of R studied ......................................... 1.05, 1.10, 1.40, 2.00
Values of ug studied ......................................... 0.83, 1.96, 3.11 cm/s
Gas holdups measured ( ) .............................. 0.03 to 0.12

Chemical Engineering Education


alkaline solution is thoroughly mixed by bubbling air for 5
minutes. During this process, about 5 mL of phenol-
phthalein solution is added to color the solution red. The
air flow is started at a desired rate. A volume of stock 1.0
mol/L hydrochloric acid, equal to R times the volume of
1.0 mol/L sodium hydroxide added earlier, is measured
out. It is necessary that R>1 in order to completely decol-
orize the solution.
The mixing process is started by quickly adding the acid
(within less than one second) to the gas-liquid dispersion.
We suggest that a small deflector (e.g., an inverted, 3-cm-
diameter, plastic filter funnel) be used so that the acid can be
tipped into the column without creating a strong vertical
eddy. As the acid is added, a stopwatch is started. The spread
of the neutral zone down the column can be observed. It may
be possible to measure zone position as a function of time,
but the most important measurement is t*, the time it takes
for the last trace of red color (alkalinity) to disappear from
the base of the column. An estimate of error in the t* mea-
surement should be made.
A measurement of the gas holdup E is required for
each gas flow rate. This is done by comparing the height
of the gassed dispersion (L) with the height of the
ungassed liquid (L0),
= (L-Lo)/L (21)

The baffles (5.08-cm diameter) were mounted in a stack on a
central support rod that was also equipped with two "spi-
ders" to ensure centering in the column. The baffle stack
could be lowered into the column from the open top. The
baffle stack is shown in place in Figure 1.

The student will be handling 1.0 mol/L solutions of hydro-
chloric acid and sodium hydroxide. The main hazards are
possible ingestion or eye damage from splashes. Skin
contact is to be avoided, although at the 1.0 mol/L con-
centration, the skin hazard is moderate. Goggles should
be worn; if any acid/base gets on the skin or clothing,
wash exposed area with plenty of water immediately
after contact. Rinse the apparatus thoroughly with water
after each lab day to prevent corrosion damage. It is
particularly important to rinse any metal parts that have
been in contact with acid or base.


On the first day, the column can be operated as an open
bubble column without the baffle stack. For a given air flow
rate and distributor, the mixing time t* should be measured
for four values of "R." We suggest values of 1.05, 1.10, 1.4,
and 2.0. Immediately after this, the four data points should



20 .0... o
60 -Present

40 r

20 -Z5- -
o- o Baffles
0'*-"-------------------- ------
0 1 2 3 4
Function ln(2R/(R-1))

Figure 2. Typical data plot of neutralization time t*
versus function of R according to Eq. (20). For these
data, Ug=0.83 cm/s.

be roughly plotted as t* versus the log function of R (Eq. 20).
If any of the points appear inconsistent, that particular ex-
periment should be repeated.
Then the air flow rate is changed to a new set value, and
four more experiments (varying R) are done. The sequence
is continued until three air flow rates have been covered. A
total of twelve experiments is expected, allowing approxi-
mately 10 to 15 minutes per experiment. Gas holdup is
measured for each gas flow rate.
Then, on the second day, another series of measurements
is performed, varying R and the air flow rate, with the baffle
stack in place. This can be shown to give much less axial
mixing than in the open column case.

Figure 2 shows typical data151 on the effects of the chemi-
cal ratio R on the measured neutralization time t*. Complete
neutralization takes longer as the value of R approaches
unity (i.e., the abscissa function approaches infinity). The
data points at the smallest value of R (typically 1.05) are
sensitive to small errors in setting the ratio R, but it can be
seen that the data lie reasonably close to a straight line
through the origin. This confirms the assumptions lead-
ing to Eq. (20). For a given value of R the values of t* are
much greater in the presence of baffles than in the
unbaffled condition, which is also to be expected. The
slopes of the data plots in Figure 2 are found by linear
regression and the value of E can then be calculated from
Eq. (20).
For each gas flow setting studied, and for the presence or
absence of baffle plates, a plot of the form of Figure 2 is
made, and values of E are calculated. Figure 3 shows typical
Continued on page 213.

Summer 1998



Mass Transfer and Axial Dispersion

in a Reciprocating-Plate Liquid Extraction Column

University of Washington Seattle, WA 98195-1750

The BSChE program at the University of Washington
(UW) includes a required 2-quarter lab sequence.
The first course emphasizes measurement fundamen-
tals, transport phenomena, statistical analysis of data, and
technical communication skills. Experiments are relatively
The second course involves open-ended assignments and
the equipment is more complex. Educational goals are to
Reinforce and integrate the material covered in lecture
Develop students' abilities to plan an experiment, carry it
out with good technique, and analyze results in a
professional manner
Refine oral and written communication skills
Students work in teams of three and have three 4-hour lab
periods to collect data. A team performs three experiments
during the quarter. For each experiment, a designated leader
submits a written plan (including a safety analysis) and
defends it in a conference with the instructor prior to the first
day of data collection. At the end of three weeks, the leader
submits a detailed written report, and another team member
gives an oral report. The leadership role and other responsi-
bilities rotate from one experiment to the next.
We have recently developed a multifaceted experiment on
an important unit operation (liquid-liquid extraction). A typi-
cal assignment asks a team to characterize one or more of the
Equilibrium solute distribution between two liquidphases
as a function of solute concentration
Mass transfer coefficients (or HTUs) for steady-state
extraction in a reciprocating-plate column as a function
of operating conditions
Residence-time distribution and the influence of axial
dispersion on column efficiency
In this paper we will describe the experiments briefly, explain-
ing how they fit in with the educational mission of the course.

Our liquid-liquid contactor is a standard Karr'l reciprocat-
ing-plate column (see Figure 1). The mass transfer zone is a
glass tube, 2.5 cm ID, 2 m long. It contains 36 perforated
stainless steel plates (approximately 60% open area) spaced
at 5-cm intervals on a central stainless shaft. An electric
motor makes the shaft oscillate vertically at frequencies of
0-180 cycles/minute (displayed on a meter and easily varied
during operation). The vertical stroke length is 0-5 cm (vari-
able, but constant for the duration of a run). We bought the
column, drive system, and internals as a package from the
Chem-Pro Equipment Company.
Kerosene is the dispersed phase (a 40-liter stainless steel
reservoir allows runs of several hours). It passes through a
rotameter and enters as large droplets at the bottom of the
column. Smaller droplets form as the kerosene flows upward
through the reciprocating plates in the contacting zone. Drop-
lets coalesce above the aqueous inlet location, and the kero-
sene flows by gravity to a collection vessel.
A feed of tap water passes through a rotameter, enters the
column about a foot below the top, and flows downward by
gravity, eventually leaving at the bottom. With the drain

June Luke received her BS degree in chemical
engineering from the University of Washington
in 1997 She is currently a first-year graduate
student at the University of Delaware. This past
summer she worked as an intern at Hercules,
Inc., where she studied polymerization reactions
employing microwave energy.

N. Lawrence Ricker received his BS from Michi-
gan, and his MS and PhD from Berkeley,. all in
chemical engineering. He worked as a systems
analyst for Air Products and Chemicals before
joining the University of Washington's depart-
ment in 1978, where he is currently Professor of
Chemical Engineering. His research specialty is
process simulation and control.

@ Copyright ChE Division of ASEE 1998

Chemical Engineering Education


valve of Figure 1 closed, the ex-
iting water flows upward through
flexible tubing to a tee, then back
down to a collection vessel. The
location of the tee (adjusted by a
sliding clamp) controls the wa-
ter/kerosene interface location
inside the column. The section
above the tee provides an emer-
gency overflow line and a vent
A 6-mm sampling port with a
rubber septum allows either
tracer injection or removal of an
internal sample at the water feed
entry point. A conductivity meter
(Omega CDTX-81, 0-200 gS/
cm) monitors the conductivity of
the exiting water phase. We
added a zone of 5-mm glass
beads between the kerosene in-
let and conductivity meter to pro-
mote plug flow in that region.
Otherwise, flow is laminar and
Taylor dispersion biases the mea-
surement of the residence-time
distribution. The conductivity
meter provides an analog output
to a microcomputer.

The system n-butyric acid/
kerosene/water offers the follow-
ing advantages:
Butyric acid concentrations

... the challenge is to
constraints of the lab,
credible results, underste
and communicate it al
students appreciate tl
challenge imn

Figure 1. Standard Karr Rec

can be measured accurately
by titration, even at high dilution. Use of a dilute
solute simplifies analysis of mass transfer experiments
and minimizes chemical consumption and waste-dis-
posal problems. Dilute aqueous solutions of butyric
acid and its salts may be sewered. Other acids could
be used, but butyric is one of the few having a distri-
bution coefficient of order unity in a nonpolar or-
ganic. This makes it easy to avoid a pinch of operat-
ing and equilibrium lines in mass transfer experi-
Butyric acid's strong, unpleasant odor is a disadvan-
tage; good ventilation is essential.

Kerosene is inexpensive, insoluble in water, and rea-
sonably safe and odorless. Nearly all of it can be
recovered and recycled. The lack of a known molecu-
lar weight forces students to think about the choice of
Summer 1998

o work within the the basis for equilibrium and
obtain consistent, mass-transfer calculations.
obtain consistent,
and their limitations, One could easily substitute a
l effectively. Some pure hydrocarbon, however.
his and rise to the > Distribution coefficient
nediately. data are unavailable in the lit-
erature; they must be mea-
sured to determine mass trans-
OVERFLOWTUBE fer driving forces. We pro-
vide the students with pub-
lished data for the analog sys-
tem butyric acid/hexane/wa-
- ADJUSTABLE CLAMP ter, however. This gives them
a basis for comparison and
allows them to estimate ap-
-propriate solvent/feed ratios
for extraction runs.
> Results are insensitive to
variations in ambient tem-
WATER FEED perature in the lab, avoiding
the need for temperature con-
Measurement of Solute Dis-
tribution This bench-top ex-
periment seems simple, but it
requires careful planning and
FEED execution. Mixtures of the
three chemicals equilibrate in
a suitable container at room
temperature. Analysis of
WATER OUTLET samples from one or both
phases follows. Students must
'iprocating-Plate Column. choose overall mixture com-
positions such that equilibrated
phases have solute concentra-
tions in the correct range (i.e., as expected in mass transfer
experiments). Other considerations include the agitation pro-
tocol, time needed to reach equilibrium (we suggest at least
one-half hour), and whether it is necessary to measure the
solute concentration in both phases. An alternative is to mea-
sure one and estimate the other by mass balance. We ask the
students to perform an error analysis that compares these alter-
Example data appear in Figure 2. The empirical relation-
ship y=0.20x"82 fits well over the range x=0 to 2 wt. %
butyric acid in the aqueous phase, y=0 to 0.8 wt. % butyric
acid in the kerosene phase. Heinonen and Tommila[21 ob-
served similar nonlinear behavior for n-butyric acid/hexane/
water, attributing it to the dimerization of butyric acid in the
nonpolar organic phase.
Team leaders usually assign one member to this task,

i Laboratory
which can run in parallel with mass transfer experiments.
Study of the full composition range is a significant effort,
however. The instructor might instead provide data from a
previous group's report, asking the current team to verify
some aspect of it. A variant is to provide two or more old
reports (good and bad), forcing the team to separate the
wheat from the chaff.
Titrations Analysis of aqueous samples for butyric acid
content is straightforward. We titrate with aqueous NaOH
(usually 0.5 N). A phenolphthalein indicator gives a sharp
endpoint. For organic samples, one approach is to add the
sample to a beaker containing water and phenolphthalein,
then slowly titrate with NaOH while stirring vigorously (us-
ing a magnetic stir bar). The butyric acid transfers to the
aqueous phase where it reacts to form sodium butyrate,
which is insoluble in kerosene. Mass transfer is slow, how-
ever, so the endpoint is less obvious. An alternative is to
equilibrate the sample with an excess of aqueous NaOH,
then titrate the unreacted NaOH with aqueous HC1. Another
is titration with NaOH in a suitable organic solvent, such as
methanol. This requires a pH meter to detect the endpoint.


A typical assignment asks the team to measure the overall
height of a transfer unit (HTU) as a function of shaft oscilla-
tion frequency. The feed is kerosene with a specified con-
centration of butyric acid (usually less than 1 wt. %), and the
solvent is tap water. We also specify the shaft stroke length
and the kerosene feed rate. The team must select the water
rates) and determine the region of operability. The main
constraint is flooding, which occurs at high feed rates or
high oscillation frequency. Another is phase inversion, i.e.,
a transition from water-continuous to kerosene-continuous
operation. It is also important to avoid a pinch of the operat-
ing and equilibrium lines. Otherwise the calculated HTU
value is very uncertain.
Students find that running the equipment is non-trivial.
The minimum HTU is usually close to the flooding point,
where small changes in operating conditions (especially in-
sufficient attention to flow control) can have dramatic ef-
fects, easily seen within the glass column. It usually takes a
team one full lab period to develop the skill needed to collect
useful data in this region. To reduce frustration, we instruct
them to avoid flooding conditions in early experiments.
We emphasize the importance of steady-state operation,
suggesting that they measure the outlet aqueous butyric acid
concentration-the easiest to titrate-periodically until the
value is essentially constant, then analyze a sample from the
kerosene raffinate to check mass-balance closure. It takes
about one-half hour (or turnover of four column volumes) to
reach steady state after a change in operating conditions.
Figure 2 shows a typical operating line for the measured

Equilibrium and Operating Lines
Sample Run

06 FOperating Line I

SEquilibrium Line @ 25 deg C
m ou e y = o.20x'

tion of plug flow in two immiscible phases, and dilute
operation. Other operating conditions were oscillation fre-
quency = 110 cycles/minute, kerosene feed = 3.0 g/s, and
water feed = 1.1 g/s. Calculation of the overall number of
transfer units (NTU) would be trivial if the equilibrium line
were straight.131 With a curved equilibrium line, however,
students must review the theory carefully, which enhances
their understanding.
For the case shown in Figure 2, the NTU is 0.83, giving an
HTU of 225 cm. Although the HTU decreases with increas-
ing frequency as expected, the values remain well above the
5-25 cm reported in the literature.'l This motivates the as-
signments for the follow-on teams, who are asked to deter-
mine the reason for the large HTUs. Possible variations
(other than changes in the chemical system) are the stroke
length, tray spacing and design, and choice of the dispersed
phase (water-dispersed operation is poor with stainless trays,
however). Axial disper Aia mission can also affect liquid extraction
performance,t[451 suggesting a study of this possibility.

There are two ways to check the extent of axial dispersion.
The more direct way is to analyze samples from within the
column. For example, if one expects strong axial dispersion
in the continuous phase, there should be a discontinuity in
the continuous-phase concentration at the feed location (a
"concentration jump," as shown in Figure 2).
Good technique is needed to remove a representative in-
ternal sample. Significant amounts of the dispersed phase
may be entrained, and the students must decide how to deal
with this. Figure 2 shows that the jump was rather small in
the sample run; the corresponding HTU is 220 cm, a de-
crease of only 2% from the plug-flow value.
The alternative is to measure the residence-time distribu-
tion (RTD). This has been advocated in CEE previously for
gas-liquid contractors16 ] Several texts provide good back-
Chemical Engineering Education

[ Laboratory

ground on calculational methods and interpretation.781'
The continuous phase is most prone to axial dispersion. To
measure the RTD, a student injects about 1 ml of KMnO4
dye tracer into the aqueous (continuous) phase at the feed
point (Figure 1). The pulse is easy to follow visually, and the
concentration transient at the outlet can be analyzed quanti-
tatively using the conductivity meter.
The KMnO4 is insoluble in kerosene, so it is possible to
measure the continuous-phase RTD during normal counter-
current operation. We have found, however, that entrained
kerosene droplets can interfere with conductivity measure-
ments. The main problem is development of an insulating
coating on the probe surface. Students can mitigate this by
operating far from the flood point (to minimize entrain-
ment), making baseline corrections on conductivity mea-
surements and cleaning the probe periodically-but this is
inconvenient. Thus, we usually assign a study of RTD in the
absence of kerosene. Variables are then the water feed rate
and the degree of agitation.
An alternative would be to measure dye concentration via
UV-vis spectroscopy, which should be less sensitive to the
presence of kerosene. We have not tried this yet because our
measurements show that the presence of the dispersed phase
has little impact on the RTD in our 2.5-cm column. This
would not be the case in general.
The students use the measured RTD to calculate a Peclet
number at each condition. Peclet numbers can be related to
the mass transfer results using the equations developed by
Sleicher,[51 which provide a correction to the assumption of
plug flow. Laddha and Degaleesan'l4 (pp. 125-127) illustrate
the calculations in detail. For the conditions of the sample
run, assuming plug flow in the dispersed phase, the observed
continuous-phase Peclet number leads to a correction of
only 2%, i.e., essentially the same as the correction based on
the measured concentration jump.


For our 2.5-cm reciprocating-plate column, axial disper-
sion increases with increasing plate oscillation frequency
and decreasing feed rate, but is never a dominant factor.
Continuous-phase Peclet numbers are of order 30 or greater
under most conditions. The presence or absence of the dis-
persed phase has only a small effect. These observations are
in agreement with those described in the literature.'91
This forces students to look elsewhere to explain the large
HTU values found in mass-transfer runs. It is instructive to
follow up with a more polar-dispersed phase having a den-
sity closer to water and a lower interfacial tension (e.g.,
MIBK), which exhibits much lower HTUs. It would also be
interesting to run RTD experiments in a column with a much
lower length-to-diameter ratio where axial dispersion would
Summer 1998

be more significant.
We occasionally ask a team to quantify flooding behavior.
It would be logical to do this prior to mass-transfer ex-
periments. Useful information on expected behavior is
available in the literature.1101
Student reaction to the RTD experiments has been very
positive. They appreciate the visualization of axial dispersion
and the automated collection of conductivity data, which makes
calculation of the RTD easy. They enthusiastically vary oper-
ating conditions over a wide range, check repeatability, etc.
They are less pleased with the phase-equilibrium and mass-
transfer experiments. Most complaints concern the odor of
butyric acid and the tedium of multiple titrations. We could
change the solute, but butyric acid has important advantages,
as noted previously. Odor can be reduced, but is difficult to
eliminate. Automated titration would increase student pro-
ductivity and morale.
In general, we want the students to have some hands-on
responsibilities in each experiment. In contrast to their a
priori expectations, the point is not to demonstrate perfect
(or even good) agreement between the textbook and reality,
nor is it to see modern industrial equipment and instrumenta-
tion in action (although such exposure is certainly worth-
while). Rather, the challenge is to work within the con-
straints of the lab, obtain consistent, credible results, under-
stand their limitations, and communicate it all effectively.
Some students appreciate this and rise to the challenge im-
mediately. Others grumble throughout the quarter, but judg-
ing by alumni surveys, their frustration with the course often
turns to praise once they have graduated.

1. Karr, A.E., "Performance of a Reciprocating-Plate Extrac-
tion Column," AIChE J., 5, 446 (1959)
2. Heinonen, K., and E. Tommila, "Distribution of Alcohols
and Carboxylic Acids Between Water and Hexane," Suomen
Kemistilehti B, 42, 113 (1969)
3. Geankoplis, C.J., Transport Process and Unit Operations,
3rd ed., Prentice-Hall, Englewood Cliffs, NJ (1993)
4. Laddha, G.S., and T.E. Degaleesan, Transport Phenomena
in Liquid Extraction, McGraw-Hill, New Delhi, India (1978)
5. Sleicher, C.A., "Axial Mixing and Extraction Efficiency,"
AIChEJ., 5, 145 (1959)
6. Davis, R.A., J.H. Doyle, and O.C. Sandall, "Liquid-Phase
Axial Dispersion in a Packed Gas Absorption Column," Chem.
Eng. Ed., 27, 20 (1993)
7. Levenspiel, O., Chemical Reaction Engineering, 2nd ed.,
Wiley, New York, NY (1972)
8. Fogler, H.S., Elements of Chemical Reaction Engineering,
2nd ed., Prentice-Hall, Englewood Cliffs, NJ (1992)
9. Karr, A.E., and T.C. Lo, "Scaleup of Large Diameter Recip-
rocating-Plate Extraction Columns," Chem. Eng. Prog.,
72(11), 68 (1976)
10. Camurdan, M.C., M.H.I. Baird, and P.A. Taylor, "Steady
State Hydrodynamic and Mass Transfer Characteristics of
a Karr Extraction Column," Can. J. Chem. Eng., 67, 554
(1989) 0

Random Thoughts...


North Carolina State University Raleigh, NC 27695

Think back to your experience as a new faculty mem-
ber. If you conjure up mainly memories of a happy
time of exciting new opportunities suddenly open to
you-new colleagueships, new intellectual challenges, etc.-
it's probably been quite a while since you were a new faculty
If you think a little more about it, you may start to recall
the hurdles you had to jump over to start a research pro-
gram-writing proposals and trying to get them funded,
attracting and learning how to deal with graduate students,
and having to churn out a large number of refereed papers
while you were still trying to figure out how to do research.
You may remember the incredibly time-consuming labor of
planning and teaching new courses and the headaches of
dealing with bored classes and poor student performance
and possibly cheating and poor ratings and a host of other
problems you never thought about when you were a student.
And you may recall sitting through endless departmental
faculty and committee meetings, wondering how you could
manage to squeeze in some time for your family and your-
self on top of everything else you had to do. Learning to
cope with all those conflicting demands on your time and
energy was probably not a fun-filled experience for you.
Few faculty members ever receive guidance on how to be a
faculty member, and it can take years to figure it out by trial-
Entry into the profession is if anything harder now than it
used to be. Even institutions that historically emphasized
undergraduate education are pushing their new faculty mem-
bers to build strong funding and publication records in their
first three years, and most institutions still do little or noth-
ing to help the newcomers make the transition from graduate
student to assistant professor. The stress on the new faculty
members can be debilitating, and those who survive often do
so at a severe cost to their personal relationships and/or
Robert Boice, head of a faculty teaching center at SUNY-

Stony Brook, has spent many years studying faculty mem-
bers in their first 3-4 years and has summarized his observa-
tions in The New Faculty Member.11 This column outlines
some of his main points.
Common characteristics of the typical new faculty mem-
bers Boice observed are that they
3 spent far less time on scholarly writing (proposals
and papers) than was needed to meet promotion and
tenure criteria for their institutions
3 admitted to going to class overprepared (with more
material than they could reasonably cover in the
allotted time) and rushing to complete everything,
often at the expense of active student participation.
Many spent nearly 30 hours per week on class
3 equated good teaching with good content
3 taught defensively, doing whatever they could to
avoid student complaints. They were primarily
concerned that students would complain about

Rebecca Brent is Associate Professor of Educa-
tion at East Carolina University. She received her
BA from Millsaps College, her MEd from Missis-
sippi State University, and her EdD from Auburn
University. Her research interests include appli-
cations of simulation in teacher education and
writing across the curriculum. Before joining the
faculty at ECU, she taught at elementary schools
in Jackson, Mississippi, and Mobile, Alabama.
She received the 1994 East Carolina University
Outstanding Teacher Award.
Richard M. Felder is Hoechst Celanese Profes-
sor of Chemical Engineering at North Carolina
State University. He received his BChE from City
College of CUNY and his PhD from Princeton. He
has presented courses on chemical engineering
principles, reactor design, process optimization,
and effective teaching to various American and
foreign industries and institutions. He is coauthor
of the text Elementary Principles of Chemical Pro-
cesses (Wiley, 1986).

Copyright ChE Division ofASEE 1998

Chemical Engineering Education

content errors
3 received student evaluations that fell well below
their expectations and blamed the results on external
factors (invalid rating systems, poor students,
unfavorable class times and sizes)
3 experienced a sense of loneliness and lack of
collegial acceptance, and had difficulty establishing
productive contacts with colleagues who could
provide guidance and support

Not all new faculty members fit this description. Boice
identified 5-9% of new faculty as "quick starters," who in
their first 2-3 years turned out enough proposals and papers
to put them in fine shape for promotion and tenure. They
also scored in the top quartile of peer and student ratings of
teaching and self-ratings of their enjoyment and comfort
levels as teachers. Unlike the majority of their colleagues,
the quick starters

3 spent three hours or more per week on scholarly
3 integrated their research into their undergraduate
3 did not spend major amounts of time on course
preparation (after their first semester, they averaged
1-1.5 hours of preparation per lecture hour)
3 lectured at a pace that allowed for active student
3 regularly sought advice from colleagues, averaging
four hours a week on discussions of research and

The main differences between typical new faculty and
quick starters are the latter group's abilities to balance con-
flicting demands on their time and to quickly establish pro-
ductive networking with colleagues. Boice has developed a
"balance program" to help new faculty members do those
things. Participants in the program commit to these guide-

1. Limit classroom preparation to a maximum of two
hours per hour of lecture.
This target is extremely difficult for many
professors, but those who manage to reach it find
that they can still cover what they want to cover,
appear more relaxed to their students, and are
better able to maintain a pace that encourages
active student involvement in class.

All of the Random Thoughts columns a teacl

Summer 1998

2. Spend 30-60 minutes a day on scholarly writing.
New faculty often feel they must have long
unbroken stretches of time to write, but the
demands of an academic career seldom allow this
luxury. Writing for a set time daily leads to steady
productivity and fewer feelings of anxiety over
failure to meet scholarly productivity expectations.

3. Spend at least 2 hours a week on discussions with
colleagues focused on teaching and research.

(Periodic meetings over lunch are convenient for
such networking.) It is difficult for most new
faculty members to meet this commitment, but
doing so pays big dividends. Good contacts
provide ideas and sometimes tangible assistance in
getting a research program off the ground and/or
improving teaching success.

4. Keep daily records of work time expenditure.

Recording helps new faculty self-monitor how
well they are meeting Commitments 1-3.

5. Integrate research interests into lectures.
Doing so leads to greater enthusiasm for teaching
as well as recruitment of students as research

Boice found that faculty going through this program ini-
tially resisted its requirements, particularly the one about
limiting lecture preparation time, but after five weeks they
began to look and feel more like quick starters. Regular
meetings with a facilitator or mentor were instrumental in
helping them stay with the program. Once they attained the
standards set out in the plan, they reported greater efficiency
and a higher level of comfort in their teaching.
The New Faculty Member offers a variety of useful sug-
gestions for supporting new faculty. We recommend it to
administrators, mentors, faculty developers, and anyone else
concerned with helping new faculty members attain the lev-
els of research productivity and teaching skill for which their
potential was recognized when they were hired.
1. Robert Boice, The New Faculty Member, San Francisco,
Jossey-Bass (1992). For additional discussions of problems
faced by new faculty members and ways their departments
can support them, see the Random Thoughts columns "Teach-
ing Teachers to Teach: The Case for Mentoring," Chem.
Engr. Education 27(3), 176-177 (1993) and "Things I Wish
They Had Told Me," Chem. Engr. Education, 28(2), 108-109
(1994). 0

re now available on the World Wide Web at
ling/ and at

iz,]a curriculum



At The University of New Brunswick

University of New Brunswick Fredericton, New Brunswick, Canada E3B 5A3

n an attempt to better prepare our students for their life
after graduation, the Department of Chemical Engineer-
ing continuously seeks ways to expose the students to
the practical aspects of chemical engineering. Factory tours
and laboratory exercises present students with early oppor-
tunities to experience the world of chemical engineering
during their first and second years.J" After studying several
of the chemical engineering core courses (material and en-
ergy balances, heat and mass transfer, fluid mechanics, and
thermodynamics), the students are given the opportunity to
apply what they have learned to real industrial problems
through an industrial project course known as "Practice
School," and during their senior year, their practical experi-
ence is extended during a Plant Design course. This paper
describes the industrially oriented approach taken in both the
Practice School and Plant Design courses.

Practice School projects can be considered as an introduc-
tion to the concepts of industrial design. They generally
accent the application of material and energy balances and
thus tend to be relatively basic from a design point of view.
But within the scope of their project, the students are ex-
pected to develop a detailed design solution. This industry-
sponsored endeavor allows the students to become familiar
with many aspects of an engineer's working life. Almost all
students participate in this course (a limited number, how-
ever, exercise the option of preparing a report on a depart-
mentally approved research topic).
Course Logistic
A two-week project in selected industrial process plants is
scheduled after the spring examination period (the last week

Guido Bendrich joined the Department of Chemi-
cal Engineering at the University of New
Brunswick after spending some nineteen years
in various industrial settings throughout the world.
He obtained his PhD from McMaster University
in 1992. His teaching and research interests are
in industrial plant design, cost estimation, plas-
tics processing, developing communication skills,
and education.

Todd Pugsley received his PhD from the Uni-
versity of Calgary in 1995 and was a Post-Doc-
toral Fellow at the Technical University Ham-
burg-Harburg (Germany) for six months before
joining the Department of Chemical Engineering
at the University of New Brunswick in 1996. His
research interests are gas-solid fluidization and
reaction engineering.

in April to the first week in May). Groups of third-year
students (jointly supervised by a faculty member and a pro-
fessional engineer employed by the company) are assigned
to engineering projects to be carried out on industrial pro-
cess units. While working in the plant, the students are
introduced not only to the problem at hand, but (perhaps of
equal importance) also to the various aspects of real-life
engineering, including time management and teamwork. The
students are also encouraged to take very detailed notes
while working in the plant.
At the end of the two-week period, the group presents their
results orally and submits a preliminary typewritten report to
the company representatives. The students are then expected
to work during the summer months on the final solution to
the design problem. Although the group members tend to
disperse to summer employment during this time, they are
able to stay in effective contact through electronic mail, fax,

Copyright ChE Division of ASEE 1998

Chemical Engineering Education

and telephone. The combination of detailed plant notes and
continued communication throughout the summer months is
critical when the students return for the fall semester since,
at that time, the results of their work are documented in the
form of a final comprehensive report that must be submitted
by the end of September. The following outline shows the
typical content of the comprehensive report.
Introduction to the Design Problem This section
presents a brief introduction to the industry and the
assigned task.
Theory A detailed discussion of the governing
equations and principles is given in this section. The
knowledge gained in the core courses is essential to
providing a concise description of the relevant theory.
Results The data taken at the plant are presented in
this section.
Data Analysis An in-depth data analysis leads to the
final design. Simulation software packages may be of
assistance to the students in the final design. Working
with industry on a viable solution to one of its prob-
lems also entails an economic evaluation of the final
Conclusions and Recommendations Suggestions for
the implementation of the final design and the possible
continuation offuture studies are put forward to the

To complete the requirements of the course, each Practice
School group is required to present its work in a 15-minute
oral presentation scheduled during the month of October.
Invitations to visit the department and attend the presenta-
tion are extended to the companies that participated, to the
other engineering departments, and to the public.
Suitable Practice-School Projects
When the Practice School course was initiated in 1970, all
of the projects were supplied by a lead-smelting company in

... students are given the opportunity to
apply what they have learned to real industrial
problems through an industrial project course
known as "Practice School," and during their
senior year, their practical experience is
extended during a Plant Design course.

the northern region of the Province. At that time, enrollment
in the department was less than it is now; today, we have as
many as thirty-five students for whom we must find suitable
projects each spring. With groups of three or four at each
industrial location, this translates into a need for ten or more
projects every year. As a result, students now work in indus-
tries from across the Province, and we rely heavily on the
industrial contacts of the individual faculty members to ob-
tain the necessary projects.
The industries that participate in the Practice School wel-
come the opportunity and, in many cases, have come to
depend on the work that is done by the students every year.
Many of the problems that our students work on are prob-
lems for which the typical process engineer in these plants
does not have the time. Examples are projects involving
process water balances or balances on components of waste
streams around sections of the plant.
As might be expected, certain projects lend themselves
better to the Practice School structure than others. Because
of the student's relatively short stay at the industrial site, the
projects must be well defined by the company, with clear
objectives that can be met in two weeks. Also, it is important
that any equipment provided by the company for a sampling
campaign or for laboratory analysis be in working order
and ready to use on the students' first day at the plant.
Furthermore, we have observed that projects that allow
the students to work independently (i.e., limited depen-
dence on plant personnel for assistance in sample collec-
tion or laboratory analysis) are most desirable. Table 1
lists the most recent Practice-School topics in which we

Recent Practice-School Projects

Integrated wood pulp and chemical
production facility

Integrated wood pulp and chemical
production facility

Crude-oil refinery
Lead-concentrate mining operation

Project Title
Design of an HCI vacuum scrubber
for the electrochemical plant

Sodium chlorate filter replacement and
an alternative method for sodium
hypochlorite removal

Desalter waste minimization
Residence time distribution and grade
recovery of the lead-upgrading circuit

Eliminate the environmental and operational hazards
associated with the filling process of concentrated HCI
into tank trucks
I) Design of an economically feasible separation process
for the removal of sodium chlorate.
2) Develop a system for the complete removal of sodium
hypochlorite from one of the plant's product streams.
Determine the desalter impact on-the effluent unit.
Evaluate the performance of the lead upgrading cells
using a CaCI, tracer.

Summer 1998 205

have been involved as faculty supervisors.

Practice School is one of the first steps our students take in
becoming acquainted with the industrial environment. Al-
though the actual time spent in the plant environment is only
two weeks long, the program ensures that essentially all of
our undergraduate students have some industrial exposure
by the end of their third year. During this two-week period
and the subsequent report-writing stage, students have an
opportunity to put the theoretical knowledge they have gained
and the skills they have developed to practical use. They
build confidence in themselves and in their ability to apply
what they have learned to solve industrial problems. These
Practice School projects ease the students into the senior
year of the chemical engineering program.


One approach to the senior Plant Design course is to make
use of prepackaged (and sometimes outdated) case studies.
The instructor might have twenty such cases to choose from,
and the course is essentially run as any other lecture. But is
this approach in the best interest of our customers, the stu-
dents? Engineers who have gone through a "traditional"
(chemical) engineering education tend to agree that addi-
tional practical components should be added to the curricu-
lum and especially to the senior design project. The expres-
sion "plant design" should immediately imply industrial ap-
plications;[21 consequently, industry should be involved in
the design course. Working with practicing engineers on
relevant industrial problems and not just trying to fulfill their
degree requirements gives students a great sense of achieve-
ment and satisfaction.
The Plant Design course, as our department defines it, is a
capstone course in which our senior undergraduates use their
knowledge to work on an industrially relevant problem.
Over the years, close ties between New Brunswick compa-
nies and our department have been established. Through
personal contacts, the department seeks to obtain industri-
ally relevant design problems for our students, while the
industries are interested in obtaining solutions for their de-
sign problems. If successful, the students and the industrial
partner will benefit from this cooperation. The various steps
involving the successful completion of an industrially spon-
sored design project are described below.

Course Logistics
During initial meetings between the industrial sponsor and
the course instructor, the design project's topic and scope are
established. The company designates an engineer to supply
relevant project information such as process and instrumen-

station diagrams, flow sheets, and plant operating data. Re-
cent design topics supplied by industry have ranged from
"Design of a Sodium Chlorate Plant" to "From PET to the
Soft Drink Bottle."
At the beginning of the term, the class is divided into
design teams (comprised of three to four members each) that
are required to solve different design problems. As the indi-
vidual groups begin working on the design problem, they are
concurrently being introduced to the concepts of group dy-
namics,m31 hierarchical plant design,[41 design of experiments,
and scale-up procedures. The steps taken to arrive at an
optimized solution of a design problem are:

* Formulation of the Project Parameters Pertinent
materials, such as P&ID, flow sheets, and any other
information relating to the design problem, are given
to the students during the first lecture. One reason for
identifying the process specifications is to make sure
that there are no surprises for either the students or
the company upon completion of the project. Where
possible, a plant tour during the second week helps
the students learn more about the plant operation and
assists them in establishing contact with the company

Literature Review and Generation of Alternative
Solutions A thorough search of the existing literature,
as well as application of relevant design and manu-
facturing codes, helps the design teams minimize
design errors and maximize their efforts. Exercises to
encourage the lateral thinking process, the back-
ground of the students in the field of engineering, and
the knowledge gained from the literature search will
yield alternative design solutions.

Development of Gantt Chart The management of
design projects within the constraints of the university
environment, i.e., the heavy course load for students,
is easier said than done. The plant design course
environment is very turbulent and is composed of
numerous design team meetings, section report
writing, conflict resolution, planning, and communi-
cation with the various parties involved. It is ex-
tremely important for the students to learn about time
management. A bar (Gantt) chart developed by each
design group assists everyone in keeping the projects

Development of the Conceptual Design After the
preliminary process flow diagrams have been
developed, an engineering panel comprised of two
practicing engineers reviews the initial conceptual
designs. The panel arrives at a "go/no-go" decision.

Chemical Engineering Education

Certain recommendations may be put
forward by the panel to enhance the
proposed design. If the engineers should
arrive at a no-go decision (which has not
happened yet, perhaps due to the students'
intensive practical preparation through
practical laboratory exercises,"j practice
school, and PEP), the students would be
asked to revise their design and report
back to the engineering panel.

Execution of the Design Calculations and
Assessment of Performance Computer
simulations are an invaluable tool for an
engineer. Simulation programs such as
CMOLD, HYSYS, and FLUENT are used
by the design teams to accomplish their
design tasks more efficiently. Many design
problems, however, still cannot be
adequately simulated. Therefore, experi-
mental work or hand-computations are
frequently necessary to obtain the
required design specifications. The
students must also keep safety aspects
foremost in their minds throughout the
design stage.

Optimum Economic Design In almost
every design case there are several
technically equivalent design solutions.
The optimum size can often be obtained
only by cost comparison of different
schemes. There is a tendency to evaluate a
parameter as being the optimum factor.
This sub-optimization may result in an
increase in the overall process equipment
cost. Care must be taken to study the
overall system and its economic implica-
tions. The students are required to
evaluate their design solution from an
economic perspective.

Preparation of Specifications The use of
specification sheets is mandatory, but due
to time constraints, they need not be fully
completed. These sheets are the basis for
the company in the continuation of the
design projects.

Ultimately, a final report summarizing the
work done on the project must be submitted.
The report's content follows a format similar
to that described by Seider and Kivnick.51






alleviate the




by a




accepting a

position in


The Plant



presents an



for students

to learn

more about

the language

and customs



The individual groups are then required to
present their final design at the Annual Indus-
trial Chemical Engineering Conference. The
conference, which is organized by the chemi-
cal engineering student body, attracts attend-
ees from the university, industry, and govern-
ment. It is intended to hone the presentation/
public speaking skills of the students, to present
the design solutions to the participating com-
panies, to allow the graduating students to "rub
shoulders" with decision makers and perhaps
to discuss employment opportunities, and to
develop new university/industry relationships
as well as bolster existing ones.

While this plant-design concept has been suc-
cessful, it should be pointed out that the notion
of an industrially oriented plant design course
is not new. Articles by Seider and Kivnick151
and Rockstraw, et al.,16' have discussed both
this concept and their affiliations with local
companies. Our effort is different because,
while New Brunswick is a highly industrial-
ized province, there is no single industry in
close proximity to the University with which
we have been able to develop a close, continu-
ous relationship. Rather, projects come from
various companies each year.
It is interesting to note that our involvement
with industry through the Practice School has
led to projects for Plant Design. The most re-
cent example is the sodium chlorate plant-de-
sign project mentioned above, which grew out
of the sodium chlorate filter and sodium hy-
pochlorite removal Practice School projects.
Furthermore, several national and international
companies currently support our effort to edu-
cate students in a more practice-oriented way.
Information technologies have made it pos-
sible for the students to work on design prob-
lems that are not available in our local area.


Students enjoy working on an industrial de-
sign project and feel pride in that a part of their
work will be implemented in an industrial op-
eration. When this course concept was first
introduced, some students commented:
"I can't design a plant addition for a com-
pany; I don't have enough experience to un-

Summer 1998

dertake this task."
"Is this for real?"
After the students had time to think about the underlying
objectives, the comments were more along the line of:
"Finally I am not working for the wastebasket any-
"Thank you for introducing us to the world of practical

Although the student comments have been positive, there
is always room for improvement. In industry, chemical engi-
neers seldom work on a project by themselves; different
disciplines are needed to ensure the successful completion of
a project. There will be electrical engineers, mechanical
engineers, civil engineers, chemical engineers, architects,
and business people working side-by-side.
In the 1997-98 academic year, mechanical engineering
students will also participate in the Plant Design course,
creating a more realistic design environment. A design project
involving the local aquaculture industry was carried out in
the fall of 1997. It involved interactions with the biology
department and was quite successful.
Expanding the design course to involve other disciplines,
however, creates some problems that need to be addressed.
Of particular concern is the issue of prerequisites. For ex-
ample, mechanical engineering students do not take courses
dealing with separation processes or reactor design, which
are both Plant Design prerequisites for our students. During
the course of a design project, process economics and safety
and environmental concerns have to be considered simulta-
neously. Having recognized that these subjects are insepa-
rable, starting in 1998-99, we will introduce a new full-year
design course to more adequately cover them. The new
course will represent the amalgamation of the Plant Design
and Process Economics and Safety courses.
Proper planning and preparation will alleviate the "culture
shock" experienced by a graduate engineer when accepting a
position in industry. The Plant Design course presents an
excellent opportunity for students to learn more about the
language and customs of professional engineers. Our hope
is that our students' transition to industry will be consid-
erably eased.

Weil71 observed that two paradigms have shaped the field
of chemical engineering throughout this century. In the 1920s,
the first paradigm of unit operations was developed. Then,
with the publication of Transport Phenomena, by Bird,
Stewart, and Lightfoott81 in 1960, came the more analytical
and fundamental second paradigm. In a recent article,
Landau[ 9 states,

"It is my opinion ... that we need a change in direc-
tion toward more relation to practice and to industry,
which perhaps might constitute the third paradigm."
Elsewhere in the article, Landau comments,
".. I believe chemical engineering's third paradigm,
if there is one, is to return the discipline closer to the
practices in industry, and to strengthen its interdisci-
plinary ties ... "
While the first two paradigms were indeed revolutionary,
the next paradigm should not be. As discussed by Dou-
"As we extend chemical engineering into new applica-
tion areas, we will need experts in each of these para-
digms. "
Whether or not one accepts the notion of an emerging third
paradigm, we do feel that our Practice School and Plant
Design courses are effective approaches to preparing chemi-
cal engineers for their future in the profession. We empha-
size practical, industrially oriented projects and are promot-
ing an interdisciplinary structure. The students are also firmly
grounded in the fundamentals of chemical engineering and
the basic sciences. The positive feedback obtained thus far
from the students encourages us to continue thinking of new
and innovative approaches for improving the existing courses.

The authors wish to thank Robin Chaplin and Derek Lister
for their invaluable suggestions during the development stage
of the plant-design course. Also, Jules Picot provided us
with his first-hand recollection of the early days of the Prac-
tice School. The most recent Practice School projects would
not have been possible without the coordination efforts of
Mladen Eic.

1. Chaplin, R.A., "Providing Industrial Experience in a Regu-
lar Laboratory Course," Chem. Eng. Ed., 31(2), 130 (1997)
2. Peters, M.S., and K.D. Timmerhaus, Plant Design and Eco-
nomics for Chemical Engineering, McGraw-Hill, New York,
NY (1991)
3. Crosby, P.B., Problem Solving, (1991)
4. Douglas, D.M., Conceptual Design of Chemical Processes,
McGraw-Hill, New York, NY (1988)
5. Seider, W.D., and A. Kivnick, "Process Design Curriculum
at PENN," Chem. Eng. Ed., 28(2), 92 (1994)
6. Rockstraw, D.A., J. Eakman, N. Nabours, and S. Bellner,
"An Integrated Course and Design Project," Chem. Eng.
Ed., 31(2), 94 (1997)
7. Wei, J., "Future Directions of Chemical Engineering," in
Advances in Chemical Engineering, 16, 51 (1991)
8. Bird, B.R., W.E. Stewart, and E.N. Lightfoot, Transport
Phenomena, John Wiley and Sons, New York, NY (1960)
9. Landau, R., "Education: Moving from Chemistry to Chemi-
cal Engineering and Beyond," Chem. Eng. Prog., 52, Jan.
10. Douglas, J.M., "The Paradigm After Next," in Advances in
Chemical Engineering, 16, 535 (1991) 0
Chemical Engineering Education

Experiment in Mass Transfer
Continued from page 201.
final data"51 on the effect of Ug on E, with and without baffle
plates. Also shown for comparison on Figure 3 is an equa-
tion that has been suggested in the literature16' for axial
dispersion in unbaffled bubble column,

E = 0.35(gug)d43 (22)
We can see that the unbaffled column data (open circles) are
in the same order of magnitude as predicted by Eq. (22),
showing the same trend of a slight increase in E with respect
to gas velocity. In the presence of the baffles, E is greatly
reduced and the values go through a shallow minimum with
respect to ug. The tendency of E to increase at the lowest values
of gas velocity is thought to be due to the formation of trains of
gas bubbles following a preferred path up one side of the
baffles, resulting in a downflow of liquid on the other side.

This experiment can be carried out in as little as one three-
hour laboratory period, although two such periods are better
for a thorough comparison between the results with and
without baffles. The experimental work is not highly de-
manding, although students must be careful to measure solu-
tion volumes accurately, and necessary safety precautions
must be followed. It is preferable that prepared standard
1.000 mol/L solutions of acid and base be used; these are
available in 20 L quantities from most scientific suppliers.
Students should be encouraged to use computers to exam-
ine the time-dependent concentration profiles in the column
by means of Eq. (11). Numerical examples can also be
developed to confirm the stated assumption leading to Eq.
(13), namely that only the first term in the series is impor-
tant when a> 0.1.
As a further exercise, the eddies of color at the neutraliza-
tion zone can be videotaped in close-up mode. Playing these

ISuperficial gas velocity, u,, cm/s

Figure 3. Axial dispersion coefficients plotted versus ug,
with and without baffles. Dashed line denotes Eq. (22)

Summer 1998

back at slow speed provides a good visualization of the
highly random nature of turbulent axial dispersion.
Our experience over a period of about eight years has
shown that this experiment provides students with good
insight into axial mixing. The present method offers two
advantages over more classical "trace injection," such as salt
or dye. First, the complexities of sampling/analysis of the
tracer are diminished, allowing more data to be obtained in a
given period, and second, the sharp color change boundary
allows students to directly observe the mixing process.

The experimental facilities were provided by the teaching
budget of the Department of Chemical Engineering at
McMaster University. The authors are also grateful to the
Natural Sciences and Engineering Research Council of
Canada for support of the costs of producing this paper.

c concentration, mol m3 or mol/L
c' concentration variable defined by Eqs (9) and (10),
mol m-3 or mol/L
d internal diameter of column, m
E axial dispersion coefficient m'rs
g acceleration due to gravity, ms'2 (=9.81)
L depth of gas/liquid mixture, m
L, depth of ungassed liquid, m
n term in expansion series
N flux, mol m s'
Pe Peclet number
R ratio (mols acid added) to (mols base added)
t time, s
t' time for neutralization at z=L, s
U superficial liquid velocity, m s'
u, superficial gas velocity, m s'
z axial distance, m (measured downward, see Figure 1)
Greek Symbols
a dimensionless time, see Eq. (12)
e gas holdup (fraction)
a free acid
b free base
1 initial value
Sas t o
r reaction zone
1. Deckwer, W.-D., Bubble Column Reactors, John Wiley &
Sons Inc., New York, NY (1992)
2. Treybal, R.E., Mass Transfer Operations, 3rd ed., McGraw-
Hill, New York, NY, p. 210 (1980)
3. Levenspiel, O., Chemical Reaction Engineering, 2nd ed.,
John Wiley & Sons Inc., New York, NY, Ch. 9 (1972)
4. Danckwerts, P.V., Gas-Liquid Reactions, McGraw-Hill, New
York, NY, p. 111 (1970)
5. Armstrong, M., "Mixing in a Bubble Column," Undergradu-
ate Laboratory Report, McMaster University, March 19,
6. Baird, M.H.I., and R.G. Rice, "Axial Dispersion in Large
Unbaffled Columns," Chem. Eng. J., 9, 171 (1975) O

100 i

o Baffles


40 F

Baffles Present

j, R"classroom




B. Wayne Bequette, Kevin D. Schott, Vinay Prasad, Venkatesh Natarajan, Ramesh R. Rao
Rensselaer Polytechnic Institute Troy, NY 12180-3590

In recent years, there has been an on-going argument that
engineering students need more open-ended problems,
more team projects, more written memos, reports, and
oral presentations, more practical problems, and more inter-
active learning (rather than complete dissemination of course
material by lecturing) in their undergraduate education.
At Rensselaer Polytechnic Institute we try to include all
these factors in our process control course. The focus of this
paper is on a case study project performed during the latter
half of the semester. We will cover not only the project, but
also a background description of the Rensselaer curriculum,
the introductory material for the project, the venues used for
distributing course material, future teaching efforts, and a

There are roughly 80-90 BS degrees granted in chemical
engineering at Renssealer each year, while there are ap-
proximately 30 environmental BS degrees granted. A num-

Ramesh R. Rao received a BE in chemical engineering from Annamalai Univer-
sity and a Masters from the Indian Institute of Technology, Kanpur. Before joining
Rensselaer Polytechnic Institute to pursue a PhD, he spent two years as a
research engineer at the Tata Research Design & Development Centre, Pune.
His current research interests focus on model predictive control applications in
chemical and biomedical engineering. (Far left)
Venkatesh Natarajan received his Bachelor of Technology from the Indian
Institute of Technology, Bombay, and his MS from Rensselaer Polytechnic Insti-
tute, both in chemical engineering. He is pursuing a PhD at Rensselaer, where his
research involves the scale-up and optimization of ion-exchange
biochromatographic systems. (Second from left)
Vinay Prasad obtained a Bachelor of Technology from the Indian Institute of
Technology, Bombay, and an MS from Kansas State University, both in chemical
engineering. He is now pursuing his PhD at Rensselaer, and his research inter-
ests lie in process design, analysis, and control, with a focus on batch chemical
processes. (Second from right)
B. Wayne Bequette is Associate Professor of chemical engineering at Rensselaer
Polytechnic Institute. He received his BS from the University of Arkansas and
spent several years as a process engineer at American Petrofina before obtain-
ing his PhD from the University of Texas, Austin. His teaching and research
interests are in process control, design, and biomedical engineering. (Right)

ber of courses in the curriculum (material and energy bal-
ances, dynamic systems, chemical process control, unit op-
erations laboratory I and II) are taught to both the chemical
and the environmental students.
A distinguishing characteristic of the Rensselaer curricu-
lum (in addition to the fact that both chemical and environ-
mental engineering take many of the same courses) is that
we have had separate courses in dynamics and control for
over a decade. Another is that the dynamics and control
courses are taught during the junior year. One advantage of
teaching these courses at that time is that students tend to
take more of a process-systems engineering viewpoint in the
senior courses (reactor design, separations processes, pro-
cess design, lab I and II).
The process dynamics course covers more material in
more depth than the "front end" of a typical single course in
dynamics and control. Particular emphasis is given to nu-
merical methods for the solution of algebraic and differential
equations, with MATLAB as the numerical analysis pack-

Kevin D. Schott received his BS in chemical engineering from the Uni-
versity of Massachusetts and his MS in electrical engineering from
Rensselaer Polytechnic Institute. He has several years of experience with
Monsanto and is completing his PhD at Rensselaer. His research in-
volves multiple-model and gain-scheduling approaches to nonlinear con-
trol (Not pictured)

Copyright ChE Division of ASEE 1998


Chemical Engineering Education

age. State space models receive much attention. Also, phase-
plane analysis and an introduction to nonlinear dynamics
and chaos is provided. A textbook for this course has been
published by Prentice Hall.1'21

A major advantage of the two-course sequence in process
dynamics and control is the proper coverage that can be
given to the topic of process control. Process control is
taught in the spring of the junior year, immediately after the
students have taken process dynamics. After a concise re-
view of modeling and dynamics, we are able to leap into
important issues of control-system design. The topics cov-
ered in the course are
Review of process modeling for control
Introduction to feedback control
Direct synthesis and open-loop "control"
Internal model control (IMC)
IMC-based PID control

The process dynamics course covers more
material in more depth than the "front end" of a
typical single course in dynamics and control.

Introduction to frequency response
Frequency response for control-system design
Control using multiple measurements
Implementation issues
Decentralized control
Multivariable control
Case-study problems in multivariable control

Each of the topics, including characteristic homework prob-
lems, is summarized in Table 1.
When we first taught this course, we also introduced dis-
crete control-system design, and in some years we included
model-predictive control (MPC). Our philosophy now is to
cover less material, but to provide more depth. Since our
course is taught during the junior year, students often select
a senior project in control or take a reading course in ad-

Process Control Course Topics

Motivation and Introduction to Control Fairly standard material on
economic, safety, and environmental incentives is presented. Simple
examples, such as "taking a shower" and surge drum level control are
discussed extensively in the lectures. Issues include objectives, mea-
surements, manipulated inputs, disturbance inputs, continuous vs. batch
(and semi-batch) and feedforward/feedback. As a homework problem,
the students select a favorite activity and analyze it in detail from a
control perspective.
Review of Process Modeling for Control This section is much shorter
than in a standard course since the students completed a dynamic-
systems course during the previous semester. Both fundamental and
input/output models (obtained by step tests) are reviewed. An example
homework problem is to develop a nonlinear model for a series of gas
surge drums. The students form a state-space model via linearization,
find transfer functions, and simulate the open-loop system using
Classical Feedback Control (PID) The concept of PID control (in
various forms) is presented, and the effect of the tuning parameters is
discussed and illustrated by example. Traditional methods such as
Cohen-Coon and closed-loop Ziegler-Nichols are covered. A typical
homework assignment is a continuation of the previous modeling and
simulation assignment (gas drums, for example), again using SIMULINK
for closed-loop simulation. The students are encouraged to explore the
robustness of their control-system designs.
Direct Synthesis and Open-Loop "Control" One issue stressed in this
section is that, because of inherent performance limitations (right-half-
plane zeros, time-delays), one cannot arbitrarily select any desired
closed-loop response and yield a physically realizable (or internally
stable) controller. We show how the open-loop control system design
approach evolves to the internal model control structure when one
accounts for disturbances and model uncertainty.
Internal Model Control (IMC) The IMC procedure is a major focus of
the course and distinguishes the course text from other undergraduate

texts. Factorization of the model, inversion of the "invertible" portion
of the model to form the ideal controller, addition of a filterfor realiz-
ability, and tuning for robustness are all covered.
IMC-Based PID Control We show how to rearrange the IMC structure
to the standard feedback structure, often resulting in a PID algorithm.
The design procedure for open-loop unstable systems is also detailed.
The control of a biochemical reactor at an open-loop unstable point is
used as a homework problem.
Frequency Response for Control-System Design One of the main
motivations for covering frequency response is that gain-margin and
phase-margin concepts lead to a better understanding of robust con-
trol-system tuning. A typical homework problem involves steam drum
level control.
Control Using Multiple Measurements Here we introducefeedforward
and cascade-control design. Again, steam drum level control is often
used for the homework problem. At this point in the semester, the
students usually have a week with a lighter load because of student
government elections. During this week we normally take a tour of the
campus boiler house, pointing out the various control loops; it is clear
to the students that an operator would not be able to operate the
boilerhouse without feedback control.
Implementation Issues Important practical issues, such as variable
scaling, proportional gain, installed valve characteristics, are covered
in this section of the course.
Decentralized Control The relative gain array is introduced as a tool
to help select variable pairings for decentralized multivariable control
structures; distillation control problems of various sizes are used as
illustrative examples. Students implement these techniques in their
case studies.
Multivariable Control There is little time to provide detailed treatment
for full multivariable control design. Usually, static and dynamic
decoupling are covered; sometimes, multivariable IMC is also covered.

Summer 1998 21'

vanced control where we cover digital control and MPC.

A number of excellent process-control textbooks are currently available,
but each lacks some of the features of the soon-to-be-published text,131 such as
In-depth coverage of the IMC procedure
Connections between open-loop design and IMC
IMC design for open-loop unstable systems
The course enrollment is typically large (roughly 120 students), which
somewhat limits the type of faculty/student interaction that can occur in
lectures, although we try to motivate as much discussion as possible by
using a "Phil Donahue" approach. There are three 50-minute lectures per
week, a 50-minute recitation by a teaching assistant, and a weekly com-
puter lab for solving homework problems. Lectures contain a mix of
analytical derivations and simulation results (the lecture hall is equipped
with a workstation, a PC, a VCR, a computer/video projector, and two
overhead transparency projectors).
Homework assignments are given weekly and are solved in groups of
three; typical assignments are discussed in Table 1. The students are
expected to provide a one-page written memo summarizing the results of
the assignment. Working in groups improves the students' interaction
skills and enables more complex problems to be solved, and providing a
written memo improves their communication skills. The homework as-
signments constitute 30% of the course grade. We place a high weighting
on the homework assignments because we feel that students learn more
about dynamics and control through interactive simulations (combined
with analytical solutions) than analytical solutions alone.
MATLAB is the software package used for numerical analysis and
simulation. The students have been introduced to MATLAB in the process-
dynamics course, and it is sometimes used in the chemical engineering
thermodynamics course. One of our first assignments directs the student to
complete a tutorial review of MATLAB.
Rensselaer has an extensive network of roughly 500 workstations (IBM
RS6000 and Sun SparcStations), with a site license for MATLAB/
SIMULINK (as well as many other packages). We reserve a computer lab
with thirty workstations for three nights a week. The lab is staffed by the
instructor or a teaching assistant.
The course homepage is used as an additional venue for distributing mate-
rial. Summaries of lecture notes, practice problems for exams, and tutorial
modules in hypertext form, are made available on the course homepage. It can
be found by linking to "Courses" from

Control Case Sudies

Mixing Tank (tutorial example) Evaporator
Dowtherm Heater Solution Copolymerization
Reactive Ion Etcher Fluidized Catalytic Cracking Unit
Drug Infusion System Wet Grinding Circuit
Rotary Lime Kilm Anaerobic Sludge Digester

oage Bns owe Thttle

G.,,, In n Oaut

Reactive Ion Etcher

Drig Infusion Control

Rcr Tempcrailure
Damper i From "I impcnrure
Poslon -. TC24--2 Cnonlolcr
Y2 r TCI FueCla Fi
F eb s I uI

Rotary Lime Kiln

R-I--- Reel ,p ..

Fluidized Catalytic Cracking Unit (FCCU)

disgested sludge
Anaerobic Sludge Digester

Figure 1. Case study instrumentation

Chemical Engineering Education

The projects are more open-ended than typical undergraduate assignments, provide more experience
working in a group environment, and further develop written and oral presentation skills ....
The case studies give [the students] the opportunity to "tie it all together" and to
understand each component of a control-system design project.

An electronic newsgroup is used to answer common ques-
tions or to post notes about the lecture material. Rather than
responding to many individual e-mail questions, a single
posting to the newsgroup saves faculty and TA time. Also, it
gives the students the capability of posing questions that can
be answered by other students (this feature is not used as
often as we would like).
The written examinations are fairly standard. Three one-
hour exams (45% of the course grade) and a three-hour
comprehensive final exam (25% of the course grade) are
required. It would be nice to give some exams on the com-

References for Five 1997 Case Studies

General references for reactive ion etcher (suggested to the students)
Bagwell, T.A., T. Breedijk, S.G. Bushman, S.W. Butler, S.
Chatterjee, T.F. Edgar, A.J. Toprac, and I. Trachtenberg,
"Modeling and Control of Microelectronics Materials
Processing," Comp. Chem. Eng., 19(1), 1 (1995)
Lee, H.H. Fundamentals ofMicroelectronics Processing,
McGraw-Hill, New York, NY (1990)
Sze, S.M., VLSI Technology, McGraw-Hill, New York, NY
Wolf, S., and R.N. Tauber, Silicon Processing fr the VLSI Era,
Lattice Press, Sunset Beach, CA (1986)

The model we use is modified from
Rashap, B.A., M. Elta, H. Etemad, J.P. Foumier, J.S.
Freudenberg, M.D. Giles, J.W. Grizzle, P.T. Kabamba, P.P.
Khargonekar, S. Lafortune, J.R. Moyne, D. Teneketzis, and
F.L. Terry, "Control of Semiconductor Manufacturing
Equipment: Real-Time Feedback Control of a Reactive Ion
Etcher," IEEE Trans. Semicond. Manuf 8(3), 286 (1995)

Models for drug infusion, lime kiln. FCCU. and an anerobic digester
are presented in
Yu, C.L., R.J. Roy, H. Kaufman, and B.W. Bequette, "Multiple-
Model Adaptive Predictive Control of Mean Arterial Pressure
and Cardiac Output," IEEE Trans. Biomed. Eng., 39(8), 765
Charos, G.N., Y. Arkun, and R.A. Taylor, "Model Predictive
Control of an Industrial Lime Kiln," Tappi J., 203, February
Hovd, M., and S. Skogestad, "Procedure for Regulatory Control
Structure Selection with Application to the FCC Process,"
AIChEJ., 39(12), 1938 (1993)
Alatiqi, I.M., A.A. Dadkhah, A.M. Akbar, and M.F. Hamouda,
"Comparison Between Dynamics and Control Performance of
Mesophilic and Thermophilic Anaerobic Sludge Digesters,"
Chem. Eng. J., 55, B55 (1994)

Summer 1998

puter, but thus far it has been too big of a challenge to
organize for 120 students.

In a typical semester, for the final course project we allow
the students (in groups of three) to select from at least five
different case studies on multivariable control. A breadth of
applications are covered, from biomedical to classical chemi-
cal processes (see Table 2). Since the students are allowed to
select from a variety of problems, they are more motivated and
able to attach physical significance to the problem they study.
During the most recent offering of the course, we decided to
revise the case-study concept, placing more emphasis on it.
During the last half of the (spring 1997) process control
course, the students worked in three-person teams on multi-
variable control projects that they selected from a choice of
five systems: reactive ion etcher; drug infusion; rotary lime
kiln; fluidized catalytic cracking unit (FCCU); and anaero-
bic sludge digester. Control diagrams for each of these case
studies are shown in Figure 1. Each project was advised by a
different member of an instructional team.
The students were given a brief description of each project.
They selected their own teams of three students each and
chose a project (project advisors were not designated until
after the groups were selected). Each project included many
phases typically associated with a control design project:
literature review, model development and process identifica-
tion, control structure selection and controller tuning for
SISO systems, multiple SISO loop tuning, and decoupling.
This approach gives the undergraduate student a sense of
what an industrial control problem involves, including work-
ing in a project-team environment with a project advisor. It
also gives the graduate students and teaching assistants ex-
perience in advising and teaching and reinforces many con-
trol-system concepts.
To illustrate the case study, we will use the reactive ion
etcher example. The control diagram is shown in Figure 1
and suggested references are in Table 3. Descriptions of all
case studies can be obtained by linking to "Educational
Material" at the homepage found at

C Literature Review (1 week)
Students are given a brief description, with control instru-
mentation diagrams, for each of the projects. They form
groups of three and perform a concise literature review to

provide background material on the unit operation of interest
and the industry where this process is dominant. They write a
concise memo, which is evaluated by the project advisor.

C Model Development (1 week)
A SIMULINK file, developed by the project advisor, is pro-
vided for each group. The open-loop diagram for the reactive ion
etcher is shown in Figure 2. The actual model for the etcher is
shown, in unmasked form, in Figure 3. Notice that constraints,
time-delays, and noise are included. To develop a model that will
be used for control system design, the students perform open-
loop step tests. Example results are shown in Figure 4.
The groups provide a short memo (with plots and transfer
functions attached), summarizing the modeling results. The
advisor evaluates the memo and makes suggestions for addi-
tional modeling studies, if necessary.
C SISO Controller Design (1 week)
In this phase, the groups perform in- 000
dependent SISO control design, usu- w,ss power disturbar
ally pairing the loops based on physical
considerations. They use one or more power i Su power
of the techniques covered in the course constrain
(IMC-based PID is the most popular
choice). The groups prepare a short writ-
ten report describing their results. It is 2 1
important here that the project advisor throttle Sum4 alve l
catch obvious mistakes before the v 50
groups close both loops simultaneously. Throttle.
C MV-SISO Controller Design (1
Here the groups use the relative gain
array (RGA) to gain insight about vari-
able pairing and how independently de- Flow easureent
signed loops need to be retuned when
both loops are closed. Failure sensitiv-
ity is considered very important in this
phase; if one loop fails (is opened or saturates), the other loop
should not go unstable. Advisor comments on the memo report
assist the groups in preparing the final written report.
C Final Written Report (1 week)
This is a formal written report with the structure of a typical
technical paper. Much of the material can be gathered (with
some rewriting) from previously written memo reports. Most
groups also take the time to perform "full" multivariable con-
trol studies, such as static and/or dynamic decoupling.
C Oral Presentation (1/2 week)
Each group prepares a fifteen-minute oral presentation (plus
five minutes for questions) that is evaluated by the project
advisor and at least one other evaluator. This gives the students
a chance to enhance their oral presentation skills. Also, it is
much easier for the project advisor to see what the students
really learned from the experience and to provide immediate

Clock time voltage
Voltage, V

power, 1 -- fluorine
Fluorine, % of range
throttle, % flow
-D- Flow, % of range
power disturbance, W

Figure 2. SIMULINK diagram for open-loop tests.

Figure 3. Etcher unmasked.



0 10 20
t, sec

44 u1-y2
0 10 20
t, sec



0 10 20
t, sec

t, sec

Figure 4. Step responses.

Chemical Engineering Education

We found that many students have no idea how to perform
a literature review. Often, an internet search was conducted
using a web-crawler (Alta Vista or a similar program). Ap-
proximately one-half of the literature reviews consisted of a
rambling essay about motivation or previous work, with no
specific citations of the literature. We asked a number of
groups to revise their literature review.
Clearly, our case studies in multivariable control require a
lot of effort and coordination of all members of the instruc-
tional team. It is important to have a robust simulation set-up
for the students to perform their initial identification tests. It is
also important to provide rapid feedback. Groups generally
turned in their memo reports on Friday, and we usually evalu-
ated them and returned them to the students on Monday.
Comments from the undergraduate students have gener-
ally been favorable. The case studies give them the opportu-
nity to "tie it all together" and to understand each component
of a control-system design project. It should also be noted
that the role of the case-study advisors shifts during the
projects, ranging at various times from boss to intelligent co-
worker to all-knowing judge and inquisitor.

Currently, the control course has been taught in a fairly
traditional lecture/recitation/computer-lab format, with three
lectures and one recitation per week. The recitation typically
covers the assignment for that week or reviews a recent
exam. Students are also expected to participate in one com-
puter laboratory session per week.
There is a move in the Rensselaer curriculum toward
"studio" or "workshop" learning, where students meet twice
a week for two-hour sessions with a faculty member and one
or two teaching assistants. The idea is for the students to
learn interactively by solving problems rather than by pas-
sively listening to lectures. Rensselaer is currently renovat-
ing or constructing a large number of classrooms to fit the
studio format, with student workstations (not just comput-
ers) where students can interact and solve problems in groups.
The instructor or teaching assistant can give "mini-lectures"
as groups encounter common stumbling blocks or can pro-
vide more background material as needed.
Since the dynamics and control sequence is taught during
the junior year, it offers an excellent opportunity to consider
process control implications in the process-design course.
We plan to do this as process-flowsheeting packages begin
to have dynamic extensions that are relatively easy to use.

We have presented an approach to using case-study projects
in a process control course. The projects are more open-
ended than typical undergraduate assignments, provide more
experience working in a group environment, and further
Summer 1998

develop written and oral presentation skills. In addition to
the learning experience for the undergraduates, we have
found that the teaching assistants, the graduate students, and
the instructor also learn from the approach.

1. Bequette, B.W., "An Undergraduate Course in Process Dy-
namics," Comp. Chem. Eng., 21(Suppl), S261 (1997)
2. Bequette, B.W., Process Dynamics: Modeling Analysis and
Simulation, Prentice Hall, Upper Saddle River, NJ (1998)
3. Bequette, B.W., An Introduction to Model-Based Control,
Prentice Hall (in press, 1999) O

BOOK REVIEW: Mathematical Methods
Continued from page 189
takes a fairly classical approach, and the authors dig into first-order
partial differential equations in Chapter 6 with relish. They offer a
particularly thorough treatment of the subject replete with examples
of waves, shocks, and weak solutions. This is obviously a favorite
topic of the authors, and many chemical engineers dealing with
packed bed or chromatographic separations will find meaty bones
to chew on in Chapter 6.
Fourier and Hankel transforms are covered in Chapter 7, but the
applications to the vibrating circular membrane and semi-infinite
strips and cylinders are not particularly stimulating for the chemi-
cal engineer. The applications of Laplace transforms in Chapter 8
are probably of greater relevance to chemical engineers.
Although the references at the end of each chapter are not exten-
sive, they are well thought out and direct the interested reader to
more comprehensive treatments of the subjects. The variety of
mathematical tools useful to chemical engineers is reasonably well
covered, and the authors point out that they felt it necessary to
exclude complex variables, statistics, and numerical methods. It would
have been reasonable to include a short summary of similarity analy-
sis because similarity solutions are so often encountered in fluid
mechanics and heat and mass transport processes. An instructor may
wish to supplement the book with examples of similarity analysis.
Some of the chapters are beyond the abilities of many under-
graduates, but chemical engineering graduate students would profit
greatly by working through the entire nine chapters. I plan to use this
text in my graduate course in mathematical methods applied to chemi-
cal engineering, replacing Hildebrand's widely used book Advanced
Calculus for Applications, because it is necessary to supplement
Hildebrand's book with chemical engineering applications. Varma
and Morbidelli do this well and at a cost that is reasonable.
One finds that the book has been carefully proofread, for it is
difficult to find typographical errors. The figures are simple and
uncluttered, and they are entirely adequate to illustrate the relevant
mathematics. There is a good set of problems at the end of each
chapter, and many chemical engineering applications are incorpo-
rated in these problems.
The rigor and sophistication of this book go well beyond the few
competing texts that claim to be advanced mathematics for chemi-
cal engineers, and I can add my humble imprimatur to those of
Professors Amundson and Aris who encouraged the authors to
write this book. 1

M= classroom




University of Illinois at Urbana-Champaign Urbana, IL 61801-3792

Providing fast and smooth transitions during discrete
process changes is of high industrial importance. For
example, in polymerization, production runs for a
particular polymer are typically of limited duration, and the
reactor conditions must be modified to produce a different
grade or type of polymer. Another type of discrete process
change occurs when a controller output saturates. Reset
windup is said to occur when the controller continues to
integrate the error signal during saturation, causing large
overshoots and oscillations.
Discrete process changes also occur during split-range
control, in which different manipulated variables become
active in different operating regimes. Split-range control is
useful when more than one manipulated variable is required
to span the whole range of setpoints.1'3"] Controllers that
provide smooth transitions during discrete process changes
are said to provide bumpless transfer.
Although industrial control systems must be designed to

Richard Braatz received his BS from Oregon
State University and his MS and PhD from the
California Institute of Technology. After a
postdoctoral year at DuPont, he became an
assistant professor of chemical engineering at
the University of Illinois. His main research in-
terests are in the modeling and control of com-
plex systems.

Serena Chung received her BS in chemical en-
gineering from the University of Illinois, Ur-
bana-Champaign in 1998 and plans to pursue a
PhD in chemical engineering. Her current re-
search interests are in the modeling and control
of crystallization.

@ Copyright ChE Division of ASEE 1998

... students are rarely taught in their
undergraduate process control courses how to
address such [discrete process change]
problems. This paper serves to close
this [educational] gap...

handle such discrete process changes, students are rarely
taught in their undergraduate process control courses how to
address such problems. This paper serves to close this gap in
the education of undergraduate chemical engineers. The pa-
per is distributed as a reading assignment to students in the
undergraduate chemical process control course at the Uni-
versity of Illinois, and the material is discussed during the
lecture immediately after the students have covered
feedforward, ratio, and cascade control.
First, the students are introduced to the process model of a
laboratory apparatus for collecting permeation data for thin
polymer films.141 The temperature must be controlled to very
high accuracy for the apparatus to provide accurate measure-
ments of diffusion coefficients. The students are taught to
control the process using multiple digital Internal-Model-
Control-Based Proportional Integral Derivative (IMC-Based
PID) controllers in the velocity form.
For homework, the students are required to derive the
control algorithm, to simulate closed-loop responses using
different controller-tuning parameters, and to propose and
discuss potential improvements (see Table 1 for the home-
work assignment). The students are also required to compare
the closed-loop response of their best controller to that ob-
tained by a default controller that was implemented on the
real apparatus.51 Although this paper focuses on IMC-Based
PID control, the homework assignment can be readily modi-
fied to teach other controller design methods.

Chemical Engineering Education

The advantages of using this control problem for training
students are

The process dynamics and performance specifica-
tions are based on a real system.
The two operating conditions are substantially
different (there are significant changes in time delay,
gain, and time constant).
A practical control algorithm is provided that can be
easily implemented in a process control laboratory
or in industry.
A MATLAB program simulating the process and the
classical control algorithm is available via the
Internet. 61


Precise control of the temperature, T, of a sample contain-
ing a thin polymer film is required to provide accurate mea-
surements of the diffusion coefficient.'41 The manipulated
variable is the power to a heating tape that surrounds the
polymer sample. Heat sinks allow the temperature of the
sample to be reduced quickly. For temperatures below 300C,
the heat sink is distilled water, and for higher temperatures,
the heat sink is gaseous nitrogen. The advantage of the
gaseous-nitrogen heat sink over the liquid-water sink is that

Homework Assignment
Antiwindup, Bumpless Transfer, and Split-Range
(The textbook referred to in Problem 2 is Ref [1],
Handout A is Ref [5], Handout B is this paper.)

Problem 1. In Handout B, derive Equations (7) and (8)
from Equation (6).
Problem 2. Draw a block diagram for the split-range
control problem described on page 583 of the textbook.
Describe the differences and similarities between this
control problem and the control problem described in
Handout B.
Problem 3. Use Netscape to download the MATLAB
program bump.m from
software. This program implements the two IMC-based PID
controllers and the process models described in Handout B.
Implement several values of the IMC tuning parameter k,
and select the X that gives the best overall performance.
Justify your selection.
Problem 4. Implement the IMC-tuning parameters listed
in Equation (20) of Handout A in the MATLAB program.
Select ) to give the best overall performance. Compare the
performance for this controller with the best performance
reported in Problem 3. List conditions for which the IMC
tuning parameters in Handout A are expected to provide
superior performance. Propose and discuss modifications
that could improve your best control algorithm.
Problem 5. Compare your best closed-loop response to
that in Figure 1 of Handout B. Discuss.

Summer 1998

it allows a wide range of temperatures to be covered by only
manipulating the heating power. The distilled-water sink
provides a more stable response for temperatures under 30C.
The heat sinks are at room temperature, which is approxi-
mately 21C with slow variations up to 1IC. For each heat
sink, temperature responses to step changes in heating power
were taken at a variety of operating conditions along the
desired temperature trajectory in order to estimate the im-
portance of nonlinearity. The process responses were linear
for each heat sink, with the transfer functions given by

1.0 e-2.4s
pl(s)=- (1)
9.5s +1

for the gaseous-nitrogen heat sink (T > 30C), and

0.068 e-'4s
2(s) = (2)
1.7 s + 1

for the liquid-water heat sink (T < 300C), where the time
constants, ti, and time delays, e,, are measured in minutes,
and the process gains, K,, are measured in C/% power. The
heating power is constrained between 0 and 100%. At steady
state, the sample is at room temperature when the heating
power is turned off.

The goal of the closed-loop system is to smoothly ramp
the temperature from stable operations at 120C to 250C (see
Figure la). For reproducible collection of diffusion data, the
temperature must stay within 0.50C of the setpoint 70 min
utes before and 50 minutes after the ramp, and within 1.5C
throughout the ramp. The control algorithm must provide
bumpless transfer between the radically different process



0 200 400 600 800
Time (minutes)




0 200 400 600 800
Time (minutes)

2 80
a 60

200 400 600 800
Time (minutes)

0 200 400 600
Time (minutes)

Figure 1. (a) Setpoint; (b) closed-loop temperature tracking (-) and
transition line between heat sinks (--); (c) power output from the
nitrogen (--) and water (-) heat sinks; (d) difference between setpoint
and controlled variable.

behaviors, Eqs. (1) and (2), that result when the temperature
crosses 300C, while satisfying the constraints on heating
power. This control problem is referred to in industry as a
split-range control problem.''

One strategy is to design an Interal-Model-Control (IMC)-
Based Proportional-Integral-Derivative Controller (PID) with
a filter for each process transfer function, implement the
controllers in digital velocity form, and switch controllers
when the heat sinks are changed. Figure 2 is the block
diagram of the system. The continuous-time transfer func-
tion for an IMC-Based PID controller is[7-9]

k(s)= Kcl+ 1+ TDs+- (3)
TIs Ti FS+1

2T+9 6 T0 o
K, =2K(=+ ; = "D=2;+; CFD2= +0) (4)
K~2K( + 0) 2 2+9 2(X+O)

The IMC tuning parameter ? provides a trade-off between
speed of response and the robustness of the closed-loop
system to measurement noise and inaccuracy in the model.
The time domain expression for the controller is

du de
F -+u(t)=u(0)+Kc e(t)+TD + Je(o)d (5)
dtdt c 0
uI 'lo J
where e is the difference between the setpoint r and mea-
sured variable m. To avoid derivative kick,[21 the derivative
of the error e=r-m is replaced with the derivative of the
measured variable m to give

TF +u(t)=u(0)+Kc e(t)-TD + e()di (6)
Io 0 J

By approximating the integral by a summation and the de-
rivatives by a first-order backward difference, and rearrang-

ing, we arrive at the control algorithm in digital form

u TF/ At
un = + un-]
l+TF/At 1+TF/At

KC imn mn-1 At
+ en -T"D -- +- ek (7)
l+TF/At At T k I
At= t -t,_- samplingtime
u, value of the manipulated variable (% Power) that
is held constant between times to and t,+i
m,, e, defined similarly.

Writing Eq. (7) for the n-1 sampling instance and subtracting
gives what is referred to as the velocity form of the algo-

Un =n-1 + /AtUn -Un2)

Kc e mn-2mn_1+mn,2 Ate
+ Fen" -en-I -"TD +-- e n (8)
1+TF/At At Tj

The main advantage of implementing the controller in this
form is that it will not integrate the controller error when the
manipulated variable reaches a constraint (for example, 0 or
100% power). For this reason, the controller will also per-
form better during transitions between different operating
conditions-that is, it will provide bumpless transfer. The
sampling time was selected as At =0.1 minute, which is con-
sistent with well-known rules-of-thumb."l The IMC tuning
parameter was selected to be X=1.0 minute to give fast
uniform closed-loop response throughout the temperature
ramp. Disturbances on the output were modeled as inte-
grated white noise, given by

do = o (9)
dk =dk- +PYk (10)
where the Yk's are normally distributed random numbers.

Figure 2. Block diagram. The setpoint signal is r, the error signal is e, the measured temperature
is m, and the effect of the disturbances on the temperature is d. The selector S switches between
the controllers k, and k2, depending on the value of the measured temperature.
Chemical Engineering Education

The coefficient p3 was set to 0.013, to be consistent with the
open-loop step tests for the apparatus.141

A simulation of the closed-loop temperature response to
programmed step and ramp trajectories is shown in Figure 1
(the program producing the plot is available from the World
Wide Web[6]). The simulated response is very similar to the
experimental response shown in Figure 3.6 of Drake."41 If
disturbances had been better modeled as entering the process
input, then the alternative IMC-Based PID controllers de-
rived in [5,7] would provide improved performance. A slightly
more complex IMC controller would be used [7,9] if zero
offset in tracking the ramp had been required.

The model of a polymer-film-diffusion apparatus was used
to teach the design of controllers that can handle discrete
process changes. An available MATLAB code16' demon-
strates that two digital IMC-Based PID controllers, imple-
mented in velocity form that switch during transitions be-
tween operating regimes, provide high performance for this
problem. This paper and the MATLAB program are pro-
vided with the hope it will encourage teaching the design of
such controllers in undergraduate courses in process control.

S.H. Chung acknowledges the support of the Hauser schol-
arship. R.D. Braatz acknowledges the support of the DuPont
Young Faculty Award.

1. Ogunnaike, B.A., and W.H. Ray, Process Dynamics, Model-
ing, and Control, Oxford University Press, New York, NY
2. Seborg, D.E., T.F. Edgar, and D.A. Mellichamp, Process
Dynamics and Control, John Wiley, New York, NY (1989)
3. Stephanopoulos, G., Chemical Process Control: An Intro-
duction to Theory and Practice, Prentice Hall, Englewood
Cliffs, NJ (1990)
4. Drake, P.A., "Surface-Enhanced Raman and Surface Plas-
mon Resonance Measurements of Case II Diffusion Events
on the Nanometer Length Scale." PhD thesis, University of
Illinois, Urbana, IL (1995)
5. Horn, I.G., J.R. Arulandu, C.J. Gombas, J.G. VanAntwerp,
and R.D. Braatz, "Improved Filter Design in Internal Model
Control," Ind. Eng. Chem. Res., 35, 3437 (1996)
6. Chung, S.H., and R.D. Braatz, Software for a Benchmark
for Studies in Antiwindup and Bumpless Transfer, Univer-
sity of Illinois, Urbana, IL;
Issrl/software/bump.m (1997) Computer software
7. Braatz, R.D., "Internal Model Control," in The Control Hand-
book, W.S. Levine, ed., CRC Press, Boca Raton,FL; 215
8. Rivera, D.E., S. Skogestad, and M. Morari, "Internal Model
Control 4: PID Controller Design, Ind. Eng. Chem. Proc.
Des. Dev., 25, 252 (1986)
9. Morari, M., and E. Zafiriou, Robust Process Control, Prentice-
Hall, Englewood Cliffs, NJ (1989) 0

Summer 1998

L%- book review

by Y.V.C. Rao
Sangam Books Limited; 601 pages (1997)

Reviewed by
Thomas E. Daubert
Pennsylvania State University

This beginning intermediate text is a welcome addition to the
limited number of texts appropriate to entry-level chemical engi-
neering courses. While the major emphasis of the book is for use in
the classroom, employment by more advanced students and practitio-
ners is suggested and is justified by the breadth of material included.
Each chapter of the book begins with a set of learning objectives
that is more helpful than the usual one-or-two paragraph introduc-
tion informing the reader of a chapter's content. At the end of each
chapter, a quantitative summary reviews the material, including the
important definitions and equations. A set of review questions
(primarily qualitative but sometimes requiring a calculation) and a
set of problems pertinent for class use complete each chapter.
The fourteen chapters of the book proceed logically from basic
definitions and concepts to more complex topics, but do not be-
come lost in esoteric arguments of little use to undergraduates. Chap-
ters 1 and 2 give the basic definitions of both thermodynamics itself
and primary concepts such as systems, processes, properties, energy
types, and equilibrium, together with the units used for thermody-
namic calculations. Review questions and problems support the text
in prompting the student to make sure they understand the material.
Chapter 3, on PvT relations of fluids, discusses real fluids to-
gether with ideal gases as a preparation to their use for later applica-
tion to calculations using the first and second laws. The selection of
relations includes the progression of cubic equations from van der
Waals to the various modified Redlich-Kwong equations, as well as
the virial equation and Pitzer corresponding states. The selection is in
line with current industrial use and what I myself would recommend.
The first law treatment is classical, beginning with calculations
of various types of processes for ideal and nonideal gases as well as
steam. Treatment of control mass and control volume analysis for
transient flow processes is much more thorough, but also more under-
standable than most treatments. Standard thermochemical calculation
methods are also included.
The second law treatment in Chapter 5 is again classical, with a
good comparison of heat engines and heat pumps and methods for
calculation of entropy. Control volume analysis and efficiency
calculations are brief, but unusually clear.
In Chapter 6, the mathematical analysis of the state principle, the
criteria for equilibrium, the Gibbs-Duhem equation, and the derived
energy properties, in my opinion, need not be as difficult as they are
presented. This is the only chapter that absolutely needs its summary
for understanding and relevance.
Relations among properties and their manipulation by Jacobeans
Continued on page 237

SR1class and home problems

The object of this column is to enhance our readers' collections of interesting and novel
problems in chemical engineering. Problems of the type that can be used to motivate the student
by presenting a particular principle in class, or in a new light, or that can be assigned as a novel
home problem, are requested, as well as those that are more traditional in nature and which
elucidate difficult concepts. Please submit them to Professor James O. Wilkes (e-mail: or Mark A. Burns (e-mail:, Chemical
Engineering Department, University of Michigan, Ann Arbor, MI 48109-2136.



Part 2: Recycle Loop with Reactor

University of Nottingham Nottingham, NG7 2RD, England

A n earlier article"' discussed teaching flexibility via a
simple heat-exchange problem. Heat exchange is a
convenient starting point for flexibility teaching be-
cause the equations and resulting calculations are easy to
handle. Other process operations can rapidly become com-
plex and a rigorous treatment very demanding.12] Hence it is
pleasing to find that a potentially complex recycle loop with
reactor can be simplified to provide a thought-provoking,
process-flowsheet-based exercise that emphasizes understand-
ing of the system's operation.
It is important to stress, even with simplification, that the
recycle loop is still a much broader example than that for
heat exchange, making the exercise more suitable as a basis
for project work or a discussion question. One of the authors
first became interested in the problem when attempting to
explain the operating bounds of an ammonia or methanol
synthesis loop during the course of final-year design projects.
The reactor used in this exercise is a simpler design than
commonly found in ammonia and methanol plants, but the
exercise still illustrates many of the issues that make design
and operation of such process systems so interesting.

The exercise considers a reversible exothermic reaction
taking place in the gas phase over a catalyst in an adiabatic

reactor. The reactor is part of a recycle loop where product is
separated from the reactor effluent and unreacted feed is
recycled (see Figure 1). It is well known that when designing
recycle loops, an economic optimum exists for reactor con-
version that balances reactor size against the combination of
product separation load and circulation rate.
Having chosen the design conversion, the circulation rate
is an immediate consequence in order to meet the required

Copyright ChE Division of ASEE 1998

Chemical Engineering Education

Warren Jones holds BSc and PhD degrees in
chemical engineering from the University of
Nottingham and is a registered Chartered Engi-
neer. He has a wide-ranging interest in both front-
end process and detailed plant design, devel-
oped initially through nine years of experience
with a major engineering and construction com-
pany. Teaching responsibilities include several
design courses, and engineering thermodynam-

Tony Wilson holds BSc and PhD degrees in
chemical engineering from the University of
Nottingham. With industrial and consulting ex-
perience in process control and batch process
engineering, and with active research in both
fields, he coordinates the department's research
in computer-aided process engineering and is
responsible for process control teaching at the
undergraduate level.

14*4 1 1

production. The choice of reactor operating temperature is
also important and may be optimized by the designer as part
of the reactor size-separation load/circulation rate trade-off.
In the case of reversible exothermic reactions, selection of

Figure 1.

20% extra flow
0.12 -
0.11 / 40 t

C- ~-- -'' "- -
S0.105 ....

0.095 30t

0.0 EOR, 35t
0.085 20% less flow
0.075 ----
500 505 510 515 520 525 530 535
Reactor Inlet Temperature K

0.115 B' "- EOR Circulation Rate

0.11 -^- BOR Catalyst
o 0.10 Activity
X 0.105 /
S0.1 "

o 0.095 BOR circulation Rate EOR Catalyst
S (-85% EOR Circulation) Activity

475 485 495 505 515 525
Reactor Inlet Temperature K

Figure 2.

a relatively low temperature permits a higher conversion,
but this must be balanced against a slower rate of reaction
leading to a large catalyst inventory. The design conversion
for reversible exothermic reactions often approaches quite
closely the upper bound imposed by thermodynamics.
Finally, the temperature will increase through the catalyst
bed, but a single temperature value is needed to character-
ize operation; for this we will use inlet temperature. It is
assumed that reactor inlet temperature can be adjusted
For a flexibility study, the catalyst inventory and other
equipment will be fixed; we are interested in investigat-
ing how the plant will perform under alternative operat-
ing conditions.
From the earlier discussion, two obvious parameters to
characterize operation are circulation rate and reactor inlet
temperature. As noted before, the reactor inlet temperatures
can be adjusted to maximize production for fixed circulation
rate (illustrated in Figure 2a.[31).

To the left of the maximum, increasing inlet temperature
improves the rate of reaction, leading to more pro-
duction. But as the temperature increases, the dete-
riorating equilibrium conversion imposed by ther-
modynamics "bites," and production peaks. Opera-
tion at the optimum inlet temperature ensures that
the minimum circulation rate to meet production is
being used; we assume the optimum temperature to
be selected in each case.

A typical industrial example is more complex and
involves four additional considerations:
* Catalyst deactivation
* Producing circulation around the loop
* Loop pressure
* Presence of an inert component in the make-up
The catalyst can be deactivated in several ways.[41
Here, we are more interested in the overall effects of
deactivation than we are in the details of one par-
ticular mechanism, so we adopt a simple model-
namely, uniform deactivation throughout the cata-
lyst inventory. The normal response to loss of pro-
duction through catalyst deactivation is to raise the
reaction rate by increasing the inlet temperature. For
reversible exothermic reactions, however, this ac-
tion further limits the conversion, and unless the
circulation rate is also increased, the production tar-
get will not be met.
The effects of catalyst deactivation are demon-
strated in Figure 2b, which compares performance
at beginning-of-run (BOR) and end-of-run (EOR).
Note particularly that the BOR circulation rate is

Summer 1998


Parts 1 and 2 of this paper introduce flexibility by providing exercises that force the student
to consider how the system will actually operate-this is an important first step for
developing a robust design. Further, a wide-ranging knowledge of basic chemical
engineering is required, making these exercises (particularly Part 2)
ideal as the basis for project work.

15% less than that for EOR; if the EOR circulation
implemented at BOR, overproduction would result
Normal design practice would be to set the loo1
parameters for EOR operation, but producing a des
ciently flexible to operate at BOR is clearly very c
Loop circulation is normally produced using a varial
centrifugal compressor, so compressor character
come important in meeting process flexibility.
Centrifugal compressor characteristics are repres
a head-capacity curve and, for recycle systems (v
loop pressure drop is 10% or less of the nominal
pressure) the loop can be treated as one of constant
and the compressor characteristics will have rou
same shape as those obtained for centrifugal pumps
Figure 3 shows the operating point to be defined
system curve intersects the compressor characters
recycle loop the system curve is based solely or
losses because the discharge returns to the suction.
Also shown in Figure 3 is the implementation of
circulation rate via compressor speed reduction. Th
laws[51 can be applied to relate compressor perfon
different speeds. Sometimes the speed reduction re
quite large, taking the compressor close to a critic
which induces synchronous whirling of the shaft; th
tion must be avoided and hence places a lower b
Typically, critical speeds are found in the range
80% of normal running speed. Two other constra
arise from compressor operation are
Maintenance of sufficient volumetric o80
flowrate through the machine to avoid
surge; for a low-head compressor (as 701
considered here) this constraint will
lie well to the left on the head-capacity 601
curve (around 50% of normal capac- E
ity) and hence will not be influential. 50'
Overspeed trip, typically set at 10 to 401
15% above the normal running speed;
this constraint obviously "bites" for 301
operating modes requiring higher cir-
culation rates. 20(

The constraints imposed by compressor
operation lead to the third parameter that
required further consideration-namely,

rate were loop pressure. Varying loop pressure is helpful when match-
ing compressor operation to process requirements because
process A reduction in loop pressure increases the circulat-
ign suffi- ing gas specific volume; hence, circulating the same
desirable. quantity of gas would require a faster compressor
ble speed speed to cope with the larger volumetric flowrate,
stics be- i.e., moving operation away from the critical speed.
SIf the reaction is one where there is volume reduc-
ented by tion as the reaction proceeds, pressure reduction
here the will give a lower, thermodynamic limited, conver-
)perating sion, and this will lead to a larger circulation rate
t density, and a move away from the critical speed.
ghly the The loop pressure cannot be set independently, however.
In a simple example where the feed contains no inert compo-
vhere the nent, the rate of reactant consumption (in the reactor) must
tic; for a balance the rate of make-up gas (MUG) flow into the loop in
friction order for the pressure to remain steady. In other words, at
steady-state, the reactor inlet temperature and circulation
reduced rate (combined with the pressure) must give a reaction rate
e affinity that just balances the MUG rate. If the circulation rate, say,
mance at were smaller, the loop would equilibrate at a higher pressure
quired is with the extent of the pressure increase being limited by the
al speed, relief valve setting on the loop.
is condi- Finally, we need to consider an inert component in the
ound on MUG. In many industrial examples, the MUG to the synthe-
sis loop is not pure reactant, but contains 1 to 2% by volume
of 60 to of an inert component. Unless the inert is removed, it will
lints that accumulate in the loop and slow the rate of reaction. In a
pressurized loop, some inert will dissolve in the product, but

0.6 0.7 0.8 0.9 1 1.1
Actual Volumetric Flowrate m^3/sec

1.2 1.3 1.4

Figure 3.

Chemical Engineering Education

most has to be purged from the loop (see Figure 1). A
relatively high purge rate ensures the inert composition in
the recycle loop is low and the rate of reaction high for the
pressure because the inert has only a small diluting effect. A
high purge rate, however, results in a large reactant loss
and, in design mode, setting the purge rate is an eco-
nomic optimization that adds the third dimension of MUG
rate to the reactor size- separation load/circulation-rate
balance already discussed.
From a flexibility viewpoint, recycle loops with purge can
fully exploit the interacting trio of parameters (circulation
rate, reactor inlet temperature, and pressure) because pres-
sure can now be set independently via the purge rate rather
than being a consequence, as previously noted.
To illustrate, if the loop is initially at steady state, reducing
the purge rate causes the inert composition to increase, so
the loop pressure must increase to restore the reactant partial
pressures. There will be a compensating reduction in MUG
demand. If the circulation rate is also increased, it will limit
the change in pressure needed.
In view of the complexity added by considering an inert in
the MUG, the main part of the exercise concentrates on pure
MUG; handling an inert in the MUG is considered at the end
of the exercise.


The reversible exothermic chemical reaction
A+B > C
takes place in the gas phase over a fixed catalyst bed in a
reactor operated adiabatically. The reaction data are summa-
rized in Table 1. The reactor forms part of a recycle loop
(shown in Figure 1) where C is totally separated from the
reactor effluent. For the first three parts of the exercise, you
may assume the make-up gas contains only A and B in

Reaction Data for the Exercise

Heat of reaction = -14000 kJ per kmol of C formed
Rate of production of C by the forward reaction = kFPAP,
kF = BOR rate constant, kmol sec bar' (kg of catalyst)'
kF = 100 exp(-94000/8.314T)
Rate of destruction of C by the reverse reaction = k Pc
k, = BOR rate constant, kmol sec' bar' (kg of catalyst)
kB = 525000 exp(-108000/8.314T)
P ,PB,PC = partial pressures of A, B, and C, bar
T = absolute temperature of the reacting mixture, K

Specific heat capacities: CA', C,,, CP = 30, 40. 70 kJ kmol-' K'

Summer 1998

stoichiometric proportions (mixture RMM=50); hence the
recycle stream also contains only A and B, and the purge
flowrate will be zero.
The recycle loop at end-of-run (EOR) is designed to oper-
ate at a nominal pressure of 50 bar, with a circulation rate
through the reactor of 2.2 kmol sec '. The EOR loop pressure
drop is 5 bar; this comprises 4 bar for all equipment exclud-
ing the reactor and 1 bar for the reactor. The EOR reactor
pressure drop may be further decomposed:
Pressure drop (based on EOR circulation rate) for
reactor if catalyst were in good mechanical condition:
0.7 bar
Allowance (based on experience) for catalyst particle
breakdown and loss of voidage: 0.3 bar
Circulation through the reactor is provided by a single-
stage centrifugal compressor whose head (He, m) capacity
(Q, m3 sec ') characteristic at 10,000 rpm is represented by

He = 530.5 + 149.52 Q 130.585 Q2
The exercises below examine the process flexibility re-
quirements for the synthesis loop as the catalyst inventory of
35000 kg slowly deactivates. Catalyst deactivation is mod-
eled by assuming EOR kF and kg values are only 30% of
those applying at BOR (see Table 1).
You may assume the heat exchanger arrangements on the
reactor feed have sufficient flexibility to produce a wide
range of reactor feed temperatures and the compressor
suction operates at 40'C. Also, assume the reactants A and
B have negligible solubility in the product.
1) Determine the maximum production of C achievable
at EOR conditions by varying the reactor inlet
2) You wish to achieve the same production rate of C as
that calculated in (1), but using BOR catalyst.
a) Calculate the circulation rate required if the loop
pressure remains at 50 bar.
b) If the circulation rate is maintained at 2.2 kmol
sec-', at what pressure would the loop need to
c) Use the insight you have gained from solving 2(a)
and (b) to explain the interrelationship between
loop pressure, circulation rate, and reactor inlet
temperature. How might you exploit this flexibil-
ity to deal with variation of catalyst activity?
3) a) Confirm the compressor speed of 10,000 rpm will
satisfy EOR operating conditions.
b) Determine the compressor speed for 2(a).
c) Determine the compressor speed for 2(b).
d) Comment on your answer to 2(c) in light of the
compressor speeds you have just calculated.

4) The MUG to the synthesis loop contains 2% of an
inert compound, I, in addition to equal proportions of
A and B. If the circulation rate is 2.4 kmol sec and
the purge rate is 0.05091 kmol sec ', determine the
loop operating pressure to meet the production target
from part 1. What would happen if the purge rate
were reduced further, and how would you mitigate
the consequences?


The solution of this problem requires a model for the
chemical reactor, which is easily generated by integrating
the differential mass and energy balances through the reac-
dFc q(kFPAPB kBPc)
dT q 14000 (kFPAPB -kBPc)
dw (FACPA + FCpB + FCCPc)

where w is the mass of catalyst and q is the multiplying
factor to account for catalyst deactivation.
At the reactor inlet, the flowrates and temperatures are
FA=FAO, FB=FBo, F=O0, T=T0, and at any subsequent point,
FA=FAO Fc, FB=FBo-Fc. Ideal gas behavior is assumed to
calculate the partial pressures, e.g., PA=PTFA/(FA+FB+Fc),
where PT is the nominal loop pressure. Any one of a number
of numerical integration packages can be used.

i Setting FA0=FBO=I.l kmol sec ', q=0.3, and PT=50 bar,
the model is run for a range of inlet temperatures T0. Fc is
noted after integration through 35 tons of catalyst and the
results recorded as a plot of Fc against To (see the middle
curve in Figure 2a). The peak indicates a maximum produc-
tion of 0.10206 kmol sec' for C at an inlet temperature of
517K. (Note the sensitivity to circulation rate changes, but
relative lack of sensitivity to catalyst inventory changes as
indicated by the other curves in Figure 2a.)

2(a).For BOR catalyst q=1.0 and PT=50 bar, the problem
requires a trial-and-error solution to find FAo and Fo, that
give the required production of Fc. A solution is found for
FAO=FBo=0.951 kmol sec with a reactor inlet temperature of
487K (see Figure 2b). (Note: maintaining the EOR circula-
tion rate gives a large overproduction.)

2(b). For BOR catalyst q=1.0 and FAO=FBO=1.1 kmol sec-',
the problem requires a trial-and-error solution to find PT that
gives the required production of Fc. A solution is found for
P,=43.64 bar with a reactor inlet temperature of 495K.

F2(c).By performing the calculations in 2(a) and 2(b), stu-

dents should be aware that there are three important vari-
ables. Reactor inlet temperature can and should be opti-
mized for all situations, but circulation rate and loop pres-
sure are related and cannot be set independently. The above
exercises take the two extreme positions of maintaining loop
pressure or maintaining circulation rate; the corresponding
maximum reduction in circulation rate or loop pressure is
then calculated. In practice, an operator would use a smaller
change in circulation rate and allow the loop pressure to
equilibrate with a reduction somewhat less than the maxi-
mum change previously calculated.

3(a.) Plotting the head-capacity curve using the given equa-
tion produces the 10,000 rpm line in Figure 3. The EOR
circulation rate is 2.2 kmol sec- or a compressor suction
flowrate of [(2.2)(0.08314)(313)]/50=1.145 m3 sec-', and as
may be seen from the curve or equation, this implies that a
He of 530.5m will be thrown up. If the loop pressure drop is
5 bar, this is equivalent to an Hs of (5)(105'/
[(9.81)(96.07)]=530.5m, i.e., Hs=Hc and the compressor
speed of 10,000 rpm will satisfy EOR operating conditions
(96.07 kg m3 is the gas density at the compressor suction
and is calculated from the ideal gas law).

3(b). In this case, the system head will be proportional to
the (circulation rate)2 because the loop pressure is main-
tained at 50 bar and thus the gas density can be assumed
constant. We neglect the effect of minor composition and
reactor exit temperature changes. Thus the system curve
passes through the volumetric flowrate [(2)(0.951)(1.145)]/
2.2=0.9899 m3sec-1 at a head of [(4.7)(530.5)(0.951 {2/2.2})Y]/
5=372.7m. Stable operation requires the compressor speed
to be reduced such that the head-capacity curve also passes
through this point. Let N be the new compressor speed and
the stable operating point must map back onto the 10,000
rpm head-capacity curve. Thus

372.7 10,0002

( 10,000)2)
530.5 + (149.52)(0.9899)\'N


Solving the quadratic gives N=8420 rpm; this is shown as
the second compressor curve in Figure 3. (This is quite a
large speed reduction and may bring the compressor quite
close to the critical speed.)

S3(c).In this case, the system head will be proportional to
circulation rate
loop pressure )
Chemical Engineering Education

because dropping the loop pressure has a strong effect on gas
density. The required point on the system curve is defined by
[(1.145)(50)]/43.64=l.3119m3sec' at a head of
[(4.7)(530.5)({50/43.64}2)]=654.6m. Using the same ap-
proach as that applied in 3(b), but this time anticipating a
speed increase, we solve the quadratic to give N=11,160 rpm.
(This is quite a large speed increase and may bring the
machine close to overspeed trip.)

3(d).Answer 2(c) advocates a middle path-some reduc-
tion in circulation rate and a corresponding loop pressure
reduction. The answers to section 3(b) and (c) show that if
loop pressure or circulation rate is maintained, the extremes
of compressor speed are also approached, i.e., 2(a) takes
operation close the critical speed and 2(b) results in opera-
tion near overspeed trip. Hence, the middle path requires a
steadier machine speed, closer to normal running speed.

4. The equations representing the reactor must be modified
to include the presence of the inert. Two changes are needed:
1. Inclusion of the inert component flowrate, F1, in the
circulation rate (FA+FB+Fc+F,); this reduces the par-
tial pressure of any reactant.
2. Inclusion of the inert component heat capacity, FICp,,
in the heat capacity (FACpA+FC,,+FcCpc+FICpi);
this provides a greater heat sink and so reduces the
temperature rise across the reactor.
Before the reactor simulation can be used, the effect of the
inert component on the recycle loop material balance must
be established. In particular we need the reactor feed compo-
sition. Figure 4 shows all the information easily derivable
from the problem statements, but two important unknowns

Component kmol/sec
C 0.10206

Reactor fee
Component km
I 2.

I-1- 2.4

Reactor effluent
Component kmol/sec
A X-0.10206
B X-0.10206
C 0.10206
I 2.4-2X

Figure 4.

remain; namely 1) flowrates of A or B in the reactor feed, X,
and 2) MUG flowrate, M. Writing component mass balances
for a reactant and the inert at the mixing point just before the
compressor gives

X=0.49M+(X-0.10206) 2.14497
-' .. 2.19588)


Solving for X and M gives X=1.089976 kmol sec' and
M=0.255028 kmol sec', and hence the reactor feed for the
simulation is A and B, 1.089976 kmol sec-' and I is 0.220048
kmol sec-'. Note particularly that the MUG rate has in-
creased by 24.9%, compared to an inert free MUG, to ac-
count for the reactants lost in the purge. Solving for the loop
pressure in this part of the exercise is also by trial and error.
A loop operating pressure of 55.62 bar will ensure the target
production is met if the reactor inlet temperature is opti-
mized at 517K.
Reducing the purge will increase the inerts composition of
the gas in the loop. At constant circulation rate, the change
must cause the loop pressure to increase in order to restore
the reactant partial pressures. For high-pressure systems, the
relief valve is typically set at 10% above normal running
pressure; hence the calculated loop pressure of 55.62 is
probably unacceptable if 50 bar is viewed as normal opera-
tion. A further reduction in purge flow would certainly cause
the relief valve to lift. To limit the pressure increase, the best
tactic would be to increase the loop circulation rate.


Parts 1 and 2 of this paper introduce flexibility by provid-
ing exercises that force the student to consider
how the system will actually operate-this is an
important first step to developing a robust de-
d sign. Further, a wide-ranging knowledge of ba-
oI/sec sic chemical engineering is required, making
x these exercises (particularly Part 2) ideal as the
4X basis for project work.

1. Jones, W.E., and J.A. Wilson, "An Introduction to
Process Flexibility: Part 1," Chem. Eng. Ed., 31(3),
172 (1997)
2. Swaney, R.E., and I.E. Grossmann, "An Index for
Operational Flexibility in Chemical Process De-
sign,"AIChE J., 31(4), 621 (1985)
3. Froment, G.F., and K.B. Bischoff, Chemical Re-
actor Analysis, 2nd ed., John Wiley, New York,
NY, p. 425 (1990)
4. Ibid. p. 219
5. Aerstin, F., and G. Street, Applied Chemical Pro-
cess Design, Plenum Press, New York, NY, p. 237
(1978) 0

Summer 1998

W essay


A Curious Application of Thermodynamics

Universidad Sim6n Bolivar Caracas 1080, Venezuela

here is a loose analogy between the intermolecular
forces that govern the observable behavior of fluid
systems and the social forces that drive human be-
havior. Based on this premise, at least in principle, we can
use thermodynamics to describe social systems. This paper
will put forth some ideas, basically similes, that will help
understanding of some common social situations, such as
divorce and racism, through thermodynamic reasoning.
The origins of classical equilibrium thermodynamics as
we now know it rest on the early findings of Watt, Clausius,
Carnot, Joule, and Gibbs, along with many others. The re-
sults obtained by these early thermodynamicists were based
on careful and systematic study of idealized thermal sys-
tems. Curiously, none of these "founding fathers of thermo-
dynamics" had an appreciable comprehension of the exact
constitution of matter. As an extreme example, one can point
out how Sadi Carnot established the second law of thermo-
dynamics without knowing the law of conservation of en-
ergy or the molecular nature of matter. In fact, he had an
erroneous idea of what heat was, even though that quantity
was the basis of his analysis. The relations and results ob-
tained in this early classical thermodynamic period are inde-
pendent of the actual nature of the systems studied and are,
indeed, very general. This happy occurrence is the reason we
can extrapolate the fundamental concepts of thermodynam-
ics to other modern disciplines (e.g., geology,'1 information
science,121 and medicine"') if the analogies are carefully made.
Nowadays, we have elucidated many secrets of the nature
of matter. Statistical thermodynamics faces the challenge of
predicting macroscopic behavior of systems through knowl-
edge of intermolecular interactions and appropriate averag-
ing among the large number of molecules that constitute a
system. It is fascinating to see how the statistical mechanical
predictions line up perfectly with the earlier classical results.
For example, for early scientists pressure was simply a prop-
erty that described a system and could be related to work or
energy. Today, we understand pressure as a result of forces
between molecules in a fluid. In other words, today we can
somehow understand the collective behavior of the system if

we comprehend the interactions on an individual level.

In nature, molecules interact among themselves by means
of forces that translate into the observable behavior of com-
mon substances. A few examples are the existence of a
particular temperature at which liquid boils, the reason that
two substances mix while others do not, and peculiar behav-
ior such as ice floating on water-all of which can all be
explained once we understand the molecular forces.
Even the most insignificant molecule of the simplest com-
pound interacts with its neighbors by means of specific
forces. The underlying cause for the presence of these forces
is the physical separation of positive and negative charges in
the atoms. The fundamental electrostatics and quantum me-
chanics needed to fully explain the nature and form of inter-
molecular forces are beyond the scope of this discussion. It
will suffice to understand that if two atoms attempt to come
too close to each other there will be an electrostatic repulsion
between them. A macroscopic manifestation of this is the
fact that matter cannot occupy the same space; in a simpler
fashion, we witness this repulsion when two billiard balls
collide. On the other hand, in atoms and molecules the
electrons are not fixed in their orbits but rather move around
in average locations. These fluctuations in position lead to
fluctuating molecular dipoles, and they account for a weak
type of attraction among molecules (sometimes called dis-
persion, or van der Waals forces). It is this attraction that
accounts for the existence of condensed phases in which

Copyright ChE Division of ASEE 1998

Chemical Engineering Education

Erich A. Muller is Professor at Simon Bolivar
University (USB). He received his engineering
and MSc degree at USM and his PhD at Cornell
University. His research programs include mo-
lecular simulation of complex fluids and the pro-
duction of software for chemical engineering
applications. He is author of the text
Termodinamica Basica (editorial Equinoccio,

molecules are closely packed. These attraction forces are of
rather short range and are basically imperceptible if the
centers of the molecules are separated by more than three or
four molecular diameters. At larger distances, the majority
of molecules do not interact directly." (See Figure 1.)
Instead of speaking of forces, thermodynamicists prefer to
talk about potentials, whose variation with respect to dis-
tance represents the force. Potentials have the advantage of
having units of energy; thus a discussion of some concepts is
simplified. A typical potential function has basically three
zones. At short distances, the potential is positive and be-
comes larger in magnitude
as molecules come closer repulsion attraction
together (indicating an in- I
creasing repulsion). At in-
termediate distances there
is an attraction. The po-
tential is negative and, in
fact, presents a minimum
corresponding to a dis-
tance at which the mol-
ecules find a balance be-
tween attractive and repul-
sive forces. Lastly, at2 "
large distances, molecules
do not interact directly
and the potential is effec-
tively zero.
Distance between centers ol
POTENTIAL Figure 1. The solid line sh
The central point of this intermolecular force for two s
a it of the distance between their
essay is that with a little cates repulsion among molec
bit of imagination, one traction. The relation to hun
can see that the social isolated individual (the sunbi
behavior of humans actions from other humans. A
may be governed by an tively close to other individual
"interhuman potential" going to the movies). Too clo
similar to that of insignifi- ally leads to repulsion.
cant molecules. To ex-
plain, let us assume we are studying people at a large fancy
party. As guests enter the ballroom, the first thing they do
(after serving themselves a drink) is to mingle, wandering
without direction. They place themselves at judicious dis-
tances, not too close but not too far away, from others-a
distance corresponding to the minimum of the interhuman
potential. If we attempt to get too close to an individual,
there will be an inherent repulsion. It is the typical discom-
fort we feel when we are approached by a "close-talker" and
are compelled to take a small step back, thus regaining the

An exception to this comment are the forces among ionic fluids
(salts dissolved in water, for example) and polar fluids where
long-range forces, due to strongly anisotropic charge distribu-
tions, are present.
Summer 1998

ian i.
sals f
se an

appropriate distance. It is clear that two people cannot oc-
cupy the same space; thus repulsion becomes infinite at
extremely short distances. On the other hand, one will rarely
be isolated at a large party and somehow will be attracted to
some of the clusters. Humans are social beings, attracted to
their fellows. Lastly, the presence of persons at a large
distance away will not affect us. We do not know, and
usually do not care, who is at the other end of the ballroom.
An interesting point of this logic is that both at a human
and a molecular level, even when the individual interactions
may be somehow different and the fluctuations around the
mean significant, the en-
no appreciable interaction semble will have seemingly
homogeneous properties if
the number of individuals
studied is extremely large.

If we start from the premise
that molecules and humans
are governed by roughly
similar potentials, thermody-
namics (which serves to
study and comprehend the
former) may well describe
the latter. Moreover, the ther-
modynamic laws applicable
to fluids may possibly de-
Sscribe some aspects of hu-
man society.
in a qualitative way the
le molecules as a function In a multicomponent fluid,
ers. A positive value indi- each type of molecule will
and a negative value, at- have its own peculiar inter-
nteractions is evident. An action. Even though the
) is at peace without inter- qualitative shape will usually
theless, he may come rela- be similar, there are differ-
ror certain occasions (like ences among species (e.g., if
encounter, however, usu- there are two types of mol-
ecules, say type A and type
B, the potential minimum might be deeper between self-self
A-A and B-B interactions than with unlike A-B interac-
tions). Additionally, other types of specific interactions may
also be present due to multipolar forces dipoless, quadru-
poles, etc.), association forces (hydrogen bonds, for example),
and others. In a complex fluid there might also be symbiotic
relations-two molecules may bond for mutual benefit (as
exemplified by electron-transfer bonds), or there may be a
strong repulsion due to dissimilar cross-interaction forces.
Similarly, in human society, our own individuality will draw
us to some other people while we strongly dislike others.
Racism Segregation (or intermolecular racism) is com-
mon in fluid systems. A classical example is an oil droplet in
water. An oil molecule, being so different with respect to its

immediate surroundings, will tend to migrate to a border,
avoiding unfavorable interaction with the homogeneous wa-
ter majority. If other oil molecules are present, they will
attempt to aggregate, thus achieving a more stable system
and forming a "molecular ghetto." If oil molecules are present
in significant numbers, a phase separation will occur, creat-
ing a region where all interactions will be favorable (since
all of the members of the phase are compatible). The border
with the aqueous phase will not be free from a high interfa-
cial tension. The analogy with a homogeneous society,
where a small racial, cultural, and/or religious group
appears is obvious. World history is plagued with ex-
amples such as the persecution of our native inhabitants
during the colonization of the Americas, the separation
of Pakistan and India, the Arab-Israeli conflict, and the
disintegration of the Balkan states.

A very heterogeneous fluid may mix under one of two
conditions: 1) by constantly adding a high influx of energy,
or by 2) modifying the interactions between the molecules.
Following the above example, we can vigorously shake a
water/oil mixture, forming a homogeneous system during
the process (much in the way one mixes a salad dressing).
But the result is not the most stable equilibrium state, and as
soon as the energy flux is suspended, the system will sponta-
neously phase separate, independent of the previous effort to
keep it homogeneous. The old Soviet block comes to mind;
it disintegrated into its "natural" communities once the strong-
hold of a central government ceased to exist. All the homog-
enizing efforts through the decades were futile since they did
not alter the basic relations between the republics.
The second method to obtain homogeneity is to modify
interactions between molecules by adding amphiphilic mol-
ecules (molecules that present one type of behavior on one
side of the molecules and a different behavior on the other
side). Common examples are soaps and detergents. These
molecules, when added to oil/water systems, may convert
the system to a stable emulsion. This suggests a rather fool-
proof way to eliminate racism and intolerance among di-
vided societies. It is essential to maximize the number of
people that can be accepted by both conflicting groups, thus
minimizing the formation of distinct "phases."* In South
American societies, existing races have crossbred since co-
lonial times and the majority of inhabitants have some amount
of racial mixture in them. In these societies, racism as such is
unheard of since this mixed group does not differentiate by
race and "accepts by equal." In a similar fashion, the exist-
ence of a large middle class in a country is the only sensible
thermodynamic path to obtain social stability.

'Strictly speaking, only a small amount ofsurfactant is needed to
emulsify oil and water. Therefore, only a small "middle" class
would be needed to get some mixing of society as stated. A large
amount of surfactant is needed to form a microemulsion where the
mixing is at a fine level, thus obtaining a stable society.

Love Other important interactions among molecules are
those of association. They are characterized by being spe-
cific, discriminatory, and producing a much stronger attrac-
tion than van der Waals forces. A typical example is the
formation of hydrogen bonds. At a molecular level, two
molecules forming a hydrogen bond will maintain them-
selves at a much closer distance than that expected for nor-
mal molecules and will be bonded for a longer time than two
non-associating molecules. Again, the analogy between this
behavior and that of humans is apparent. There are certain
humans who form tightly bonded pairs or clusters for rela-
tively long times. Buddies or lovers, certain humans form
bonds among themselves that have the same characteristics
as hydrogen bonds.
Hydrogen bonds are not chemical, but physical, in nature
and thus may be fragile. Usually, the collision with a third
molecule produces a perturbation large enough for the bond
to break. To keep the energy at a minimum (the thermody-
namic equilibrium condition), it is possible that one of the
molecules of the broken pair bonds with the third one (mo-
lecular infidelity and divorce). There are also metastable and
short-lived bonds (molecular one-night stands).
Macroscopically, an associating fluid is denser and will
persist as a condensed phase at higher temperatures than its'
non-associating counterpart. In a similar fashion, societies
where the concept of family is deeply rooted tend to be more
solid and stable. Models of collective societies like the clas-
sic Spartan or the modern Israeli kibbutz are based on the
thermodynamically inconsistent idea that they are solid due
to their number and collective behavior. The solidness of a
society depends on the strength of the particular bonds among
their members. Here, for example, is the understandable
force of the Mafia. There is no human force greater than
love, and therefore it must be the base of a stable society.
Hate, on the other hand, is a clearly repulsive force, and it
tends to disintegrate societies the way a high temperature
disperses a liquid into a gas by boiling.
Resistance to Change Another aspect of society that can
be considered from the viewpoint of thermodynamics and
statistical mechanics is the ability to relate the capacity for
change to transport coefficients. It takes a while for society
to move from one state to another. The laws of irreversible
thermodynamics indicate that the rate of change depends
basically on two parameters: the driving force and the trans-
port coefficient,* which is a property of the substance. In a
society, the driving force is the real need for change. The
transport coefficient depends on the overall cultural level of
the members, their education, and their willingness to un-
dergo change. Large changes may be due to large values of
the driving force, of the transport coefficient, or both. Po-
litical revolutions are characterized by large driving

SFor example, the flow of a liquid will depend on the pressure
gradient (driving force) and its viscosity (transport coefficient).
Chemical Engineering Education

forces; on the other hand, the so-called German and Japa-
nese miracles following WW II can be explained by high
transport coefficients.
The heterogeneity seen in all societies has clear resem-
blance to fluid transport, say through a pipe where molecules
move at very different relative velocities, even when the
flow seems uniform. We notice that molecules near the wall
do not move while those at the center move with the highest
speeds. In society, we have individuals who cling to the
walls of the "status quo" while other, more progressive
types, usually lead the way.


Among the first things a student of thermodynamics learns
are the statements of the laws of thermodynamics, postulates
written in stone that indicate the way in which energy is
transferred. The first law guarantees that we cannot create
energy-a lesson all humans have experienced. Things are
not done by themselves; someone has to do them. The sec-
ond law is more difficult to put into human terms and is
therefore frequently ignored, with tragic results for society.
It places restrictions on the flow of energy, giving it a pre-
ferred direction. This is done by defining a mathematical
quantity called entropy, which upon being measured during
a spontaneous change in an isolated system must increase
(almost in the same fashion as that thing we call time). A
common, albeit undesirable, analogy of entropy is as a mea-
sure of disorder in a system.
The second law tells us there is no reasonable expectation
that an isolated system may order itself spontaneously. The
janitor of our school knows this quite well since every day
he places chairs in a perfect Cartesian ordering, only to find
them in total disorder at the end of the day. The only way to
order either molecules or humans is through a transfer of
energy between the system and its surroundings. Societies,
being a collective of human beings, cannot escape the conse-
quences of the second law. To obtain an ordered society
(low entropy), one has to balance the entropy generation by
inputting large quantities of work into the system via a
strong government. The Marxist experiment has lasted for
over eighty years, but was thermodynamically inconsistent
and doomed to failure from the start. Dialectic materialists
believed that societies would eventually tend to become
ordered and organized systems in which the government
would eventually disappear. The second law states quite the
opposite-a society without government tends to total disor-
der and anarchy. Curiously enough, Marxist societies evolved
according to thermodynamic postulates, evolving into strong
dictatorships. It seems that large empires (complex sys-
tems with lots of order and very low entropy) may not
survive without disordering the environment and will
eventually collapse when the energy transfer through its
boundaries is stopped.

Summer 1998

Civilization Human society has evolved from a rather
chaotic ensemble of cavemen to organized civilizations with
a large number of specialized individuals. Energetically
speaking, the more complex a society becomes, the higher
its energy dissipation rate. This seems to contradict the sec-
ond law in the sense that isolated systems seek chaos and a
homogeneous energy condition. But human societies are
open systems, with significant interactions with their sur-
roundings. They can only be maintained by an abnormally
large amount of energy. One can easily imagine the cata-
strophic consequences of a city left without electricity. The
fact that some societies tend to decrease their entropy while
increasing their energy consumption implies that there must
be some parts of the biosphere that suffer the consequences,
increasing their entropy in a corresponding way. In fact, the
appearance of civilization is regarded as a mechanism to
hasten the thermal death of our universe.3] Individuals them-
selves[41 and other animal societies, such as ants,151 can be
studied in the same way.

The obsession of a thermodynamicist to explain the world
in which he lives in terms of his basic laws is an old, long,
and endless one. Many of us attempt to explain economy,
understand computers, analyze art and poetry, or prove the
existence of God through our skewed optic. The article by
Kyle[61 does a nice job reviewing some of these attempts.
In spite of this, and without wanting to appear ostenta-
tious, it may be suggested that sociological studies must
recognize the existence of two points of view: a global, or
collective, one and an individual one. Generalities or laws
can be applied on the larger scale since they conform to
ensemble averages of individuals' behavior. In this sense,
human communities parallel simple fluid systems, which we
believe are better understood.

The author is grateful for the comments and discussions
with Professors N.F. Carnahan of Rice University, N. Ruiz
of the Universidad Central de Venezuela, and J.-M. Ledanois,
K. Jaff6, and S. Gonzalez of the Universidad Sim6n Bolivar.

1. Bevensee, R.M., Maximum Entropy Solutions to Scientific
Papers, Prentice-Hall, New York, NY (1993)
2. Janes, E.T., "Information Theory and Statistical Mechan-
ics," Phys. Rev., 106, 620 (1957)
3. Zotin, A.I., and I. Lamprecht, "Aspects of Bioenergetics and
Civilization," J. Theor. Biol., 180, 207 (1996)
4. Schrodinger, E., What Is Life?, Cambridge University Press,
Cambridge, UK (1944)
5. Jaffe, K., and C. Fonck, "Energetics of Social Phenomena:
Physics Applied to Evolutionary Biology," II Nuovo Cimento,
6. Kyle, B.J., "The Mystique of Entropy," Chem. Eng. Ed.,
18(2), 92 (1988) 0





Carnegie Mellon University Pittsburgh, PA 15213-3890

Practicing engineers have long known that it takes
more than just technical competence to succeed and
advance in a career. This fact was most recently ac-
knowledged by the Accreditation Board for Engineering and
Technology (ABET) in its Engineering Criteria 2000. Be-
sides reaffirming the importance of mathematics, science,
engineering, and design in an engineering curriculum, ABET
explicitly lists other educational objectives such as the abil-
ity to function on multidisciplinary teams, an understanding
of professional and ethical responsibilities, the ability to
communicate effectively, a knowledge of contemporary is-
sues, and the acquisition of a broad enough education to
enable the student to understand the impact of engineering
solutions in a global/societal context.
One of the challenges that engineering educators face is
how to achieve these desired outcomes within the constraints
of a four-year curriculum. One possible solution is to break
the problem into two parts: first, we must make our students
aware of the importance of non-technical skills, and second,
we need to help them develop these skills through meaning-
ful exercises that are integrated into their technical courses.
In order to raise the students' awareness of professional
skills and to provide them with some context of an engineer-
ing education, in 1997 we introduced a 2-unit (equivalent to

* Current Address: City University of Hong Kong, Tat Chee
Avenue, Kowloon, Hong Kong

1-credit) course that juniors are required to take in the fall
semester. In this article, I will share my experiences in
teaching this course for the first time.

Course Objectives
The class meets once a week for fifty minutes. Through a
series of 11 discussions, I expect my students to
Articulate what they would like to get out of a college
education and what kinds of careers they would like to
Identify their strengths and weaknesses
Develop skills that will make them better students in
the near term and better workers in the long term
Become more confident infinding theirfirstjobs
Be in a better position to make the transition from the
college to the next phase in their lives
Simply put, the course is about professionalism and about
the skills that are essential to having a productive career.
I adopted The Career Tool Kit: Skills for Success' as the
textbook for the course. As stated on its back cover, the book
"develops the practical and interpersonal skills essential in
modem business settings." The course schedule (see below)
roughly follows the flow of the ten chapters in the book. We
only had time to cover selected sections of each chapter in
class, but I asked the students to read through the entire
book. I also told the students about several other books that
cover similar topics.12-51
We spent each week discussing a particular topic. The key
word here is discussing. In each class meeting I gave the
students a handout listing several items for discussion. After

@ Copyright ChE Division of ASEE 1998

Chemical Engineering Education

Edmond I. Ko is Professor of Chemical Engi-
neering at Carnegie Mellon University, where
he has been since 1980. He received his BS
from the University of Wisconsin-Madison and
his MS and PhD from Stanford University, all in
chemical engineering. His research interests are
in the sol-gel preparation and characterization
of catalytic materials.

I must admit that I approached the teaching of this course with some trepidation... [but] the fear turned
out to be unfounded. Students reacted very positively to the discussion of issues
that are relevant to their study, their lives, and their careers.

providing some background for the
chosen topic, I became more like a
moderator than a lecturer, encourag-
ing the students to express their view-
points and sometimes even encourag-
ing them to argue with each other.
The topics and selected discussion
items are shown in Table 1.
The order of these topics follows
the chapters in the book, but there is
some flexibility in moving the topics
around. Certain topics follow each
other naturally; for example, time
management and money management
fall into the grouping of "resource
management." In the two weeks that
I was out of town, I arranged for two
guest lecturers to speak with the stu-
dents. One was a writing consultant
who talked about writing as an intel-
lectual exercise, and the other was a
career consultant who did a workshop
on internships.
Because of time constraints, I chose
not to repeat certain topics that stu-
dents had already been exposed to
elsewhere in our curriculum. For ex-
ample, professional ethics and resume
writing, two topics that would nor-
mally belong to a course on profes-
sionalism, are covered in our first-
year introductory course and our
sophomore seminar course, respec-
tively. Together, these three courses
provide a firm foundation for the pro-
fessional development of our students
early in their college careers.
Grading Criteria
I designed the grading system in
such a way that students don't have to
compete with each other. Instead, ev-
eryone was given 100 points at the
beginning of the course and lost points
only when they displayed "unprofes-
sional behavior," defined as
Missing a class (minus 5 points)
Showing up late for a class
Summer 1998

Items for Discussion

Universal Work Skills
Skills needed by the American work force
ABET Engineering Criteria 2000
Life-long learning

Self-Image and Motivation
Diversity in the Workplace
Being adaptable
Being tolerant
Being thoughtful
Being unprejudiced
Being empathetic
Setting Priorities and Managing Time
Setting goals
Taking inventory
Identifying time wasters
Budgeting and Investing
Gross vs. net income
Fixed vs. discretionary expenses
Savings vs. investment
Concentration and Memory
Staying focused
Avoiding distractions
Making associations
Using mnemonic devices
Communication in building and maintaining relationships
Nonverbal communication
Getting along with people
Staying Focused and Managing Stress
Keeping promises
Avoiding procrastination
Taking care of yourself
Networking and Tracking Down Career Leads
Establishing contacts
Finding mentors
Talking with people
Job searches
Things to do before, during, and after an interview
Making the Transition to and Moving Up in the Workplace
Striving for excellence
Taking the initiative
Being adaptable
Continuing to grow
Looking at the big picture

(minus 3 points)
Turning in homework late
(minus 3 points)
Turning in homework that is
unsatisfactory or missing
one (minus 5 points).

The penalty was not imposed if
the student provided me with a
valid excuse (e.g., medical rea-
son) in writing.
I told the students that I would
hold myself to the same profes-
sional standard, i.e., that they
each would get 5 points if I
missed a class without prior no-
tice or a valid excuse, showed
up late for class, or did not re-
turn their homework in the fol-
lowing class meeting.

At the end of the semester, a
student who lost less than 11
points received an A; less than
21 points, a B; 31 points, a C.

Students were required to sub-
mit an essay after each class
meeting. The essays were graded
satisfactory/unsatisfactory; the
guidelines for a satisfactory es-
say were
It should be no more than
one page long
It should be professionally
formatted and presented
It should be well written (at
least free of typographical
and grammatical errors).

Other than giving students
ample opportunity to write, the
essay assignments were intended
to encourage them to be reflec-
tive, a point that is best demon-
strated by the following ex-

0 "Describe two accomplishments that you are most proud
of and explain briefly why. These accomplishments can
be things you did at home, at school, in a job, or in a
"State two of your short-term and two of your long-term
goals. Describe the extent to which these goals have
been influenced by others (your family, friends, or teach-
ers). "
"Identify two items in our discussion on savings and
investmentfrom which you gained the most. For each of
them, indicate how you plan to use this information to
shape your behavior when you start your career."
> "Think of an instance in your life when miscommunica-
tion has cost you time, money, or peace of mind. Analyze
the situation and describe the cause for miscommunica-
tion and how it could have been avoided."
"Identify an idea that meant the most to you from our
discussion in class. Give an example of how you will use
that idea in order to achieve success in your career.
Indicate specific changes you plan to make as well as
how you will continue, or reinforce, existing behavior."
Time Commitment
Students spent, on average, less than two hours a week
outside of class on the reading and writing assignments. For
myself, the time spent on developing the course was compa-
rable to that needed for a typical technical course on a per-
credit basis, but it actually took less time to teach the course
than it does a technical course because there were no prob-
lem sets or exams.


Reading the students' essays each week was one of the
most enjoyable grading experiences I have had as a teacher.
First of all, it was clear from the essays that most of the
students paid attention and followed the discussion in class.
More importantly, students were uniformly open and honest
in expressing their views and sharing their life experiences.
In fact, I came to know many students in the class very well
by reading their reflective essaysquite an accomplishment
in a course where I met with the fifty-two students only one
contact hour a week. Throughout and after the course, many
students asked me to write letters of recommendation for
them-another indication that the format of the course pro-
moted student-faculty interaction.

At the end of the course, 47 out of the 52 students received
A's. Although a student could miss one class and one assign-
ment and still get an A, 21 of them did not lose a single
point! I found this to be encouraging because it showed that

students could and would meet high expectations that were
clearly articulated to them.

Students' Comments
I conducted a mid-semester evaluation and found that at
seven weeks into the semester, most of the students felt that
they understood the goals of the course and what was ex-
pected of them. They also found continuity among the lec-
tures, the reading assignments, and the homework. At the
formal end-of-semester course evaluation, the students gave
an overall course rating of 4.5 out of 5 (for comparison, the
department average is 4.0 and the university average is 4.1).
Perhaps more telling are the students' (voluntary) comments,
some of which are reproduced below.
"The course really prepared me for the professional
life. I really believe the material from the course is
worth keeping for life."
"Ifelt that this course has helped me grow as a
person and has significantly aided me in learning how
to be successful in my job search and in my career."
"This course must definitely stay since it really
helped in getting to know what the real world is like."
"I liked this course because we discussed things that
we never learn in other classes. I definitely think I
benefited from this class."
"I thought this class was a really good use of our
time. This has been my most beneficial class so far in
Students also made specific recommendations for areas
that they felt needed improvement. For example, many of
them found the textbook not to be very useful. In retrospect,
I could have chosen a book intended for an audience closer
to chemical engineering juniors at Carnegie Mellon (for
example, reference 4). And despite my conscious attempt
not to lecture in class, many students felt that there should be
more time spent on discussion, especially among the stu-
dents themselves. I later realized that most of the discussion
did take place between students and myself.

Pre-Test and Post-Test
The critical question, of course, is whether the course
helped students to learn. For that purpose, I conducted a pre-
test and post-test by asking the students to answer the fol-
lowing questions both at the beginning and at the end of the
El What are the essential skills necessary for a successful
career in engineering?
E[ What personal qualities can enhance or detract one's
El How do you set goals and prioritize?
El What time management strategies do you use, if any?
El What communication strategies can enhance relation-

Chemical Engineering Education

ships? Hinder relationships?
E What is networking (in the human sense) and how can
you best accomplish it?
El What should you do before, during, and after an inter-
Note that these questions were designed to test only for
increased student awareness of the important skills (which,
after all, was the course's main goal) and not for actual
mastery of these skills. Furthermore, these questions do not
provide a direct correlation to all the course objectives stated
earlier. Despite these caveats, an analysis of students' an-
swers (done by an independent, objective assessor'61) showed
that the course was most effective in increasing student
awareness of issues related to the interview process, goal
setting and prioritization, and effective communication; it
was less effective in helping students to identify the skills
and qualities that would most help them to be successful in
their careers, and in teaching them networking skills; and it
was least effective in conveying information regarding time
management strategies.
In most cases there is a correlation between a course's
effectiveness and students' prior knowledge (as shown in the
pre-test). In other words, if students have good prior knowl-
edge about a particular topic (such as time management),
then the course was less likely to enhance their knowledge in
that area. This finding may be simple common sense, but it
highlights the importance of assessing prior knowledge in the
planning stage of such a course. Overall, the course did add
some value, even though many of the students had at least
some good prior knowledge on every pre-/post-test question.

I must admit that I approached the teaching of this course
with some trepidation, not knowing how well chemical engi-
neering students would react to a "non-technical" course
offered by their home department, let alone a course that
involved numerous writing assignments. The fear turned out
to be unfounded. Students reacted very positively to the
discussion of issues that are relevant to their study, their
lives, and their careers. They participated actively in class
discussion, wrote openly about their aspirations and fears,
and were delighted to have the opportunity to place their
education in context. I encourage other departments to
consider offering such a course to their students. My course
syllabus can be found at

1. Carter, C., S. L. Kravits, and P. S. Vaughan, The Career
Tool Kit: Skills for Success, Prentice Hall (1995).
2. Harris, C., Hired! The Job-Hunting/Life-Planning Guide,
Prentice-Hall (1996).
3. Wilkes-Hull, M., and C. B. Grosswait, Professional Develop-
ment: The Dynamics of Success, Fifth Edition, Wadsworth
Summer 1998

4. Goldberg, D.E., Life Skills and Leadership for Engineers,
McGraw-Hill (1995).
5. Kravetz, S.,Welcome to the Real World: You've Got an Edu-
cation, Now Get a Life! Norton (1997).
6. L. M. Naples, LMN Evaluations (the cost for this assess-
ment was $600) O

BOOK REVIEW: Thermodynamics
Continued from page 223

and Bridgman relations are very concisely and logically developed
in Chapter 7, together with an excellent discussion of the Clapeyron
equation. Chapter 8 uses the derivations of the previous chapter to
develop relations for properties of real fluids using each of the PvT
relations of Chapter 3 to derive departure functions for both gases and
liquids. Graphical and equation representations are given. Property
tables and diagrams are briefly discussed.
Chapters 9 through 13 deal with phase equilibria. The first chap-
ter defines and calculates partial molar properties, chemical poten-
tials, and fugacity coefficients, the latter by applying the definition
to the various equations of state. Mixing rules and calculations of
both thermal and equilibrium properties for real fluid mixtures fol-
low. Chapter 10 discusses stability of equilibrium systems as well as
pure fluid phase transitions, vapor pressure, and the phase rule.
Properties of solutions from ideal to very nonideal, simple phase
equilibria predictions, and the full Gibbs-Duhem equation and its
use, including derivation of excess free energy models for activity
coefficient correlation, are given in Chapter 11, together with activ-
ity coefficient prediction methods.
Chapter 12 discusses vapor-liquid equilibrium in a methodical
and logical way and is a high point of the book. Basic relations used
to equate fugacity for both low-pressure and high-pressure systems
are detailed with many examples. Tests for VLE thermodynamic
consistency are discussed. Qualitative discussions of both vapor-
liquid and vapor-liquid-liquid equilibria are discussed and illus-
trated. A short treatment of dilute solution laws and liquid-vapor-
solid solubilities is contained in Chapter 13.
Chemical reaction equilibrium is the subject of Chapter 14, which
discusses basic free energy-equilibrium constant reactions, homo-
geneous gas reactions and the effects of variables, adiabatic reac-
tions, and phase-rule analysis of and calculation of equilibrium for
simultaneous equilibrium reactions. A short discussion of simple
liquid phase and heterogeneous reactions concludes the chapter.
This chapter could be improved by including more material on
solid-gas reactions and a discussion of solution of simultaneous
reaction equilibria by free-energy minimization.
Appendices include pure component data properties from vari-
ous sources as well as thermodynamic data for steam and common
refrigerants. The pure component data section should be updated to
the data now accepted as the most accurate.
In summary, the book is a credit to the author and to his profes-
sion. In my opinion, it is definitely competitive with the leading
first textbooks in chemical engineering thermodynamics. Faculty,
students, and practitioners will all find material of value. The only
negative is that the book is softbound and poorly glued; my copy
split after very little use. O

["a =classroom


Georgia Institute of Technology Atlanta, GA 30332

Perhaps one of the most agonizing tasks associated
with an educator's work is that of assigning grades to
students. The task is somewhat complicated by the
traditional need to convert numerical scores into letter grades.
Ironically, letter grades are then typically combined in some
fashion and converted back to a numerical score (such as a
grade-point average, or "GPA"). Throughout this paper, the
latter quantity is understood to mean the conventional GPA
computed on the basis of an "A" letter grade being worth
four points, a "B" grade worth three points, etc.
While most experienced engineers (educators and practi-
tioners alike) would probably agree that a person's GPA in
college is quite irrelevant to the progress of one's career only
a few years down the road, the fact remains that the GPA is
often the single most important factor in evaluating a student's
performance. Both companies and graduate schools weigh
this quantity heavily in their respective selection processes.
And for the student still in school, the GPA is about the only
thing that can be pointed to for measurement of scholastic
performance (at least in most students' minds).
So, it behooves faculty members to at least appreciate this
all-consuming importance of grades, rightly or wrongly, in
the student perception of education. Until a better system
comes along, we must make the best of what is probably a
suboptimal grading methodology. Anything that tends to
minimize this distraction from real learning should be a
welcome tool to any educator. It is within this framework
that the following suggestions are offered.

Most professors have probably had the experience of be-
ing asked one or more of the following questions after re-
turning a quiz or examination to the students, graded on the
basis of 0 to 100 points (or some other numerical scale).
El Do you curve your grades?
El How do you curve your grades?
El What letter grade do I now have in the course?
E Do you do negative curving of your grades?
Using the term "curve" seems quite ubiquitous for most
students, even though professors often have difficulty in
responding to the above questions. The first two questions
are likely to be asked after a particularly difficult exam (one
with an average of around 50, for example), while the last
question often arises after a relatively easy exam with an
average student score of 80 or more.
The questions also seem to be asked more frequently in
sophomore-level or lower-division courses where letter grades
are not as high as they are in upper-division courses (junior-
and senior-level). But it is not unusual to see a good student
at any level drop a course (even a required one) after getting
a score of 60-70 on an exam with a class average of 50 or
less, for fear of "ruining" his or her GPA.

This instructor has found it useful to distribute some guide-
lines to students immediately after returning their first ex-
aminations. The guidelines are based on the mean (or aver-
age) and standard deviation of the actual distribution for a
given examination. The latter two quantities are also re-
ported to the students (in addition to the median and range).
My experience has been that questions regarding "curving"
of the numerical grades to letter grades are virtually elimi-
nated after distribution of the guidelines. Unfortunately, some
good students still feel compelled to drop the course (par-

Copyright ChE Division of ASEE 1998

Chemical Engineering Education

Jude T. Sommerfeld is Professor in the School
of Chemical Engineering at Georgia Tech. He
received his BChE from the University of Detroit
and his MSE and PhD degrees, also in chemical
engineering, from the University of Michigan.
His industrial and academic experience has been
primarily in the area of computer-aided design,
and he has published over 100 articles in this
and other areas.

While most ... would probably agree that a
person's GPA in college is quite irrelevant to the
progress of one's career only a few years down
the road, the fact remains that the GPA is
often the single most important factor in
evaluating a student's performance.

Guidelines for Converting Numerical Quiz Scores to Letter
Grades in Sophomore ChE Courses
[M=class average (or mean); a =standard deviation]

Quiz Score
Greater than (M + 1.0 a)
Between (M + 0.2 a ) and (M + 1.0 a)
Between (M 1.0 a) and (M + 0.2 a )
Between (M 2.0 a)and(M 1.0 )
Less than (M 2.0 a )

Class GPA

Normal Quality
Distribution % Points

15.9 0.636
26.2 0.786
42.0 0.840
13.6 0.136
2.3 0.000

Guidelines for Converting Numerical Quiz Scores to Letter
Grades in Junior ChE Courses
[M=class average (or mean); a =standard deviation]

Normal Quality
Distribution % Points

Quiz Score

Greater than (M + 0.6 a) 27.4
Between (M 0.2 a) and (M + 0.6 a) 30.5

Between (M 1.5 a ) and (M 0.2 )
Between (M 2.25 a) and (M 1.5 a )
Less than (M 2.25 a)
Class GPA


35.4 0.708
5.5 0.055
1.2 0.000

Guidelines for Converting Numerical Quiz Scores to Letter
Grades in Senior ChE Courses
[M=class average (or mean); a -standard deviation]

Normal Quality
Distribution % Points

Greater than (M + 0.5 a)
Between (M 0.4 a ) and (M + 0.5 a )
Between (M 2.0 a ) and (M 0.4 a )
Between (M 2.5 a) and (M 2.0 a )
Less than (M 2.5 a )
Class GPA

30.8 1.232
34.7 1.041
32.2 0.644
1.7 0.017
0.6 0.000

ticularly if they know it is being offered again in the follow-
ing quarter).
Separate guidelines have been developed for sophomore-,
junior-, and senior-level courses in chemical engineering.
They are presented in Tables 1, 2, and 3, respectively. The
normal, or Gaussian, distribution was not specifically as-
sumed in the development of these guidelines. The distribu-
tion percentages given in the tables were obtained from
tabulating integrated values for the standard normal distribu-
tion; that is, one with a mean value of zero and standard
deviation of unity. The class averages presented in the tables
are then derived from the distribution percentages. Such
tabulations can be readily found in many mathematical hand-
books"' and textbooks on statistical methods.[2] Thus, for
example, in Table 1 the probability of or percentage of
sampled values greater than one standard deviation above
the mean (corresponding to a grade of "A" in a sophomore-
level class) would be equal to 15.9%.
The guidelines for sophomore-level courses yield a class
GPA of about 2.4 (on the 4.0 basis mentioned earlier), as
indicated in Table 1. This class GPA compares with values
of 2.77 and 2.93 for junior- and senior-level courses (Tables
2 and 3, respectively). The lower value for sophomore-level
courses reflects the fact that many students in these courses
are still maturing and fleshing out their career options and
thus may not be that suited for (or interested in) chemical
engineering. Typically, such courses would include material
balances, energy balances, numerical methods, and fluid
The actual relationship between these guidelines for the
three class levels and the normal distribution is depicted in
Figure 1. The curve plotted therein is for the standard normal
distribution with a mean value of zero and a standard devia-




S 01 Mean

-0.1. F D.-- C B-> A Soph.

-0.2 F4-J- C --C -B ->A Junior

-0' FED+ -C -i+ B A Senior

-3 -2 -1 0 1 2 3


Figure 1.



Grade Quiz Score


Summer 1998

tion of unity. One can clearly see the reduction in the cutoff
point (number of standard deviations above the mean) be-
tween "A" and "B" grades in the progression from sopho-
more to senior level and, consequently, the increase in their
frequency of occurrence (insofar as they lie to the right of
the mean). Similarly, in this progression the frequency of
occurrence of "D" and "F" grades, being to the left of the
mean, is seen to decrease.
The class averages computed for upper-division courses in
Tables 2 and 3 are somewhat closer to each other in addition
to being higher than those for sophomore courses in Table 1.
In all three cases, however,
the computed class GPAs
are probably slightly lower TAB
than the actual grade distri- Distribution of Letter G
bution resulting in most Computer-Aide
courses. In a sense, this is
good news for an instructor Letter Grades
insofar as dashing of stu- Year Quarter A B
dents' expectations is mini-
mized (better this situation 1991 Fall 2 7
than the reverse). By way 1992 Spring 6 9
of comparison, the median 1993 Winter 10 25
GPA of all junior chemical 1993 Fall 2 8
engineering students at 1994 Spring 13 20
Georgia Tech (173 total) 1995 Winter 14 24
enrolled in the fall quarter 1995 Fall 7 9
of 1997 was 3.05, versus a 1996 Winter 8 15
computed junior course
GPA of 2.77 from Table 2.
It should also be kept in 1997 Winter(A) 15 11
mind, of course, that with 1997 Winter (B) 6 10
the greater range below this 1997 Fall 8 5
median than above it, the
mean class GPA would be Totals 99 153
somewhat lower than the Percentages 28.37 43.84


There has been no attempt to optimize the numerical coef-
ficients in Tables 1 through 3 in order to achieve a particular
shape of distribution. My primary purpose in presenting this
information to students is to answer legitimate questions and
to allay unnecessary fears. Clearly, if one wished, much
effort could be devoted to adjusting these parameters, per-
haps to match historical or even projected data-but that is
not the focus of this paper.
Nonetheless, one detailed comparison of these guidelines
with experience may be interesting to the reader. Table 4
presents grade results from a dozen different offerings of a
senior-level computer-aided process design course offered
at Georgia Tech. These data span the 7-year period from
1991 through 1997 and pertain to a total of 349 students. The

class size in these various sections varied from a low of 11 to
a maximum of 53, with an average of 29 students.
The data in Table 4 indicate course GPAs ranging from
2.88 to 3.21. The average GPA for all students in all of the
sections is 2.99, compared to the value of 2.93 suggested by
the Table 3 guidelines. Also, the percentages of "A" and "C"
grades from the 7-year history given in Table 4 are both
slightly less than those resulting from the Table 3 guidelines,
while the percentage of "B" grades actually given is some-
what higher. There have been no "F" grades given in this
course during the 7-year period, and the small percentage of
"D" grades (1.7%) is exactly in
conformance with Table 3.

4 One clear benefit of these
es in a Senior-Level guidelines cannot be overempha-
sign Course sized. In assigning letter grades
at the end of a term, most faculty
r thereof) Total Class probably spend some time ago-
D F Students GPA nizing over the cutoff points be-
tween "A" and "B" grades and
1 0 11 2.91
1 0 1 2.91 between "B" and "C" grades. It
0 0 23 2.91 is not uncommon to hear from
1 0 50 2.88 the highest "B" or highest "C"
0 0 11 3.09 student at the beginning of the
0 0 46 3.00 next term, with a view to pos-
2 0 53 2.94 sible grade negotiation. I have
0 0 23 3.00 found these guidelines useful both
0 0 30 3.03 in the initial assignment of letter
1 0 27 2.93 grades and in any subsequent
0 0 34 3.21 grade-negotiation discussions.
1 0 23 2.91 One last benefit to the entire
0 0 18 3.17 educational enterprise should be
noted. Assuming that distribution
6 0 349 2.99 and discussion of the guidelines
1.72 0.00 100 with the students is performed
early in a term (about the time,
say, of the first quiz), it serves as
a pep talk or morale booster for
the students. This is particularly true with sophomore stu-
dents who are still accustomed to more standardized tests,
such as found in college freshman courses or high school
SAT examinations. Rampant dropping of courses, particu-
larly required ones, is an expensive exercise for students
(and parents?), not to mention the drain on institutional
resources. My experience with publicizing the guidelines
has been reduced student withdrawal rates from required
chemical engineering courses.

1. Spiegel, M.R., Mathematical Handbook of Formulas and
Tables, McGraw-Hill, New York, NY (1968)
2. McNeese, W.H., and R.A. Klein, Statistical Methods for the
Process Industries, ASQC Quality Press, Milwaukee, WI
(1991) 0
Chemical Engineering Education

d Des






The following 163 departments contribute to the support of CEE with bulk subscriptions.

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

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

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

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


The Engineering
Sof Chemical

Lanny D. Schmidt
S This text focuses on the analysis
of chemical reactors while
simultaneously providing a
description of industrial chemi-
cal processes and the strategies by which they oper-
ate deal for upper-level undergraduate courses in
chemical reactor engineering and kinetics. It provides
a concise up-to-date alternative to similar texts In
addition to the analysis of simple chemical reactors, it
considers more complex situations such as multistage
reactors and reactor-separation systems. The evolu-
non of chemical engineering from petroleum refining,
through petrochemicals and polymers, to new appli-
cations is described so that students can see the rela-
tionships between past, present, and future technolo-
gies Numerical methods are used throughout to con-
sider more complex problems. Worked examples are
given throughout the text and over 300 homework
problems are included
I 1 .II 1

Mathematical Methods in
Chemical Engineering
Arvind Varma and
Massimo Morbidelli
This text builds on students knowledge of calculus, lin-
ear algebra and differential equations. employing
appropriate examples and applications from chemi-
cal engineering to illustrate the techniques Numerous
high-qualiTy diagrams and graphics support the con-
cepts and solutions Many examples are included
throughout the text ana a large number of well-con-
ceived problems at the end of each chapter rein-
force the concepts presented The topical coverage
provides a good balance between material which
can be taught in a one-year course and the tech-
niques that chemical engineers need to know to
effectively model analyze and carry out numerical
simulations of chemical engineering processes, with
an emphasis on developing techniques which can be
used in applications.
' : I di : C .: ll-,
T, LI .

Yh~~ d ~

ll lld -n


Introduction to
Theoretical and

Fluid Dynamics
C. Pozrikidis
"There Is clearly a great deal of
material covered here, and
covered well.. Fantastic as a

reference....A text that any serious student of fluid
dynamics would like to own "
-Michael D. Graham Chemical Engineering
This is the first text to combine theoretical and com-
putational aspects of fluid dynamics in a unified and
comprehensive treatment The theoretical develop-
ments are carried Into the realm of numerical compu-
tation, and the numerical procedures are developed
from first principles The book simultaneously illustrates
the application of numerical methods to solving a
broad range of problems drawn from diverse areas
and discusses the development of pertinent compu-
tational algorithms it assumes no prior experience in
computational fluid dynamics and provides refer-
ences for specialized topics.


Polymer Chemistry
An Introduction
Third Edition
Malcolm P. Stevens
Ideal for students and Industrial chemists who work
with polymers this text provides a comprehensive
and up-to-date introduction to the chemistry of
macromolecular substances, with particular emphasis
on their synthesis and the properties underlying their
commercial importance The text covers significant
recent developments in modern polymer science
Malor topics include polymer synthesis nomenclature
molecular weight reactions of polymers, recycling of
polymers methods used for characterizing and test-
ing polymers morphology stereoregular polymers.
polymer blends, heterocyclic polymers inorganic
polymers, and natural polymers
' I,_, e--. --r 1998 ." : 95 I ,

To be cosdee fo r 00 exmiato 00 py writ on scoo letterhea giin
course~~~~ name exece enrllmnt and 0ecsio dae*oCleeeaeoriaor
OXOR UNIERIT PRS 198 MaioAeu Ne Yok0 Y 01
Toodrb crdi card cal 0ole 1-80 5175 Price and puliato dae subec to~ change
Brws or orde onln htt://ww 6up-sarg

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