Front Cover
 Front Matter
 Table of Contents
 Chemical Engineering at Polytechnic...
 Joseph Reynolds of Manhattan...
 The Chemical Engineering behind...
 The Devil's in the Delta
 An Internet-Based Distributed Laboratory...
 A Realistic Experimental Design...
 Forced Convection Heat Transfer...
 Future of Chemical Engineering:...
 Turning New Faculty Members into...
 Incorporating Six Sigma Methodology...
 Book Review
 Introducing Non-Newtonian Fluid...
 Implementation and Analysis of...
 Back Cover

Chemical engineering education
http://cee.che.ufl.edu/ ( Journal Site )
Full Citation
Permanent Link: http://ufdc.ufl.edu/AA00000383/00170
 Material Information
Title: Chemical engineering education
Alternate Title: CEE
Abbreviated Title: Chem. eng. educ.
Physical Description: v. : ill. ; 22-28 cm.
Language: English
Creator: American Society for Engineering Education -- Chemical Engineering Division
Publisher: Chemical Engineering Division, American Society for Engineering Education
Publication Date: Winter 2007
Frequency: quarterly[1962-]
annual[ former 1960-1961]
Subjects / Keywords: Chemical engineering -- Study and teaching -- Periodicals   ( lcsh )
Genre: serial   ( sobekcm )
periodical   ( marcgt )
Citation/Reference: 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 rights reserved by the source institution and holding location.
Resource Identifier: oclc - 01151209
lccn - 70013732
issn - 0009-2479
sobekcm - AA00000383_00170
Classification: lcc - TP165 .C18
ddc - 660/.2/071
System ID: AA00000383:00170

Table of Contents
    Front Cover
        Front Cover
    Front Matter
        Front Matter 1
        Page iii
    Table of Contents
        Page 1
    Chemical Engineering at Polytechnic University
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
    Joseph Reynolds of Manhattan College
        Page 10
        Page 11
        Page 12
        Page 13
    The Chemical Engineering behind How Pop Goes Flat: A Hands-on Experiment for Freshmen
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
    The Devil's in the Delta
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
    An Internet-Based Distributed Laboratory for Interactive ChE Education
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
    A Realistic Experimental Design and Statistical Analysis Project
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
    Forced Convection Heat Transfer in Circular Pipes
        Page 39
        Page 40
        Page 41
        Page 42
    Future of Chemical Engineering: Integrating Biology into the Undergraduate ChE Curriculum
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
    Turning New Faculty Members into Quick Starters
        Page 51
        Page 52
    Incorporating Six Sigma Methodology Training into Chemical Engineering Education
        Page 53
        Page 54
        Page 55
        Page 56
    Book Review
        Page 57
        Page 58
    Introducing Non-Newtonian Fluid Mechanics Computations with Mathematica in the Undergraduate Curriculum
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
    Implementation and Analysis of Hemodialysis in the Unit Operations Laboratory
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
    Back Cover
        Back Cover 1
        Back Cover 2
Full Text

chemical engineering education

Joseph Reynolds
... of Manhattan College


u Random Thoughts: Turning New Faculty Members Into Quick Starters (p. 51)
V1 Brent, Felder
M Incorporating Six Sigma Methodology Training into Chemical Engineering Education Ip. 53)
� Dal
F- a An Internet-Based Distributed Laboratory for Interactive ChE Education (p. 24)
c< Guo, Kettler, AI-Dahhan
� C The Chemical Engineering Behind How Pop Goes Flat: A Hands-On Experiment for Freshmen (p. 14)
.! n " Hohn
> -
S ._ The Devil's in the Delta (p. 19)
0 - E Luyben
-r ^Implementation and Analysis of Hemodialysis in the Unit Operations Laboratory (p. 65)
o.- Madihally, Lewis
C- 0
2 i Future of Chemical Engineering: Integrating Biology Into the Undergraduate ChE Curriculum (p. 43'
LU Mosto. Savelski, Farrell. Hecht
S A Realistic Experimental Design and Statistical Analysis Project (p. 31)
E -c Muske, FMyers
- u Forced Convection HeatTransfer in Circular Pipes (p. 39)

SPolytechnic University

aMg -teaching tips

This one-pagecolumn will present practical teaching tips in sufficientdetail thatChE educators can
adopt the tip.The focus should be on the teaching method, not content. With no tables orfigures
the column should be approximately 450 words. If graphics are included, the length needs to be
reduced.Tips that are too long will be edited to fit on one page. Please submit a Word file to Phil
Wankat , subject: CEE Teaching Tip.


Purdue University * West Lafayette, IN 47907

Both industry and ABET require that engineering gradu-
ates can communicate. Clearly the best way to achieve this
is to have frequent assignments throughout the curriculum
requiring writing and oral presentations. Unfortunately, oral
presentations tend to require a significant amount of class
time. An alternative oral presentation is the "elevator talk."
The scenario: a student steps into an elevator with someone
she needs to persuade or sell. For example, the student may
want to convince the person to hire her. She has from one to
two minutes to do this.

dents (askfor a
the time (two
minutes), gave
them a copy of
the scoring ru-
bric (Table 1),
and told them
to prepare a
talk that they
will present
ously, without
visuals. There
was no written
assignment. In
class, assigned
the "boss" for
each person.
Students were

told to assume that they knew the boss well enough to talk
to. Presenters and bosses went to the front of the room and
stood in the elevator. Talks were timed for a strict two min-
utes. Since two minutes is actually fairly long, most students
finished early and had to do something-perhapsjust stand
there-for the remaining time. If they weren'tfinished attwo

minutes, the elevator door opened anyway and they had to
summarize very quickly.

The students saw the relevance of elevator talks and were
well prepared. Grading the talks with the scoring rubric was
straightforward and I was able to finish the grading while
the next pair walked to the front. Since it takes less than 30
secondstochangespeakers, 20two-minutetalkscan bedone
in a 50-minute period.

While not eliminating the need for more formal presen-

stations, eleva-
tor talks can
provide an
easy way to
include oral
tion in courses
that normally
would not
have time.
Grading all of
the talks with
the scoring ru-
bric and then
saving cop-
ies provides
evidence for
ABET that all
students have
been assessed
and can do
oral presenta-

tions, at least at the barely acceptable level.

1. Mitchell, B.S., and VJ. Law, "Community-Based Presentations in the
Unit Ops Laboratory,"Chem. Eng. Ed., 39(2), 160 (2003)

SCopyright ChE Division of ASEE 2006

Scoring Rubric for ElevatorTalks. Adapted from Mitchell and Law.!'
Attribute Not Barely Meets Exceeds
Acceptable Acceptable Expectations Expectations
Logical topic Disjointed; no Parts out of Organized by Superior;
order organization order guidelines enhancescom-
Appropriate Far too long or Somewhat Appropriate
time use too short long or short length
Objective Not stated Poorly stated Clearly stated
Background & Neither stated Only one Both stated Both clearly
Significance stated stated
Conclusions None Present, but Logical & Logical& supe-
not logical clearly stated riorexplanation
Presentation Many Some No distractions Superior
mechanics* distractions distractions presentation
Response to Notresponsive Incomplete Clear and Complete
questions (if any) direct
Focus on person Not focused; Some focus; Focused with Totally
speaking to distracted, no some eye good eye focused; excel-
eye contact contact contact lenteyecontact
*voice, poise, mannerisms

Author Guidelines for the



The laboratory experience in chemical engineering education has long been an integral part
of our curricula. CEE encourages the submission of manuscripts describing innovations in the
laboratory ranging from large-scale unitoperations experimentsto demonstrationsappropriate
for the classroom. The following guidelines are offered to assist authors in the preparation of
manuscripts that are informative to our readership. These are only suggestions, based on the
comments of previous reviewers; authors should use their own judgment in presenting their
experiences. A set of general guidelines and advice to the author can be found at ourWeb site:

c Manuscripts should describe the results of original and laboratory-tested ideas.
The ideas should be broadly applicable and described in sufficient detail to
allow and motivate others to adapt the ideas to their own curricula. It is noted
that the readership of CEE is largely faculty and instructors. Manuscripts must
contain an abstract and often include an Introduction, Laboratory Description,
Data Analysis, Summary of Experiences, Conclusions, and References.
* An Introduction should establish the context of the laboratory experi-
ence (e.g., relation to curriculum, review of literature), state the learning
objectives, and describe the rationale and approach.
* The Laboratory Description section should describe the experiment in
sufficient detail to allow the reader to judge the scope of effort required
to implement a similar experiment on his or her campus. Schematic dia-
grams or photos, cost information, and references to previous publica-
tions and Web sites, etc., are usually of benefit. Issues related to safety
should be addressed as well as any special operating procedures.
If appropriate, a Data Analysis section should be included that concisely
describes the method of data analysis. Recognizing that the audience
is primarily faculty, the description of the underlying theory should be
referenced or brief.The purpose of this section is to communicate to the
reader specific student-learning opportunities (e.g., treatment of reac-
tion-rate data in a temperature range that includes two mechanisms).
* The purpose of the Summary of Experiences section is to convey the
results of laboratory or classroom testing. The section can enumerate,
for example, best practices, pitfalls, student survey results, or anecdotal
* A concise statement of the Conclusions (as opposed to a summary) of
your experiences should be the last section of the paper prior to listing

( Ih nr, all I. unimi i ml, Education
Department of Chemical Engineering
University of Florida * Gainesville, FL 32611
PHONE and FAX : 352-392-0861
e-mail: cee@che.ufl.edu

Tim Anderson

Phillip C. Wankat

Lynn Heasley

James O. Wilkes, U. Michigan

William J. Koros, Georgia Institute of Technology

Nicholas Rosinia


John P. O'Connell
University of Virginia

E. Dendy Sloan, Jr.
Colorado School of Mines

Kristi Anseth
University of Colorado
Thomas F. Edgar
University of Texas atAustin
Richard M. Felder
North Carolina State University
H. Scott Fogler
University of Michigan
Carol K. Hall
North Carolina State University
Steve LeBlanc
University of Toledo
Ronald W. Rousseau
Georgia Institute of Technology
C. Stewart Slater
Rowan University
Donald R. Woods
McMaster University

Winter 2007

Chemical Engineering Education
Volume 41 Number 1 Winter 2007

2 Chemical Engineering at Polytechnic University
Edward N. Ziegler, Jovan Mijovic

10 Joseph Reynolds of Manhattan College
Helen C. Hollein

51 Turning New Faculty Members Into Quick Starters
Rebecca Brent, Richard M. Felder

14 The Chemical Engineering Behind How Pop Goes Flat: A Hands-On
Experiment for Freshmen
Keith L. Hohn
31 A Realistic Experimental Design and Statistical Analysis Project
Kenneth R. Muske, John A. Myers
39 Forced Convection Heat Transfer in Circular Pipes
Ismail Tosun
53 Incorporating Six Sigma l .I. ...I. I. Training into Chemical Engi-
neering Education
Lenore L. Dai

43 Future of Chemical Engineering: Integrating Biology Into the Under-
graduate ChE Curriculum
Patricia Mosto, Mariano Savelski, Stephanie H. Farrell, Gregory B. Hecht

19 The Devil's in the Delta
William L. Luyben
24 An Internet-Based Distributed Laboratory for Interactive ChE Education
Jing Guo, David J. Kettler, Muthanna Al-Dahhan
65 Implementation and Analysis of Hemodialysis in the Unit Operations
Sundararajan V Madihally, Randy S. Lewis

59 Introducing Non-Newtonian Fluid Mechanics Computations With
Mathematica in the Undergraduate Curriculum
Housam Binous

57 Book Review

CHEMICAL ENGINEERING EDUCATION (ISSN 0009-2479) is published quarterly by the Chemical Engineering
Division, American Societyfor EngineeringEducation, and is edited at the University ofFlorida. 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 0 2005 by the Chemical Engineering Division, American Society for
EngineeringEducation. 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 ifnotified within
120 days of publication. Writefor information on subscription costs andfor back copy costs andavailability. POSTMASTER:
Send address changes to Chemical Engineering Education, Chemical Engineering Department, University of Florida,
Gainesville, FL 32611-6005. Periodicals Postage Paid at Gainesville, Florida, and additional post offices (USPS 101900).

rj[ department
---- U s_____________________________________

Chemical Engineering at

Polytechnic University

T he Brooklyn Colle-
T giate and Polytechnic A
Institute was chartered
in 1854, when the city of . .
Brooklyn's rapidly growing
population was 30,000 and
Brooklyn was separate from :
New York City. This was
roughly 30 years before the
completion of the Brooklyn
Bridge and prior to the Civil '
War. The stated purpose of the
first Polytechnic Board was to
establish "an educational insti-
tution in our midst,... to give 7 0
our sons an education as would i ".
qualify them in a far higher ':
degree, through an enlarged, "' i
liberal, and thorough training i
in a course of practical, scien-
tific, and classical studies, to
enter upon the active pursuits
and duties of life," and "that its .. !
location should be as central, -'
and as easily accessible as pos- i
sible by public conveyance, i
from all parts of the city ...."
In its earliest years, the college
drew students from the man-
sions and substantial homes of
the "Heights," the "Hill," the
"Eastern District," and other
parts of Brooklyn.
The Scientific Program lead-
ing to the Bachelor of Science
degree was established for
those bent in the direction A preserved picture of life
of science and engineering,
which in addition to theory included more than "200 laborato-
ry experiments, field trips, and exercises." Between 1885 and
1890 the "Scientific" course of study was ultimately divided

r ..:."

at Polytechnic's main campus at the beginning of last century.

into three areas of specialization: Engineering (Mechanical
and Civil), Electrical Engineering, and Chemistry. The latter
had offerings in applied and fundamental areas

@ Copyright ChE Division of ASEE 2007

Chemical Engineering Education

In 1898, Brooklyn became a part of New York
City. Like Brooklyn, Polytechnic's services and
influence have gone far beyond the borders of
the Borough, through the university's worldwide
contributions to science, engineering, and educa-
tion. Polytechnic's modem Brooklyn site is still
only two blocks from the Brooklyn Bridge, and
all three of the city's major subway systems have
stations within a few blocks of the Polytechnic,
maintaining the spirit of its original charter.
Chemical engineering at Polytechnic University
had its formal beginnings more than a century ago
when the Department of Chemistry became the
Department of Chemistry and Chemical Engineer-
ing at the Polytechnic Institute of Brooklyn, or
PIB. I.W. Fay was the first head of the combined
department in 1905, with only one chemical engi-
neer on the staff-John C. Olsen. In those days,
extensive use was made of eminent professionals
in local industries as "consulting professors."
In 1925, the chemical engineering program at Polytechnic
became one of the first engineering programs to receive
accreditation by a national professional society, when the
American Institute of Chemical Engineers (AIChE) listed it
among the first 15 accredited curricula in chemical engineer-
ing. In 1931, a separate Department of Chemical Engineering
was established, with Olsen as its first head. That same year
Olsen was elected president of the AIChE, which he helped
found (and for which he served as secretary for its first 23
years). Since then, more than 2,800 bachelor's, 1,000 master's,
and 350 doctoral degrees have been awarded in chemical
engineering at Poly.

In 1932, then-28-year-old chemical engineer Donald F.
Othmer was hired into Olsen's department, after an impressive
five years of work at Eastman Kodak in Rochester, N.Y.
Kodak was the world's largest manufacturer of cellulose
acetate, a key ingredient in photographic film. A critical step
in creating cellulose acetate is concentrating the acetic acid
used in production. While in Kodak's employ, Othmer was
tapped to find ways to increase the acetic acid concentration
in various sources available to the company. Initially knowing
little about the subject, but always curious, Othmer designed
an experimental device to observe how acetic acid is distilled.
The apparatus he built became famous as the Othmer Still
and continues to be used to study the properties of mixtures
being distilled. The early version of the still was typical of
Other's hands-on, low-cost approach to science: He not only
conceptualized and designed the apparatus, but also learned
glass-blowing so that he could build it himself.
The Othmer Still allowed chemists and engineers to mea-

The Chemical Engineering Laboratory about 1919.

sure accurately for the first time concentrations in the vapor
and liquid phases in equilibrium.
Other also contributed greatly to the science of azeotropic
distillation, which introduces a third chemical during the
distillation process to improve the purity of the product and
reduce energy consumption. Thanks to Othmer, distillation
is now a science. His geometrical and mathematical instincts
were applied further to devise a figure in which the effects of
temperature on vapor pressure of various compounds could
be correlated as straight lines on a single sheet of paper, the
now-famous Othmer plot.
As a well-known chemical engineer, Othmer succeeded to
the chairmanship in 1937 and remained head of the depart-
ment until 1961, when he stepped down to devote more time
to teaching and research. He has authored hundreds of articles
and held numerous patents for chemical engineering applica-
tions. Around 1945 Raymond E. Kirk, head of the Depart-
ment of Chemistry, and Othmer, heading the Department of
Chemical Engineering, decided to embark on a project as
co-editors of an encyclopedia that would be a comprehensive
guide to industrial chemistry and chemical engineering. The
Kirk-Othmer Encyclopedia of Chemical Technology is now in
its 5th edition and comprises 27 volumes. It is the first place
chemists and chemical engineers turn when they are starting
a new project. It has everything from the commonplace to
the esoteric, from how to make batteries and beer to how
to reduce nitrobenzene. A set may be found in the library of
virtually every major university in the world.
When Othmer died in 1995 he bequeathed more than $175
million dollars to the Polytechnic, which remains as of today
the largest donation ever given to the university. Much of the
gift went to improving and expanding the university labora-

Winter 2007

tory and classroom facilities, with some to construction of a
new dorm and gymnasium. An interesting side note: If the
remaining professors had joined Othmer 30 years earlier in
the investment club he started at Poly-based on the advice
of a family friend named Warren Buffet-they all would've
been rich; but even sharing a fraction of the membership
fee of $25,000 would have been rather difficult for a faculty
member in those days.
Don Othmer supervised and inspired more than 60 doc-
toral students, many of whom went on to distinguished
careers in their own rights. He supervised research in the
fields of thermodynamic property estimation, distillative
and extractive separations, fluidized bed design, and energy
optimization. Having no biological children, he was quoted
as saying he regarded himself most fortunate to have been
blessed with so many brilliant "academic children" whom he
could recognize with almost-paternal pride. One of Othmer's
former students was Ju Chin Chu, who from 1950 to 1966
supervised fundamental distillation experiments on more
than 100 industrially important chemical mixtures. Chu,
in turn, must have passed along a high regard for research
and genes as well: His son, Steven Chu, won the Nobel
Prize in Physics in 1997.
Speaking of Nobel Prize winners, a corecipient of the 1995
Nobel Prize in Physics, Martin L. Perl, earned his chemical
engineering bachelor's at Poly in 1948 (followed by a Ph.D.
from Columbia). Perl was honored for the discovery of the tau
lepton, one of nature's most remarkable subatomic particles with
a mass 3,500 times that of the electron. In 1982, Perl's promise
had already been recognized closer to home: That year, he was
awarded the Wolf Prize for Physics for the Class of 1948.
In the '50s Othmer was able, through fund raising and
departmental equipment gifts from industrial colleagues, to
persuade Warren L. McCabe to come to Poly and become
administrative dean. A leading educator and consultant for-
merly at Cornell University, McCabe is, of course, famous
for the McCabe-Thiele diagrams of binary distillation, as
well as being coauthor of McCabe, Smith, and Harriott's Unit
Operations of Chemical Engineering.
Othmer already had a master craftsman of laboratory
equipment on staff, W. Fred Schurig (Poly '33, '35, and
'46) who constructed one of the finest teaching laboratories
in America-Poly's Unit Ops Laboratory. While at Poly and
after he retired, Schurig designed and built laboratories for
many schools throughout the Americas. Schurig became
known for his discipline and attention to detail, which he later
attributed to Othmer's influence.
Perhaps the most famous of Othmer's doctoral students,
Joseph J. Jacobs, also earned all three of his degrees at the
Polytechnic, receiving his Ph.D. in 1942. Jacobs developed a
system that could manufacture soap in 15 minutes compared
to the traditional process that required between three and

seven days. Jacobs was an assistant professor at Polytechnic
for a while and then headed west to San Francisco to take
a position assisting in the engineering of liquid fertilizers.
After doing consulting work for two years at Kaiser Alumi-
num and Chemical Company-at which he helped develop
caustic soda-Jacobs started his own business. In 1947, he
founded Jacobs Engineering Group Inc., an international
firm that Fortune magazine ranked No. 1 in 1999 as the
most admired engineering and construction company. In ad-
dition to authoring numerous articles on chemical engineer-
ing and economics, Jacobs made substantial contributions
to the study of social issues-including aging parents of
adult children-and authored two autobiographies. He was
recipient of the United Engineering Society's 1983 Herbert
Hoover Medal, which recognizes the civic and humanitar-
ian achievements of professional engineers. The university
also established the Joseph J. and Violet J. Jacobs Chair in
Chemical Engineering, and in 2002 opened the Joseph J. and
Violet J. Jacobs Building on campus, housing a full gymna-
sium and athletic center as well as state-of-the-art laboratories
and classrooms.
Another well-known doctoral student of Othmer, Ger-
hard Frohlich, earned his Ph.D. in chemical engineering
at Poly in 1957. He was the second member associated with
the Polytechnic to become president of AIChE, elected in
1999. Earlier he had been named corporate vice president
and general manager of Central Engineering at Hoffman-La
Roche, where for many years he engaged in the development,
design, and construction of chemical and pharmaceutical
facilities. In commenting on the importance of AIChE,
Frohlich said, "We must think globally, accept cradle-to-
grave stewardship of products, and strive for sustainable
development. Professional societies can lead the way by
facilitating dialogue among industry, government, academe,
and the public. By working together in new and more flex-
ible ways, using renewable resources, and learning from
advances in chemistry and biotechnology, we can make
products that enhance the quality of life and protect the
environment. If we commit to doing so, the new millennium
looks bright indeed."
Yet another of Othmer's students, Robert F. Benenati, had
a long and successful career as a professor who challenged stu-
dents to do more than they ever thought possible, particularly
in his design class. Warren Seider, a Poly graduate now at the
University of Pennsylvania, is in turn one of Benenati's former
students, and is coauthor of the major design text Product and
Process Design Principles, now in its 2nd edition.
In the '60s and '70s James J. Conti (Polytechnic '54, '56,
and '59) and Irving F. Miller were department heads through
a financial crisis, in which PIB merged with the NYU school
of engineering to form the Polytechnic Institute of New York,
with its main campus remaining at the Brooklyn site. In 1985,
the school was renamed Polytechnic University.
Chemical Engineering Education

Below, Don Othmer poses with a gold-plated
version of the invention he created in 1928,
the Othmer Still. Left, Othmer is seen with
Raymond Kirk, head of Poly's chemistry
department circa 1945 and co-editor of the
pair's comprehensive encyclopedia, now in
its 5th edition.

Rounding out the era, Leonard Stiel has carried the Othmer
tradition into the computer age; his work is cited widely in
the literature of thermodynamic and transport properties of
fluids and mixtures. Stiel officially retired a few years ago
but now as a research professor he's still very active at Poly
in education and as a consultant.

In the early part of the 20th century, many prominent chem-
ists dismissed the idea that molecules with molecular weights
in the thousands or millions could exist. Today, polymers are
everywhere, in everyday materials such as plastics, nylon, and
rubber. The year 1939 marked the introduction of a polymers
course in the chemical engineering department. That year,
chemical engineering professor Paul F. Bruins joined Poly
from the University of Iowa and offered the first graduate
course in polymer technology in the United States, paving the
way for what has become one of the most famous polymer
programs in the world. Bruins was affectionately called the
"walking encyclopedia" of plastics, and he wrote and edited
extensively. He was known to take his colleagues for a spin
in his small aircraft during the day and return in time to teach
his polymer course in the evening.
Much of today's widespread acceptance of polymers, their
chemistry, and their engineering is the result of work by the
Polymer Research Institute (PRI) of Polytechnic University.
Herman Mark, a pioneer in the study of giant molecules,
established the PRI in 1964. The institute brought together a
Winter 2007

* number of polymer researchers to create the first
academic facility in the United States devoted
to the study and teaching of polymer science.
Many scientists associated with the institute later
went on to establish polymer programs at other
universities and institutions, contributing significantly to the
development and growth of what has become a vital branch of
chemistry, engineering, and materials science. Under Mark's
leadership, the institute became the premier U.S. destination
for polymer chemistry, attracting students from all over the
world. But its effect wasn't limited to simply establishing
the importance of polymer chemistry and contributing
many of its fundamental discoveries-like colonists, PRI
alumni went on to found a number of polymer institutions
at other locations.
The American Chemical Society (ACS) recognized the
institute's pioneering efforts by designating it a National
Historic Chemical Landmark. Such designations recognize
important places, discoveries, and achievements in the his-
tory of chemistry. Other landmarks have included Joseph
Priestley's Pennsylvania home, penicillin, and the National
Institute of Standards and Technology. PRI holds a special
place in ACS Past President Eli M. Pearce's heart, as from
1982 to 1996 he served as its director. Pearce started at Poly-
technic in 1973, had a joint appointment in the Departments
of Chemistry and Chemical Engineering, and is currently a
university research professor. Pearce is confident about PRI's
future: "When you read the [National Research Council report
'Beyond the Molecular Frontier'], it's clear that the most ex-
citing developments in science and technology are occurring
at the interfaces. Over the years, considerable contributions
were made to the engineering side of polymerization research
and education."

Two other Poly engineering faculty who have made long-
term contributions to the polymer-engineering field are Chang
Dae "Paul" Han, a former department head (1974-82), and
Jovan Mijovic, present department head. Han published
widely in polymer and chemical engineering journals, and
wrote two books: Rheology in Polymer Processing and
Multiphase Flow in Polymer Processing, both published by
Academic Press. Mijovic in the course of his illustrious career
has published widely in polymerjournals and supervised doz-
ens of doctoral students in the study of polymeric materials
properties and states, and more recently has been investigating
complex chemical and biosystem dynamics, nanotechnol-
ogy, and nano-materials. He has led the department into the
chemical and biological engineering era while continuing his
yeomanlike efforts as a dedicated, distinguished teacher and
researcher. He's committed to maintaining Poly's tradition
of excellence.

The Polytechnic has performed many research investiga-
tions concerned with the understanding of fluidized bed fun-
damentals. Fuidization is used widely in petroleum refining,
power generation, and in the chemical industry. Frederick
Zenz received his Ph.D. at Poly in 1961 and taught a graduate
course throughout the '60s entitled "Fuidization," eventually
writing and publishing the seminal work Fluidization and
Fluid Particle Systems with Othmer.
In the early portion of his career, Poly's Edward Ziegler
earned an international reputation for his research in fluidized

Acclaimed alumnus
.- _" Dr. Joseph Jacobs with a blueprint
-- - for one of the many important
projects in which he was a partici-
pant. In addition to establishing a
Joseph J. and Violet J. Jacobs Chair
in Chemical Engineering, in 2002 the
university opened the Joseph J. and
Violet J. Jacobs Building on campus.
U It houses a full gymnasium and ath-
letic center as well as state-of-the-art
Sl laboratories and classrooms.

bed transport and reaction engineering
M E R modeling. His heat transfer model is
used in the design of fluidized bed coal
combustors. Ziegler, along with former
Poly professor Rutton Patel (now with
I ExxonMobil) supervised students in
Sthe fluid bed research area, and the two
AMAX -. often became members of each other's
guidance committees.
Later Ziegler's interests turned toward
environmental applications and specifi-
cally air pollution engineering control. He has co-edited the
5th Edition of the Pfafflin-Ziegler Encyclopedia of Envi-
ronmental Science and Engineering, published in January
2006, and authored a number of its articles. He started on the
encyclopedia's first edition some three decades ago, together
with his co-editor James Pfafflin- a former member of Poly's
Department of Civil Engineering. Over his career Ziegler has
taught more than 1,000 students in the undergraduate lab, and
more than 800 graduate students, mainly in his Chemical Re-
actor Design and Air Pollution Engineering Control courses.
He's been the thesis and project adviser to numerous master's
and doctoral students as well as advising undergrads.
Starting in 1986 Allan Myerson headed the department
and eventually became dean of the School of Chemical and
Materials Science. Myerson encouraged interdisciplinary
studies between the engineering and science departments. He
also was active in crystallization and nucleation research and
edited the Handbook of Industrial Crystallization.

A major renovation of the chemical engineering lab took
place in 2001, when Walter P. Zurawsky's considerable
transport phenomena knowledge, research experience at
AT&T Labs, and equipment construction skills were put into
play. Professor Ziegler had been teaching the lab for many
years after his mentor Fred Schurig retired. With the help of
the Othmer gift to the school, Zurawsky and Ziegler planned
a student-friendly, state-of-the-art experimental teaching facil-
ity with new distillation columns, process control equipment, a
Chemical Engineering Education

controlled fermenter, and membrane separations experiments.
The new, highly automated distillation experiment is, by the
way, currently used to investigate the efficiency of concentrat-
ing acetic acid (shades of Donald Othmer?) using sieve trays
and packed columns. The senior CBE students perform 20
experiments in their final two semesters. Many of the scaled-
down versions of traditional chemical engineering operations
were retained and are still used to study classical theories and
industrial correlations, but with modern instrumentation. A
computation room was fitted with the latest PCs having Lab
View, Microsoft, and MatLab software to help store, transmit,
and analyze the data. The ASPEN Engineering Suite is avail-
able to all CBE students on their local network, primarily for
use in the senior design courses.

Over the past 40 years, chemical engineering curricula have
embraced an engineering science paradigm that spans from
molecular-level interactions and transformations to large-scale
systems. Indeed, it is an appreciation of, and a willingness
to work over, many decades of scale that is one of the distin-
guishing traits of the chemical engineering discipline. This
ability to adapt to work on many scales has allowed chemical
engineers to have productive interactions with a wide range of
other science and engineering disciplines, and will be essential
for the application of engineering principles to biologically
based processes. The rising need to convert advances in biol-
ogy into new processes and new industries makes it imperative
that we adopt biology as an enabling science.
Interest in integrating biology and chemical engineering,
or CBE, is growing nationwide. For example, the number
of biologically oriented presentations at the AIChE annual
meetings increased from less than 10% to close to 50% in
only four years. Many chemical engineering departments
across the country have changed names to reflect a growing
interest in, and overlap with, biology (examples include Johns
Hopkins, Cornell, the University of Pennsylvania, the Univer-
sity of Wisconsin-Madison, Northwestern, and RPI). Many
such departments have started to require a biology course as
part of their curriculum, but there are still very few that have
made a full commitment to developing a curriculum in which
biological systems and processes are fully integrated across
the curriculum, as we are doing.
Several years ago the engineering faculty within our de-
partment reviewed and revised the chemical engineering
curriculum to reflect what it felt was the emerging importance
of biology. The specific aim was to develop an exemplary
educational program (B.Sc.) in chemical and biological engi-
neering that builds on the traditional strengths and paradigms
of chemical engineering while embracing biology as a pillar
along with mathematics, chemistry, and physics. So substan-
tive are the changes that we, too, undertook a program name
change from chemical engineering to chemical and biological
Winter 2007

engineering. The CBE program was initiated with the fresh-
man class of 2003.
We firmly believe that chemical engineering principles can
and must be applied to biological systems and to the develop-
ment of new processes based on biology. The task we face
in implementing this new curriculum is substantial, but we
are eminently qualified and confident of the success of the
proposed program.
The courses for our new CBE program are shown in Table
1 (next page). The program has been approved by the faculty
of our department, by the faculty of Polytechnic University,
and by the State of New York. By careful choice of electives
and several course substitutions, CBE students can adjust their
schedules to satisfy medical school requirements if they have
an interest in pursuing medicine as a career.
The task we face is to meld, as seamlessly as possible, sys-
tems and processes of biological relevance into our engineer-
ing curriculum. We regard the systems-oriented, multiscale
approach to problems that is the hallmark of chemical engi-
neering as the primary strength we have to offer. We believe
it is essential that our students remain strong in engineering.
It is our further belief that by exposing our students to biology
and bio-processes in addition to more conventional chemical
processes, we will produce better, more versatile engineers.
As part of our new curriculum, we have introduced re-
quired courses in biology and biochemistry and are revising
virtually all of our engineering courses to include biological
applications and examples. Technical electives in the junior
and senior year provide opportunities for elective courses,
particularly new electives focusing on engineering in biology
such as System Biology, Protein Engineering, and Drug De-
livery. Although these new elective courses will be primarily
aimed at CBE students, they will be open to other engineering
and science students.
We are the only chemical and biological engineering pro-
gram in New York City and we have seen phenomenal growth
in our undergraduate enrollment over the past two years: from
41 undergraduates in 2004 to 110 in early 2006- a whopping
150% increase. The CBE program is acknowledged as the
most demanding major on campus. A GPA of 2.5 is required
to remain in the major (2.0 elsewhere) and the students'
response has been hugely enthusiastic. We have had the high-
est percentage of students on the Dean's List and named as
valedictorians in recent years.

Polytechnic provides an important educational opportunity
for students who tend to be under represented in engineering.
Given our downtown Brooklyn location, our student popula-
tion has always included a large cross section of the population
of Brooklyn and the other boroughs of New York City. As
different ethnic groups have immigrated to the United States,

Chemical and Biological Engineering Curriculum

Freshman Fall Freshman Spring
MA1014 Calculus I MA1114 Calculus II
CM1004 Gen. Chemistry for Engineers Intro to Cell & Molecular Biology
EN1014 Writing & Humanities I CBE1214 Intro to Chem & Bio Engineering
EG1004 Intro Engineering & Design EN1204 Writing & Humanities II
Sophomore Fall Sophomore Spring
MA2012 Linear Algebra I MA2112 Multi-variable Calculus A
MA2132 Ordinary Differential Equations MA2122 Multi-variable Calculus B
PH1004 Introductory Physics I PH2004 Introductory Physics II
CBE2124 Analysis of Chem & Bio Processes CS1114 Intro to Prog. & Problem Solving
CM2234 Industrial Organic Chemistry CM2514 Chemical & Biological Equilibria
Junior Fall Junior Spring
CM3314 Biochemistry I CBE3324 Chem & Bio Separations
CBE3103 Math Methods for Chem & Bio Eng. CBE3214 Chem & Bio Reactor Engineering
CBE3314 Physical Rate Processes Technical Elective
HI2104 Modern World History HU/SS Elective
CBE3622 Chem & Bio Eng. Thermodynamics
Senior Fall Senior Spring
CBE4113 Engineering Laboratory I CBE4123 Engineering Laboratory II
CBE4413 Process Dynamics & Control CBE4623 Chem & Bio Process Design II
CBE4613 Chem & Bio Process Design I CBE4713 Engineering Polymeric Materials
HU/SS Elective Engineering Elective
Technical Elective HU/SS Elective

Poly's student popu-
lation has changed,
always mirroring the
ethnic mix of the city.

In addition to the Annual Income < $20k $20k to $80k > $80k
ethnic diversity that Polytechnic 28.4% 51.6% 20%
is part of Polytechnic,
SUNY 13.0% 31.0% 56%
we are proud to note
that over the past de-
cade nearly 50% of the students who have graduated from and Albert Einstein Coll
our chemical engineering program are female. We fully few. Adding to the list, a
expect that this trend will continue with our new program in on the grounds of Bellev
chemical and biological engineering. Although there have Manhattan) was announce
been advances nationwide, women are still grossly under- York Times a few weeks
represented in engineering. River Science Park will
Polytechnic is a private university, but our role in the New The facility is being deve
York region is, de facto, one that would be expected of a Equities, Inc.-whose ct
public university. As shown in Table 2, we educate a much Sudarsky, is a Polytechn
greater percentage of students from lower-income households Board of Trustees of Pol
than the state university system does. Washington Monthly cal institutions and the n
ranked Polytechnic University second in the nation (out of graduates with avenues fo
245 national universities) in social mobility, ties for collaboration, an

Our location in New York
provides us with exceptional
opportunities. We are in a
region with many excellent
biomedical institutions in-
cluding Rockefeller Univer-
sity, Memorial Sloan-Ketter-
ing Cancer Institute, SUNY
Downstate Medical Center,
lege of Medicine, to mention a
new $700 million science park
ue (just across the East River in
ed on the front page of The New
ago. The focus of this new East
be the biotechnology industry.
loped by Alexandria Real Estate
[airman of the board, Dr. Jerry
lic alumnus and a member of the
ytechnic. These excellent medi-
ew science park will provide our
r continued education, opportuni-
d potential employment.
Chemical Engineering Education

Student Family Incomes: Polytechnic and
the State University of New York

The explosive potential of CBE
has been recognized by the Poly-
technic trustees, our new president,
Zieer Jerry MacArthur Hultin, and the
Edward Ziegler members of the administration. These
Pivotal individuals have
made a major commit-
ment to our department's
continuing growth. Three
years ago, Jose M. Pinto
S joined our faculty from the
University of Sao Paulo
in Brazil. Pinto received
his Ph.D. from Carnegie-
Jovan Mijovic Mellon and is interested
in modeling and optimization of
chemical and biological processes
and systems biology. In fall 2004,
Stavroula Sofou joined the faculty.
S Sofou received her Ph.D. from Co-
l / lumbia University in New York City
SP and spent three years as a post-doc at
Jose Po the Memorial Sloan-Kettering Cancer
Research Center. Her principal
interest focuses on the use of
engineering principles for drug
delivery for cancer cure.
We announced two additional
faculty positions in fall 2006.
Stavroula Sofou Rasti Levicky, formerly of Co-
lumbia University, was named the Donald E Othmer Assistant
Professor of Chemical and Biological Engineering. Levicky
received his Ph.D. from the University of Minnesota and
has a strong interest in the field of biological polyelectrolyte
Winter 2007

systems, nanosized micro array biosen-
sors, and bio-diagnostics. Jin Ryoun Kim
is the Joseph J. and Violet J. Jacobs Assis-
tant Professor of Chemical and Biological
Engineering. He got his Ph.D. from the
University of Wisconsin at Madison. His
interest is in the area of protein en-
gineering and particularly those that
aggregate and cause Parkinson's and
Alzheimer's diseases.
Finally, we are very proud to an-
nounce that on Jan. 1, 2006, our
department was officially renamed
The Othmer-Jacobs Department of
Chemical and Biological Engineering,
in recognition of enormous contribu-
tions to our discipline made by these two
chemical engineering giants. Visit our
Web site: .


on campus
Library on
far left, and
the Othmer

Rasti Levicky

For practical reasons, this article
mentions only a few of the people that Leonard Stiel
are part of the history of chemical
engineering at the Polytechnic.
We would like to acknowledge
those dedicated present and
former professors, students and
alumni, and their supporters,
without whom Polytech would
never have attained its success- ' ' ,
ful international reputation.
We thank our colleagues in the Walter Zurawsky
department for their help writing this article, and greatly
appreciate the efforts of Christopher Hayes. 7

rj[ =educator
---- U s_____________________________________

Joseph Reynolds

of Manhattan College

Manhattan College * Riverdale, NY 10471

Joseph Reynolds earned a bachelor's degree in chemis-
try from Catholic University of America in 1957 and a
Ph.D. degree in chemical engineering from Rensselaer

Polytechnic Institute in 1964. He taught high school chemistry
and physics full time at LaSalle Academy in New York City
from 1957 to 1959, then taught college chemistry part time
for Catholic University (Troy extension) while pursuing his
doctoral degree at RPI. Joe excelled as a student and was
� Copyright ChE Division of ASEE 2007
Chemical Engineering Education

Joe, at right, posing with
Lou Theodore, his long-
time friend, collaborator
and fellow faculty member.

inducted into the Phi Beta Kappa, Tau Beta Pi, Sigma Xi, and
Phi Lambda Upsilon honor societies. His many accolades in-
clude listings in American Men and Women in Science, Who's
Who in Technology Today, Who's Who Among America's
Teachers, International Who's Who in Engineering, Who's
Who in the East, and Who's Who in Engineering.
Since 1964, Joe has been a member of the chemical en-
gineering faculty at Manhattan College, where he holds the
rank of professor of chemical engineering. It caused some
excitement among Joe's colleagues when Br. Thomas Scanlan
was appointed president of Manhattan College, because Br.
Thomas had been one of Joe's students in a freshman chemis-
try course that he taught in Troy. (Fortunately, we understand
that Br. Thomas earned an "A" in the course.)
Joe served as chairperson of the Department of Chemical
Engineering for seven years (1976 to 1983), and also was
called upon to serve as acting chair for brief stints totaling
another two and a half years while his successors were on
sabbatical leave. As part of his academic duties, Joe has
served for many years as moderator of the student chapter
of the American Institute of Chemical Engineers (AIChE),
and has been president of the college's Sigma Xi Chapter. He
has also served on a large number of college committees, but
says his favorite is the Board of Trustees' Facilities Planning
Committee because this membership ensures his invitation
to the President's Christmas Dinner (best wine selection and
food service, by far).
Since completing his doctoral research at RPI on "The Effect
of High Pressure on the Infrared Spectra of Solids," Joe has
collaborated for more than 30 years with Dr. Louis Theodore
at Manhattan College on various environmental research proj-
ects. Many of Joe's books and research publications include
Winter 2007

undergraduate students as coauthors. His current research
interests are in the air pollution control and hazardous waste
incineration areas. He has coauthored numerous text/reference
books, including Introduction to Hazardous Waste Incineration,
2nd Edition (2000), Accident and Emergency Management
(1989), and Handbook of Chemical and Environmental Engi-
neering Calculations (2002), all from Wiley-Interscience, New
York. He has developed computer software, which is available
commercially and currently used in the EPA's training program,
to simulate hazardous waste incinerator (HWI) performance.
His publications include problem and solution workbooks that
he uses in the courses that he teaches at the college, as well
as EPA training manuals for the HWI software. Joe has also
served as a consultant for several private companies and is
presently a consultant/expert witness for the Department of
Justice and the U.S. Environmental Protection Agency. He
has been active for most of his career in the Air and Waste
Management Association (AWMA), formerly the Air Pollu-
tion Control Association (APCA), where he presents papers
and chairs sessions at annual meetings as well as coordinating
associated continuing-education programs.

Manhattan College offers both B.S. and M.S. degrees in
chemical engineering, and Joe has always taught the under-
graduate courses by choice. He has taught the Engineering
Materials course and directed its associated laboratory for
his entire career at the college, and currently teaches Process
Calculations, Engineering Thermodynamics, Fluid Mechan-
ics, and Computer Aided Simulation and Design in Chemical
Engineering. During his tenure, he has taught nearly every
course that the department offers (or previously offered)
including Chemical Engineering Thermodynamics, Heat

Proudly posing
with students
at a poster

Transfer, Chemical Engineering
Laboratory I-II, Physical Metal-
lurgy, Physical Chemistry I-II,
Computer Methods in Chemical
Engineering, Computer Science
and Programming, and Fortran
Programming for Chemists.
In the classroom, Joe is very
much an "in your face" kind of
teacher. He teaches several of
the required sophomore courses
for chemical engineering ma-
jors, giving out grades from "A"
to "F," as deserved. The good
students stay, the others repeat
or change majors. The current seniors have created a bul-
letin board with pictures and the facts as they see it for the
chemical engineering faculty. Their advice for students of
Dr. Reynolds' classes includes: "Participate as much as pos-
sible. This will lessen your chance of being randomly called
on during class."
Joe's courses are well organized and fast paced. He sets high
standards, gives fair-but-tough tests, and assigns homework
due at every class. In the old days, he distributed the home-

work assignments for the entire semester
on day one, but when the Internet was
relatively new, he forced the students to use
it by sending out assignments via e-mail
only. The seniors advise, "check your e-
mail every day, at least twice a day. There
will always be something new in there."
This practice has the added advantage of
getting the students to read messages about
AIChE meetings and parties, which gets
them involved in departmental activities
from freshman year on.
Joe is one of the teachers who makes
effective use of the computer projector

and PowerPoint slides for each of his lecture courses: He
expects students to listen and respond during his presentation
instead of just madly copying information. His PowerPoint
presentations are available for all of his students through
the course Web sites on the Blackboard system. Many of
Joe's current and former students credit him as being a truly
outstanding teacher, a fact that is supported by his numer-
ous teaching awards and consistently excellent course and
teacher evaluations.

Joe's focus on excellent teaching must have set a good
example, because several of his former students also pursued
careers in academia. Among them are Dr. Ruben Carbonell
(B.E. 1969), now KoSa Professor of Chemical Engineering at
North Carolina State University; Dr. Sonia Kreidenweis (B.E.
1983), now professor of atmospheric science at Colorado State
University; Dr. John Blaho (B.E. 1983), now associate profes-
sor of microbiology at Mount Sinai School of Medicine; and
Dr. Marco Castaldi (B.S.ChE 1992), now assistant professor
of earth and environmental engineering at
Columbia University.
s much as
His former students turned academicians
will lessen
credit Joe in various ways for encouraging
being their graduate education and influencing
ed on their decision to pursue research and teach-
ing at the college level as a profession. Br.
Thomas participated as an undergraduate
in Joe's research at RPI, and credits this
n seniors to
early experience with giving him "an un-
ling to take derstanding of the way that research and
Oldss' class, scholarly activities reinforce teaching and
vice versa." Sonia Kreidenweis co-authored
her first publication with Drs. Reynolds
and Theodore in the Journal of the Air Pollution Control As-
sociation, based on her undergraduate research at Manhattan
College. She credits Joe as being "the first to suggest [that]
I apply to graduate school and go on for a Ph.D."-a degree
she subsequently completed at California Institute of Tech-
nology. Ruben Carbonell says that as an undergraduate, he
"looked up to Dr. Reynolds as a role model of an excellent
professor," which greatly influenced his decision to pursue a
career in college teaching.
Chemical Engineering Education

"Participate as
possible. This
your chance of
randomly calle
during class."

-Advice front
students plan

Top left, Joe and wife Barbara.
Top right, Barbara and daughter
Megan on a family trip to
Toledo, Spain-where both
Reynolds daughters got to prac-
tice their fluent Spanish. Right,
Joe and daughter Marybeth
undertaking a favorite family
activity-skiing-at Steamboat
Springs, Colorado.

All is not academics for this
overachiever, however. Joe's favor-
ite recreational activities include
skiing and jogging. He can be seen .
early mornings jogging around
his Bronx neighborhood near the
college. His equally active fam- 4
ily-wife Barbara and daughters
Megan and Marybeth-has accompanied him on the annual
Manhattan College ski trip every January since the girls were
infants. The foursome has also made the AWMA (APCA)
meetings in June an annual event. One of Barbara's favorite
activities is international travel, and the family has made
so many trips to Ireland that Megan and Marybeth recently
obtained dual citizenship.
Joe is as proud of his family's achievements as of his own.
Barbara has retired from Fordham Preparatory School in
the Bronx after 35 years of teaching. Megan and Marybeth
both earned baccalaureate degrees in chemical engineering
with honors from Manhattan College, so Joe is one of our
most enthusiastic alumnae parents. Megan recently received
a master's degree from Thunderbird, the Garvin School of
International Management in Phoenix. After working in Spain
for the pharmaceutical industry, she is currently working for
Merck in New Jersey. Marybeth completed her master's in
Winter 2007

Public Policy at Georgetown University and currently works
for Cancer Care in New York City. Both daughters are fluent
in Spanish and have studied other languages as well, i.e.,
Russian for Marybeth and Portuguese for Megan.

Joe is well known for his quick smile and easygoing manner,
as well as for his endearingly annoying habit of correcting
everybody's grammar-often in midsentence. The seniors
say, "Use proper English. He will call you on it every day!"
This applies equally to his faculty colleagues.
Joe's story is unusual in that he is an outstanding teacher and
a respected researcher at a primarily undergraduate institution.
When he was honored with a Bonus et Fidelis Medal on his 25th
anniversary at Manhattan College, he was interviewed about his
experiences. Asked about the most rewarding part of his career,
his response was immediate: working with students. 7

ri]j1 classroom
---- U s_____________________________________



A Hands-On Experiment for Freshmen

Kansas State University * Manhattan, KS 66506-5102

One of the endemic problems specifically in chemical
engineering, as well as in the field of engineering in
general, is the low retention rate of undergraduate
students. This attrition is especially noticeable in the first two
years of undergraduate studies, as roughly 50% of freshmen
entering chemical engineering do not make it to their senior
year.11 While students have varying reasons for transferring
out of science and engineering fields, one of the most com-
mon is a loss of interest in science and engineering.[2] In most
chemical engineering departments, students do not take a core
chemical engineering course until their sophomore year, and
don't become immersed in chemical engineering until their
junior year. This means that underclassmen who switch ma-
jors due to a loss of interest in science and engineering do so
without a good understanding of chemical engineering.
To combat the retention problem, many chemical engineer-
ing departments require an introductory course in chemical
engineering during the first semester of the freshman year.

Typically these courses serve to introduce students to the
department and its procedures, and give a broad overview
of some applications in chemical engineering. From a brief
survey of course descriptions and syllabi found on the In-
ternet, it appears that many of these courses use field trips
to chemical plants and presentations by guest speakers to
give students more of a perspective on the discipline. While
these are excellent activities to which students in chemical
engineering can be exposed, one problem is that they are,

Keith Hohn is an associate professor of chemi-
cal engineering at Kansas State University.
He received his bachelor's degree from the
University of Kansas and his Ph.D. from
the University of Minnesota, both in chemi-
cal engineering. His research interests are
heterogeneous catalysis and its application
in hydrocarbon conversion and hydrogen

� Copyright ChE Division of ASEE 2007

Chemical Engineering Education

for the most part, passive activities. Students are generally
hearing someone tell them what chemical engineering is or
are seeing pieces of process or laboratory equipment. They
are not touching, designing, or building anything. Hands-on
activities are relatively rare, though some departments have
used them successfully. [37
There are numerous reasons why hands-on projects are not
incorporated into freshman chemical engineering courses
more often. First of all, freshmen do not generally have the
background to apply many chemical engineering principles.
Secondly, it is difficult to package a true chemical engineer-
ing application into something that freshman students can
manipulate since chemical engineering frequently deals with
very large and sometimes hazardous processes. Finally, many
interesting activities would require extensive laboratory and
calculational time (on the order of the laboratory experiments
taught in chemical engineering lab courses). The requirement
for a useful hands-on activity that could be incorporated
into a freshman course is one that is interesting, safe, easily
understood by students with limited chemical engineering
knowledge, fairly simple, and capable of being completed
in a reasonable amount of time. This paper details such an
experiment that in fall 2003 and fall 2004 was incorporated
into a freshman chemical engineering course at Kansas State
University (CHE 110, Current Topics in Chemical Engineer-
ing). This experiment has students study the often-encoun-
tered phenomenon of carbonated soft drinks that have lost
their fizz (here in Kansas, we call that flat pop). Students
design and carry out experiments to study one aspect of this
phenomenon. The efficacy of this exercise in teaching stu-
dents what chemical engineering is and in increasing student
enthusiasm for studying chemical engineering was measured
by a semester-end survey.

Freshman students are generally familiar with the phe-
nomenon of carbonated beverages going flat, and have some
intuitive understanding as to why it occurs. Most will know
that the loss of carbonation leads to a flat beverage, and some
will recognize that carbonation is simply the absorption of
CO2 into the liquid phase. What students will not be familiar
with are the chemical engineering principles behind how
pop goes flat and how chemical engineers use many of these
principles to design chemical processes.

There are numerous chemical engineering principles
involved in the loss of carbonation. This is truly a rich mass-
transfer problem. Loss of carbonation depends on two factors:
the gas-liquid equilibrium for CO2 and the rate at which mass
transfer of CO2 from the liquid to the gas phase occurs. The
gas-liquid equilibrium is represented by Henry's Law:[81

H= Pco2 (g)/ Cc (1) (1)

where Po2 (g) is the CO2 partial pressure in the gas phase,
Cco2 (1) is the concentration of CO2 in the liquid phase, and
H is the Henry's Law constant.
Given enough time, CO2 will leave the liquid solution and
enter the gas phase until the above equilibrium relationship is
fulfilled. Temperature plays an important role, as the Henry's
Law constant decreases with increasing temperature. For car-
bonated beverage bottles left closed for long periods of time,
equilibrium is the most important factor in how the carbonated
beverage goes flat. The volume of the head space is clearly
important here, as the partial pressure in the entire volume
must satisfy the equilibrium relationship. Large head space
volumes lead to a large loss of CO2 from the liquid.
Mass transfer kinetics can be important in such situations.
There really are two types of mass transfer occurring in this
system: mass transfer of CO2 from the liquid to the gas and
mass transfer of CO2 through the bottle to the outside atmo-
sphere. For the standard polymer used to construct carbonated
beverage bottles, polyethylene terephthalate, the rate of mass
transfer of CO2 through the bottle is small. This would not be
the case, for instance, if low-density polyethylene was used to
make the bottle. The rate of mass transfer from the liquid to
the gas becomes important in loss of carbonation if the bottle
is opened and closed often within a short period of time. In
this case, there is not enough time to reach equilibrium, so the
CO2 lost from the liquid phase is the amount that went into
the gas phase in the time between openings. Mass transfer of
CO2 into the gas phase can be represented by:'81
Nc = KG (P -P *) (2)


Pco * = HCc02 (3)
Nco2 is the flux of Co2', PC02 is the partial pressure of CO2
in the gas phase, KG is the gas-phase mass transfer coefficient
times the mass transfer area, PC2 * is the partial pressure of
CO2 at the gas-liquid interface, and Cc02 is the concentration
of CO2 in the bulk liquid.

This activity was incorporated in CHE 110 (Current Top-
ics in Chemical Engineering) for fall 2003 and 2004. This
is a one-hour introductory chemical engineering course that
freshmen and transfer students are required to take for a letter
grade. Four of the 16 contact hours were spent on the CO2
absorption activity. The remaining time was dedicated to
lectures on curriculum requirements, advising and enrollment,
how to seek internships and full-time positions, applications
of chemical engineering, and field trips to a dairy processing
facility and the chemical engineering laboratories.
Students were presented with the topic of carbonated
beverages going flat by having a number of students take
the "Pepsi challenge," in which they sampled two different

Winter 2007

beverages and determined which tasted better. To show why
carbonation is important, one of the beverages was flat while
the other was fresh. Brief discussion of what made the fresh
beverage better ensued. This was followed by a discussion of
why carbonated beverages go flat, which introduced the idea
of CO2 absorption and set up a discussion of mass transfer
and gas/liquid equilibrium.
Students were then shown two ways to quantify the mass
transfer of CO2 from carbonated beverages. The first method
was based on an article by Crossno.[91 Briefly, a balloon filled
with 50 ml of 1M NaOH was affixed to a flask containing
150 ml of a carbonated beverage. The beverage was stirred
and left for ~24 hours to drive the CO2 out of solution. CO2
was adsorbed into the sodium hydroxide solution to form so-
dium carbonate. Titration of that solution to the first colorless
phenolphthalein endpoint neutralized the excess sodium hy-
droxide and converted all of the sodium carbonate to sodium
bicarbonate. Continuation of the titration to the methyl orange
endpoint converted the sodium bicarbonate to water and CO2.
The amount of HC1 required to go from the phenolphthalein
endpoint to the methyl orange endpoint gave the amount of
CO2 in the carbonated beverage.
The second method was to replace the original bottle cap
with a cap in which a pressure gauge had been placed. This
cap allowed the pressure in the head space to be measured
as a function of time.
During demonstration of the two methods, laboratory
safety procedures were highlighted and a handout was given
on these procedures. Following the demonstrations, the
students were told to form groups (self-selected) of four or
five students. Each group was told that they were to identify
and select one research topic related to the mass transfer of
CO2 in carbonated beverages. Several topics were suggested
to them, although they were encouraged to brainstorm their
own project ideas. They were then instructed to identify what
experiments and measurements they needed to do in order to
address the research question. They were finally told that they
would be required to report their results in both a written and
an oral report. Final written reports were turned in the last day
of class. Oral reports were given during class time in front of
the whole class in the last two or three weeks of the class.
Performance on the project was a major factor in the
students' final grade. In the first year of implementation, an
overall letter grade was assigned for the reports, which was
given roughly equal weight with attendance. In the second
year of implementation, the project was assigned 200 points
out of a possible 500 points, with the remainder of the points
for attendance. Students were required to turn in several
reports during the semester to ensure that they were making
progress on the project. The reports and their point value are as
follows: firing memo (described in the following paragraph),
10 points; description of experimental objectives, 10 points;

detailed experimental plan, 20 points; preliminary results
report, 20 points; rough draft of final report, 10 points; final
written report, 100 points; oral report, 30 points.
The students were given little information on working in
teams the first year, and this led to a few problems (described
in the results section). To address this problem in 2004, each
team was asked to meet and discuss the team's expectations
for individual team members. They were also asked to lay
out what specific actions would be taken if students did not
meet those expectations, leading up to a possible ultimate
action of "firing" the individual. They were then required to
write a document (a "firing memo") detailing this discussion
and all team members had to sign it. In addition, students
were required to rate their peers in a number of areas, such
as attendance at team meetings, contribution to reports, and
attitude, and turn in their ratings with the final report. Students
consistently rated low by their peers received a deduction of
their project grade, with the severity of the deduction deter-
mined by how low their ratings were.

Because students were allowed to choose their own research
topics, topic selection varied. Topics included:
C Does the commercially available Fizzkeeper work?
C How does temperature affect CO2 absorption?
) Estimate the mass transfer coefficientfor CO, loss
from carbonated beverages.
C Estimate Henry's Law constantfor CO, in carbonated
C Determine effect of different container materials
I .. i , it ,.. glass, and PET) on carbonated bever-
ages going flat.
C Determine how different PET beverage containers
affected the loss of CO, over time.
C How does the . ,,, i, of time the cap is left off a bottle
affect the rate at which the carbonated beverage goes

The experiments the students conducted and how they ana-
lyzed their data varied for the different projects. Most groups
addressed their research question empirically. For example,
several groups plotted CO2 concentration and/or gas pressure
vs. time for different conditions (i.e., different temperatures,
with and without a fizzkeeper). These groups did not use the
mass transfer equations described above.
Other groups relied on the mass transfer equations to ad-
dress quantitative questions, such as estimating the Henry's
Law constant or the mass transfer coefficient. The group that
estimated the Henry's Law constant measured concentration
of CO2 in the liquid phase and pressure in the gas phase for
several different samples, and attempted to fit these data with

Chemical Engineering Education

a single value of the Henry's Law constant. The group that
attempted to estimate a mass transfer coefficient measured the
gas pressure over time after the bottle had been opened and
closed (to start with atmospheric pressure). From the known
volume in the head space and the measured pressures, they
could calculate the change in moles of CO2 in the gas phase.
Next, the students solved Eq. (2) by separation of variables,
assuming that the concentration of the liquid (and therefore
Pco2 *) was a constant over time at the value they measured
after the mass transfer experiment. They then plotted their
experimental data using the resulting equation, and found
KQ from this plot. Essentially, they plotted the logarithm of
the partial pressure vs. time, which yielded a linear plot, the
slope of which was KG. This analysis assumed that all of the
mass transfer resistance was in the gas phase, which likely was
not the case. Making this assumption helped in the analysis,
however, since the students could readily measure the gas-
phase pressure over time.
The titration procedure was problematic for some students.
Sometimes students found that the balloon containing NaOH,
in which CO2 was absorbed, had been sucked into the flask
when they returned to the laboratory for titration. Sub-at-
mospheric pressures had apparently been created inside the
balloon due to loss of CO2 from the gas phase, causing the
balloon to shrink and eventually completely collapse. Stu-
dents also reported some problems with getting reproducible
results with the titration. These problems were likely due to
human error in most cases. There were fewer complaints in
the second year, possibly because a longer period of time
was given for completion of the project (nearly the entire
semester, as opposed to only six weeks) which allowed for
more repeat trials.
Student work showed promise, but analysis was often too
simplistic or relied on too few data points to draw a conclu-
sion. This provided a good opportunity, however, to present
important concepts such as estimating error and the need for
a good experimental design with replication. In the second

Assessment Results for CO2 Absorption Activity
Aspect Assessed Fall 2003
Average response to: 'This session improved 4.07 (out of 5) 7
my understanding of what chemical engineering
is and what chemical engineers do."
Average response to: 'This session increased 3.64 (out of 5) 7
my enthusiasm for studying chemical engineer-
Number of students listing the CO, activity in N/A
response to the following: "Of all the activities
we did in class, which three did you find the
most useful?"
Number of students listing the CO, activity in N/A
response to the following: "Of all the activities
we did in class, which three did you enjoy the
Winter 2007

year, students were asked to lay out a detailed experimental
plan for the data they would take to address their research
question, and were given feedback on the appropriateness
of their plan. In addition, preliminary reports provided more
opportunity to give feedback on whether they were analyzing
their data properly. Their oral presentations showed a good
deal of sophistication, with all groups using PowerPoint pre-
sentations with imbedded graphics. It is obvious that they had
previously given PowerPoint presentations in high school, as
no time was spent teaching about the tool.

Survey Results
A detailed survey was given to the students at the end of the
semester to evaluate both the course in general and individual
class activities. The results of this survey were used to assess
the effectiveness of the hands-on CO2 absorption experiment
in educating freshmen about chemical engineering and in-
creasing their enthusiasm for studying chemical engineering.
Table 1 summarizes student responses.
As seen in this table, students generally felt that the CO2
absorption activity improved their understanding of chemi-
cal engineering and increased their enthusiasm for studying
chemical engineering. In addition, the CO2 absorption activ-
ity was mentioned by 15 students (out of 36 students who
responded) as one of the three most useful activities in the
course (along with a field trip to a dairy processing facility and
a lecture on biotechnology), and by 17 students as one of the
three most enjoyable activities (along with the field trip to the
dairy processing facility and a tour of the chemical engineer-
ing laboratories). It is interesting, but perhaps not surprising,
that the most enjoyable activities had the students going out
to see applications of chemical engineering or engaging in a
hands-on activity rather than listening to a lecture.

Most students seemed to enjoy the exercise. The opportunity
to work with a "real world" engineering
problem energized a number of the students.
The students trying to evaluate the efficacy
Fall 2004 of the Fizzkeeper, for example, devoted
.23 (out of 10) a great deal of time (as well as a large
amount of sealant products) to attempting
to produce a bottle that would allow them
.05 (out of 10) to use the Fizzkeeper while simultaneously
measuring the pressure in the head space
of the bottle. It appeared that the students
15 (out of 36 with a more applied, rather than theoretical,
mindset appreciated the activity.
Student comments on the end-of-semes-
17 (out of 36 ter surveys were mostly positive, and also
respondents) provide some insight into why students
enjoyed the activity. Comments reflected

the following positives about the activity:
1. Provided an opportunity for a hands-on/laboratory

2. Allowed students to work in a group

3. Gave an idea as to what chemical engineers do

The opportunity for students to work in groups was particu-
larly well received. This was a great way for freshmen to get
to know their colleagues, make friends, and form study groups
for introductory science and engineering courses. Students
were forced to work in groups to decide what experiments
to run, to conduct those experiments, and to write the final
report on the project, leading to closer interactions than what
usually occur in a lecture course.
A few negative comments were noted. Comments in 2003
indicated that group dynamics were an issue. Some students
felt as if they had done all the work while other students had
done very little. To address these concerns, the next year more
time was spent discussing group work, and peer review of
group members was implemented. Another negative com-
ment, noted in both years, was that the project goals were not
well defined. This may, in part, be caused by the open-ended
nature of how the project was implemented. Student groups
were allowed to select their own projects with little input
from the instructor. Perhaps more input is needed when the
groups are selecting projects to ensure that the topic chosen
will yield good results and that the groups properly define
their objectives.

CO2 absorption in carbonated beverages can be used as a
hands-on activity in an introductory chemical engineering
course to educate students on chemical engineering. This

activity allows students to investigate a relatively familiar
phenomenon, a carbonated beverage going flat, using engi-
neering analysis. The CO2 absorption activity was successfully
implemented in a freshman introductory course at Kansas
State University. Students responded positively to its impact
on their understanding of and enthusiasm for studying chemi-
cal engineering. Most students also listed this activity as one
of the most fun and useful activities in the course. Student
comments indicated that they valued the hands-on nature of
the activity and enjoyed working in groups on a significant
"real world" engineering project.

1. Unpublished data, based on comparison of enrollment of freshman
chemical engineering students in the fall with enrollment of junior
students in a class taught at the junior level two years later.
2. Seymour, E., "Revisiting the 'Problem Iceberg'- Science, Mathemat-
ics, and Engineering Students Still Chilled Out," Journal of College
Science Teaching 24, 392 (1995)
3. Hesketh, R.E, K. Jahan, T.R. Chandrupatla, R.A. Dusseau, C.S. Slater,
andJ.L. Schm alzel, | ii..i. . ll..i . I,,: . 1I..... ..i 1 1 .1 .. . . .. i..11.c
Freshman Clinic at Rowan University, "Proc. 1997Ann. Conf. ASEE,
Seattle (1997)
4. Marchese, A.J., R.P Hesketh, K. Jahan, T.R. Chandrupatla, R.A. Dus-
seau, C.S. Slater, and J.L. Schmalzel, "Design in the Rowan University
Freshman Clinic," Proc. 1997Ann. Conf. ofASEE, Seattle (1997)
5. Ramachandran, R.P, J.L. Schmalzel, and S. Mandayam, "Engineering
Principles of an Electric Toothbrush," Proc. 1999 Ann. Conf ASEE,
Charlotte (1999)
6. Farrell, S., R.P Hesketh, and M.J. Savelski, "A Respiration Experiment
To Introduce ChE Principles," Chem. Eng. Ed., 38(3), 182 (2004)
7. Moor, S.S., E.P Saliklis, S.R. Hummel, and Y.-C. Yu, "A Press RO
System. An Interdisciplinary Reverse Osmosis Project for First-Year
Engineering Students," Chem. Eng. Ed., 37(1), 38 (2003)
8. Henley, E.J., and J.D. Seader, Equilibrium-Stage Separation Operations
in Chemical Engineering, John Wiley & Sons, New York (1981)
9. Crossno, S.K., L.H. Kalbus, and G.E. Kalbus, "Determination of
Carbon Dioxide by Titration, "J. Chem. Ed., 73, 175 (1996) 7

Chemical Engineering Education

Mj =1 laboratory



Lehigh University * Bethlehem, PA 18015

As I enter my 40th year of teaching, it seems appropri-
ate to remind teachers and students of a fundamental
error that occurs with surprising frequency. This error
is particularly evident in courses that cover a wide variety of
chemical engineering topics and pull together subjects sup-
posedly learned in previous courses. The senior design course
and a chemical engineering laboratory with a variety of experi-
ments fit this type of course. In teaching these courses I have
frequently encountered quite bright students who misuse the
deltas. Since the differences among the various deltas should
be obvious and not at all confusing, it is remarkable that errors
of this type crop up so frequently. But they do.
This paper will describe a particularly useful experiment
in the undergraduate Lehigh University chemical processing
laboratory that uses all three of the deltas and, therefore, helps
to cement in the minds of students the fundamental differences
among the three kinds.
Winter 2007

The title of this paper originates from the old expression
"The devil is in the details." (Some of you may also remember
Fip Wilson's famous portrayal of Miss Geraldene with her
expression, "The devil made me do it.")

William L. Luyben earned degrees in
chemical engineering from Penn State
(B.S., 1955) and Delaware (Ph.D., 1963).
His industrial experience includes four
years with Exxon, four years with DuPont,
and four decades of consulting with
chemical and petroleum companies. He
has taught at Lehigh University since 1967
and has participated in the development of
several innovative undergraduate courses.
He has authored 10 textbooks.

� Copyright ChE Division of ASEE 2007

A brief review might be useful to clarify the issues and
applications addressed in this discussion.
"In Minus Out" Delta
An "open" flow process has streams entering and streams
leaving. Mass, component, and energy balances can be applied
under either steady-state or dynamic conditions. For example,
a steady-state energy balance for a distillation column with a
single feed and two products is
The delta in this equation is
AH = Bh + DhD Fh
where the streams leaving the column are the distillate (with
flow rate D and specific enthalpy hD) and the bottoms (with
flow rate B and specific enthalpy hB), and the stream entering
the column is the feed (with flow rate F, and specific enthalpy
hF). Of course, appropriate and consistent units must be used
for flow rates and specific enthalpies. If the flow rates are in
moles per time, the specific enthalpies must be in energy units
per mole (e.g., Joule, kcal, Btu).
In a heat exchanger, streams are heated or cooled. Under
steady-state operations with no phase change, a stream enters
at temperature Tin and leaves at temperature Toot. If the mass
heat capacity cp is constant and the mass flow rate is FM, the
AH for the stream is
AH= FMc,(Tout - T,)
If there is a phase change, for example if steam is entering
as a vapor with specific enthalpy Hn and leaving as liquid
condensate with specific enthalpy hont through a steam trap,
the AH for the steam is
AH = Fseam (ho.t - Hn)
In fluid flow systems, the appropriate deltas are differences
in pressure, elevation, velocity, and density between the inlet
and the exit conditions.

"Driving Force" Delta
Transport processes occur because of differences in driv-
ing forces. In heat transfer, the difference is between hot and
cold temperatures. In mass transfer, the difference is between
large chemical potential and small chemical potential (partial
pressure, concentration, or activity). In momentum transfer,
the difference is between high pressure or velocity and low
pressure or velocity.
For example, consider a perfectly mixed vessel that is
surrounded by a jacket. The temperature of the liquid in the
vessel is Tv .ss Suppose the jacket is completely filled with
condensing steam at temperature Tteam. The driving force for
heat transfer is
AT = Tsteam - Tvss
The heat-transfer rate Q that results from the driving force
AT is
Q = UAHAT = UAH (Tteam - T,,,
where U is the overall heat-transfer coefficient and AH is the
heat-transfer area of the vessel wall. In this example, the jacket
temperature is the same at all positions in the jacket.
If the vessel is cooled or heated by a fluid flowing through
the jacket or through an internal or external coil in plug flow,
the temperature of this fluid changes with position. Therefore
the temperature driving force changes, and a log-mean tem-
perature difference must be used.
AT T - AT,


where the two deltas are the temperature differences at the
inlet and outlet ends of the jacket or coil.
AT1 = Td - T,
AT, = Tv, - Tcout
Now the heat-transfer rate is

Steam In (Tco,)Top

(T-oo,) tti In

Steam Trap

Condensate Out

Figure 1. Heated or cooled agitated vessel.



n, AT

If a circulating cooling water system is used with a high
rate of circulation, the temperature in the jacket is essen-
tially constant at some temperature, T,. The heat-transfer
rate is

Q = UAHAT= UAH (T,,, - T )

In this type of system, a cold makeup water stream at TCn
is added to the circulating loop, and water is removed at
the jacket temperature Tj. A circulating cooling water
system has superior dynamics compared to the once-
through system. The high circulation rate maintains a
high coolant-side film coefficient that does not change
Chemical Engineering Education

with the load on the system (the makeup water flow rate), so
time constants are less variable.
"Time" Delta
The variables in a dynamic process change with time, so we
can talk about changes in time, At, and changes in properties,
Ax, between their value at one point in time and their value at
a later point in time. For example, when the liquid in a vessel
is heated at startup, the temperature changes with time.
AT = T,, - Tt

At any point in time the rate of change of temperature is
dT AT T12 -1
dt At t2 - t1
if the time increment between t2 and t1 is small.

The process consists of a stirred vessel, 1 m in diameter
containing 785 kg of water. The rpm's of the agitator can be
varied to see the effect on the inside film coefficient. A spiral
coil is wrapped around the outside of the vessel, making nine
wraps around the circumference. Figure 1 gives a sketch of the
apparatus. The tank is equipped with a 0.3 m, 6-blade impeller
with four baffles. The heat-transfer area is 3.14 m2.
The liquid in the vessel is initially at ambient temperature. It
is heated by introducing steam at the top of the coil. Conden-
sate leaves at the bottom through a steam trap. Temperatures
inside the vessel and at the inlet and outlet of the coil are
monitored by a data acquisition system.
When the temperature of the vessel reaches about 90 �C, the
steam is shut off and cooling water is introduced. The water
enters at the bottom of the coil and leaves at the top.
Data and Analysis
Figure 2 shows typical temperature vs. time trajectories
for the batch heating and cooling. The temperature in the
coil during heating is shown as being constant at the steam
temperature. This is actually not the case because it takes
some time for the coil to become completely full of steam.
This complicates the analysis of the heating step because the
temperature profile along the coil is not known until it is full
of steam. We consider this later in this paper.
The analysis of the cooling step is much more straight-
forward, and our discussion for the purpose of illustrat-
ing the "devil deltas" will concentrate on this part of the
batch cycle.
The flow rate of cooling water is constant and can be mea-
sured by the old-fashioned "bucket and stop watch" method.
The inlet and outlet cooling water temperatures are measured,
as is the vessel temperature.
Winter 2007

At any point in time, there are two ways to estimate the
instantaneous heat-transfer rate. From the measurement of
the cooling water flow rate and inlet and outlet temperature,
the heat-transfer rate at that point in time is

QCW = FCWCp (TC.out - Tc,, )
This is the "out minus in" delta. At time equal 30 minutes in
Figure 2, this "out-minus-in" delta is
ATot m = Tc,,ou - Tc,, = 38 -15 = 23 C�
The heat-transfer rate can also be estimated by the time rate
of change in vessel temperature. This uses the "time" delta.
At time equal tn, the instantaneous rate of heat transfer to the
fluid in the vessel is

S=Tvessel )(t= )- Tvessel (t=t_)
Qvessel(t=t,) = M lCp t

Since the heat-transfer rate varies with time, the slope of
the temperature vs. time curve varies during the batch cooling
step. Having two independent ways to estimate the rate of heat
transfer improves reliability of the estimated film coefficients.
Figure 2 shows this delta at 30 minutes is about 3.5 �C per
minute (the slope of the vessel temperature line).
ATe,, = 3.5 �C /minute
Using this value, the instantaneous heat-transfer rate is
calculated to be 192 kW. The heat-transfer rate from the flow
rate of cooling water (2 kg/sec) and the inlet and outlet cooling
water temperatures is very close to this number.
The temperature of the cooling water in the coil varies along
its length, so a log-mean temperature difference is used. This
is the "driving force" delta,
In --
�42 T2

Figure 2. Temperature profiles and temperature deltas
during cooling.

Figure 3. Using time delta for heating phase with
constant U and Tte.
where the two deltas are the temperature differences at the
inlet and outlet ends of the coil.
lATi_ T,_, - T,
AT = Tv - Tcout
The log-mean temperature difference at this point in time is

LMa , AT1


(70- 38)- (70-15) 42.5

n 70-38

With a heat-transfer area of 3.14 m2, the overall heat-transfer
coefficient is

U= 192 1.44kWm 2K
AHATL (3.14)(42.5)
The vessel inside film coefficient can then be calculated by
accounting for heat conduction through the vessel wall and
estimating the film coefficient inside the coil using the Dittus-
Boelter equation and an appropriate equivalent diameter.
Even with a constant agitator speed, there is some varia-
tion of the overall heat-transfer coefficient and the inside film
coefficient with time. This occurs because the varying vessel
temperature affects the viscosity of the water.
Approximate Analysis for Heating Step
If the temperature in the coil was constant during the heat-
ing period and the overall heat-transfer coefficient was also
constant with time, the analysis would be quite simple. It
would use "time" deltas in a way that may not be obvious. The
situation is analogous to the steady-state flow of fluid down the
length of a heated tube. In that situation, the appropriate driv-
ing force for calculating the heat-transfer rate is a log-mean
temperature difference using the temperature differentials at
the inlet and exit ends of the tube. The log-mean temperature
difference assumes a constant heat-transfer coefficient. The
independent variable is length.

In the batch heating situation, the independent vari-
able is time, but the heat-transfer equations are the same.
Therefore, total heat transfer can be calculated by the
change in vessel temperature from some point in time to
another point in time. The driving force can be calculated
using a log-mean temperature difference based on the
difference between the constant steam temperature and
the temperature of the vessel at the two points in time.
The similarity between length and time coordinates is
understood if you visualize a little particle of fluid flow-
ing down the tube in steady-state flow. It sees a constant
steam temperature and is heated as it flows along. This is
exactly the same as the batch heating of a vessel.
It should be emphasized that the analysis discussed
in this section makes two important assumptions. First,
the steam temperature is constant. Second, the overall
heat-transfer coefficient is constant.
Figure 3 shows how the vessel temperature changes during
heating. It starts at T when time is t and ends at T when time
is t2. The total amount of energy added during this period is
Energy = M .Cp (T2 - T )

The average heat-transfer rate over this period is
= Energy Mss1 Cp (T2 - T)

t2 - t

t2 - t

The instantaneous energy balance on the fluid in the vessel
MvesselC Vssel = AHU Tstm - Tesse )
This linear ordinary differential equation can be integrated
to give

Tvessel(t) = Tsteam cle
The constant of integration c1 is evaluated at the initial condi-
tion where T , = T1.

c = (T - T ,a

The time dependent vessel temperature is

Tvessel(t) = Tsteam + 1 - Tsteam ) e
Evaluating this equation at time equal t2 where T vi = T2

T2 = T, T + ( T - T, )e'

Rearranging gives

T2 - steam
T - etam
Zl - steam


Chemical Engineering Education

Taking the natural log of both side of this equation gives

In team t =- AHU (t
eam - T1 .essel c
Rearranging and substituting the previously defined equation
for Q give

T - T
In steam 2
Ts,,, - 1T,

AHU(T - T)
Mveselcp (T2 T )
(t2 - t)

1n t T, - T, ^ AHU(T
steam - T, Q

Q =UAH ) UAH (team
In steam - T2
team = UA (T)


T2 ) - (TTseam - T)

In steam - T2
Steam - 1

In our experimental apparatus, the temperature through the
entire coil is not equal to the steam temperature for about half
the heating period. In addition, the change in viscosity due to
changes in temperature results in variations in the heat-transfer
coefficient. So, the simple analysis described above can only
be applied for the period toward the end of the heating step.
During the initial part of the heating step when the tempera-
ture of the exit stream from the coil is not equal to the inlet

temperature, the full heat-transfer area is not being used for
steam condensation. Thus it is uncertain how to calculate an
internal heat-transfer coefficient. One approximate method
is to assume that the active heat-transfer area varies linearly
with time during this period.
Once the temperature of the exit stream from the coil be-
comes equal to the inlet temperature, either the approximate
method discussed in this section or a rigorous approach can
be applied. The rigorous method evaluates the inside film
coefficient at each point in time by getting the heat-transfer
rate from the rate of change of the vessel temperature, and
using the differential temperature driving force of Tteam minus
T and the full heat-transfer area.
For example, in Figure 2 at time equal 20 minutes, the
differential temperature driving force is 100 - 85 �C and the
slope of the Tv si curve is about 1.3 �C per minute.

This paper has attempted to provide a clear distinction
among the three deltas that are used in chemical engineering.
Although they are obvious to the experienced engineer, they
are often misapplied by young students.

Discussions of this experiment with Kemal Tuzla are grate-
fully acknowledged. 7

Winter 2007

Mr[ laboratory
---- U s_____________________________________



for Interactive ChE Education

Washington University * St. Louis, MO 63130
practical experimentation that processes real signals is Jing Guo received his B.S. degree in chemical engineering in 1997 and his
M.S. degree in 2000, both from Beijing University of Chemical Technology.
essential to helping students understand theory given He received his Ph.D. degree in chemical engineering from Washington
in textbooks and giving them skills to deal with real University in St. Louis in 2005, where he worked on the experimentation of
catalysis in multiphase reactors, including trickle bed reactor and packed
problems successfully. An indispensable part of the chemical bubble column. He also developed a modeling program to simulate the
engineering curriculum, the experimental class is designed multiphase reactions for applications ranging from bench-scale to com-
to train all students at the same time and in an effective way merci-scale.
for acquiring face-to-face interaction. This conventional ap- David J. Kettler was awarded double bachelor degrees in biomedical
proach, however, imposes difficulties on students with time and chemical engineering from Washington University in St. Louis in 2001.
During his study, he was also responsible for developing the Process
or distance constraints. Moreover, due to both safety and Control Laboratory's homepage and the Simulink Virtual Laboratory as an
security reasons, access to labs cannot be totally free and is interactive series of workshops.
restricted in time to ensure the presence of supervision person- Muthanna H. AI-Dahhan is an associate professor of chemical engineer-
nel. Interesting proposals have been made to use the Internet ing at Washington University in St. Louis and associate director of the
r v e p in in difr ty o Chemical Reaction Engineering Laboratory. He received his B.S. degree
for various educational purposes, including different types of in 1979 from the University of Baghdad in Iraq, M.S. degree in 1988 from
virtual laboratory Web sites,['] interactive simulations,[2] and Oregon State University, and Ph.D. degree from Washington University
access to real instruments and test benches through a remote in St. Louis in 1993. His research interests are in the fields of chemical
reaction engineering related to multiphase reactor systems, mass transfer,
connection.[35] In fact, some implementations of remote moni- and process engineering. He is author of more than 80 papers in the field
touring and control through the Internet have already reached of chemical engineering.

� Copyright ChE Division ofASEE 2007

Chemical Engineering Education

the teaching laboratories of physics161 and electrical engineer-
ing.l1 For chemical engineering laboratories, this capability
is now available at University of Tennessee at Chattanooga,E81
University of Texas at Austin,[91 and MIT.1101
With appropriate planning, teachers and students can run
Web-connected experiments on a flexible schedule, which
provides educational facilities and opportunities for those
students whose schedules might be asynchronous.18 "1 An-
other advantage of such remotely accessible laboratories is
that teachers and students at another institution can have ac-
cess to laboratory facilities without incurring the full cost of
developing such resources. Rather than several universities
spending money on the same equipment for the same experi-
ments, cooperating universities may each carry out one unique
experiment and then form an experiment pool.12' 131 Using
such highly automated experiments for remote operations
can allow a drastic reduction in the amount of personnel time
required for those particular experiments. It is reported that
online laboratories hold promise of being up to two orders
of magnitude cheaper than conventional ones.1141 Having
access to tutorials, pictures, past data files, data processing
tools, and graphs tracking the dynamic process variables,
these Web pages provide students with sufficient resources
that can be viewed simultaneously by all class members.1101
Such expanded access allows the students and instructors to
spend less time communicating the operating procedure and
more time investigating the experimental results. Remote
learning has evolved into a new model of high quality aimed
at engaging students in a distinctive learning technology that
helps build a solid foundation.19, 5]
Advances in available computer software and interfacing
techniques enable remotely operated laboratory experiments
to be constructed at relatively low cost.1161 In this paper, we
report on the in-house development of remote control and
measurement methods for a chemical engineering labora-
tory on unit operations, which is offered to undergraduate
students at Washington University in St. Louis. A client-server

Remote Client HTTP Server
Client Applcations Network Network Instrument Management
(Measurement Management Management (Board Drivers,
& Control Client Side Server Side Command Process)
User Interface)

Figure 1. Diagram of the client-server architecture
employed to implement the remote control measure
Winter 2007

architecture devoted to instrument management through the
Internet is built with Visual Basic and LabTECH program-
ming tools, providing a novel approach in comparison to the
Java and LabView software employed in other references.18
101 The architecture is described along with the specification
and design of a geographically distributed system based on
standard commercial components. Used for the required
undergraduate process control course, a tracer experiment
is restructured to illustrate the connection between physi-
cal instruments and the server-client Internet system. The
experimental data is archived for subsequent viewing and
analysis, and the responses of students to the online experi-
ment are assessed.

To achieve a standard component distribution system,
we adopted Internet technologies that allow portability and
independence through different client hardware/software
architectures. A standard portable language is instrumental
for independence of the application from the client system
on which it is executed. An Internet browser can now be con-
sidered a standard component of any computer installation.
Therefore, our approach will automatically work with any
widely available hardware/software environment.
The connection between the server and client program
is made by a TCP/IP Winsock socket located within both
programs, which functions much like a phone receiver/dialer
on each side of the Internet. The server and the clients are
connected on the same local area network (LAN) within the
laboratory or campus. Remote connections can even be set up
between the server and a single user working at home.
TCP/IP defines the physical interconnection, data transfer,
and message routing management. It is the typical protocol
suite adopted in the standard Internet.11 The server sends
measurement data to the client the same way the client sends
control commands to the server, by creating a string of num-
bers representing all the commands or measurements and
sending them through the TCP/IP socket. Like-
Lab Setup wise, once the client receives the measurement
string of numbers and the server receives the
control string of numbers, the string is parsed,
and each measurement or command within
the string is sent to its appropriate subroutine
within the client or server code.
S A block diagram of the proposed solution is
shown in Figure 1. The clients are hosted on
a user's personal computer while the server
runs on a laboratory computer and manages
an automatic control and measurement system
that embeds programmable instruments. Both
client and server computers run programs that
are logically split into two layers. One layer
e in both client and server sides deals with user

H a
II Cntrol

I-- -Measurement




[An] advantage of such remotely accessible laboratories is that teachers and
students at another institution can have access to laboratory facilities
without incurring the full cost of developing such resources.

interface and instrument management, while the other layer
deals with network intercommunication. The server is directly
connected to the instruments that measure physical quanti-
ties. In this work, the server computer, connected physically
to the instruments, makes available a set of remotely callable
procedures that perform all standard activities (address, read,
write, status poll, etc.). The client's command generator is-
sues commands according to the parameter set specified by
the user, and sends them via the TCP/IP client socket to the
server. The experimental results sent back by the server are
then handled and displayed in the client window. The sockets
of the client program and the server program are connected
by using the server computer's IP address on the Washington
University network. The same local port number must be
specified within both the client and server sockets. Socket con-
nections and the TCP/IP communication protocols transmit
the instrument control commands, parameters, and reports
between client and server.
Only one user group is allowed to connect at one time,
because physically only one experimental run can be done
on the reactor at a time. At the session's termination, the
socket connection is closed and the server can accept a new
connection on the same port to start a new session. If the
client/server connection is broken or remains idle more than
five minutes, the server application shuts down the system.
If the power shuts down, a system of safety interlocks in the
physical system prevents the system from running indefinitely.
To protect the server, several techniques can be used, e.g.,
access restriction based on user identification, firewalls, and
encryption. For simplicity and cost reasons, we adopted an
approach based on access restriction through user verifica-
tion of both the password and the IP address of the gateway
through which client connects to the server.
On the server site, the structure offers great flexibility.
Developed in Visual Basic, the programs require only the
addition of a very small number of statements necessary
for establishing and closing the interface-related functions
of corresponding network functions. When new instru-
ments are added to the instrument library, it is easy to add
a measurement or control variable with small modifications
to existing programs. The software related to any newly
connected equipment can be added to the system without
recompiling or modifying the application core. About two
or three lines added to the server and client programs will
add numbers representing the additional measurement or
control variables to the string sent through the TCP/IP

At the client site, because the whole core of the software
application (i.e., the components required to share, engage,
and release the resources) resides permanently on the server
computer, it is not necessary to install any special software
tool. Once a new client connection is accepted, the user lo-
cally runs the command necessary for selecting and driving an
instrument. As a consequence, the proposed structure makes
the application portable and safe for remote users.
The server is a Pentium-IV computer with a 1 GHz pro-
cessor, provided with two independent Universal Serial Bus
(USB) ports. The server runs on Windows 2000 Professional
and uses drivers from Data Translation to access the Data
Translation DT9804 interface board. The server connects
to the interface board using a USB cable, and the interface
board has analog input/output and digital input/output ports
for connection to the physical control hardware of the reactor
system. The overall system has been devised to assure reliable
communication between the client and the server, and between
the server and the physical resources available. A LabTECH
ControlPro 12.1 Runtime program receives data and sends
commands to the control hardware via the DT9804 interface
board. Once the control hardware and server are physically
connected to the DT9804 board and the board's drivers are
installed, LabTECH can be set up to control the hardware
by dragging and dropping control icon blocks from its menu
into its workspace.

The lab setup icon shown in Figure 1 represents any real
instrument that requests automatic control and measure-
ment. The proposed server-client system structure can be
extended to fit a variety of requirements and serve different
experiments. As a test case performed on implementation of
the whole system, a tracer study experiment is carried out
remotely in real time over the Internet, using a tubular reac-
tor in the Chemical Engineering Laboratory at Washington
University. The purpose of the tracer test is to experimentally
determine the ideality of a real continuous flow reactor.

In an ideal plug flow reactor, the fluid flows through the
reactor with a pistonlike motion and no axial mixing. A real
tube reactor, however, cannot reach this ideal state. In the
experiment, a conductive tracer was injected into water just
before the reactor entrance and the conductivity of the solu-
tion mixture was measured at the reactor entrance and exit.
The mean residence time, tracer response curve variance,
dimensionless variance, and axial dispersion coefficient can
Chemical Engineering Education

then be calculated. Through data analysis, the tube reactor
was compared with an ideal plug reactor.
Figure 2 displays the physical hardware built for the tracer
study experiment. The students open the feed flow valve,
adjust the water flow rate by varying valve position, start
the tracer feed pump, adjust the tracer injection duration,
open the tracer injection valve, and then inject the tracer. Air
pressure (30-100 psi) is used as the driving force to control
the feed flow valve and the tracer injection valve. The feed
flow rate is measured by a turbine flow meter downstream,
and the measurement is sent to the client application. The
conductivity measurement will rise and then fall back to the

steady state value, at which point the students may close the
client application.
Table 1 lists all of the process variables used as signals in
the tracer experiment. Digital signals are either on or off when
equal to 1 or 0, respectively. Analog signals send (Output)
or receive (Input) signals within a defined range of values,
shown in the "Range" column in Table 1. The output signals
control the instrument setup, while the input signals are
variable measurements obtained at a specific condition. The
types of all variable signals are listed in the "Type" column.
The elements that launch and receive signals are listed in the
"Origin" and "Destination" column, respectively.

Water in Feed Flow Valve

Air in

..- Flow
------ -- Pneumatic line
------- Computer signal

KC1 Tank Pump

Figure 2. Overview of the physical experimental setup
and its connection to the Web.

The first application in the server computer is
the LabTECH Runtime program Traceexe.ltc.
It initiates the connection between the physical
laboratory setup and the server computer. This
connection channel receives measurement sig-
nals from the USB port on the interface card and
issues commands to control the setup operation.
All measurement variables are classified as analog
inputs in the Traceexe.ltc program. By specifying
the correct interface point, each analog input block
in Traceexe.ltc receives the proper signal from the
interface card.
The second application in the server computer,
ServerTracerStudy.exe, activates the server site
and enables it not only to transfer the remote cli-
ent signal to the physical setup, but also to receive
measurements from Traceexe.ltc by continuously
using the GetLT function. This function uses a

Process Variables Used in the Tracer Experiment
Variable Range Units Type Origin Destination
Reactor Feed 0-1 Volts Digital Output "Client_TracerStudy.exe" Control Hardware
Tracer Feed Pump 0-1 Volts Digital Output "Client_TracerStudy.exe" Control Hardware
Inject Tracer 0-1 Volts Digital Output "Client_TracerStudy.exe" Control Hardware
Tracer Injection 1-3 sec Analog Output "Client_TracerStudy.exe" "Server_TracerStudy.exe"
Reactor Feed 0-100 % Analog Output "Client_TracerStudy.exe" Control Hardware
Valve Position
Run Time >0 sec Analog Input "Traceexe.ltc" "Client_TracerStudy.exe"
(Reactor) Influent >0 mS Analog Input Control Hardware "Client_TracerStudy.exe"
(Reactor) Effluent >0 mS Analog Input Control Hardware "Client_TracerStudy.exe"
(Reactor) Feed >0 1/min Analog Input Control Hardware "Client_TracerStudy.exe"
Flow Rate
Winter 2007

IRun Time 128.900 Isec

IFeed Flow Rate 0.17608 Ilpm

Influent Conductivity 1591.491 ImS

IEffluent Conductivity 155.389 ImS

1' I
Figure 3. User interface for the client application.

built-in LabTECH application called LT-Speedway to "grab"
the analog input data received by Traceexe.ltc. The program
ServerTracerStudy.exe takes four measurement variables it
receives from its GetLT function and combines them into
one string of text, called OutputString. Once the client
connects to the server computer using the client program
Client_TracerStudy, then Server_TracerStudy.exe sends
OutputString across the Internet once every 100 milliseconds
to the client program, using a timer within the server program
called Timer2.
These two applications must be running before a student
can access the experiment using the client program, Cli-
entTracerStudy The student downloads this program from an
Internet page and stores it on the remote computer. Once the
student double clicks on the related icon, the client program
opens up and connects to ServerTracerStudy on the server.
Every command the user manipulates sends text data from the
client TCP/IP socket across the Internet to the server TCP/IP
socket. The server program sends measurement data acquired
from the LabTECH Runtime program Traceexe.ltc back to the
client through TCP/IP socket and sends the control variable
commands acquired from the client to Traceexe.ltc, where it
is executed by that program on the control hardware.
The controlled variables, reactor feed "Valve Position" and
tracer "Injection Duration," are analog outputs in Traceexe.
Itc. Before injecting the tracer, the student sets the values
of analog outputs by using a scroll bar on the user interface
of the client program, Client TracerStudy.exe, as shown in
Figure 3. The student must also open the reactor "Feed How
Valve," turn on the "Tracer Feed Pump," and "Inject Tracer"
by pushing the respective buttons on the client user interface.
These are digital outputs in Traceexe.ltc. Whenever the stu-
dent pushes one of the buttons (digital) or slides one of the
scroll bars (analog) on the client user interface, the current
values of all the digital outputs are combined with the analog
outputs as a text string called InputString in the InputData
function in the client program. Then InputString is sent to the

Feed Flow Valve

|Feed Valve Position 501%

1:: r:::::: ac Eea:EiulmB::::::

Infection Duration 21sec

IWater Flow Valve 0

Water Valve Position 50

Tracer Recirc. Pump [0

Iniect Tracer 0

Infection Duration 10

IWater Flow Rate 0.00549

Ilnfluent Conductivity 1592.376

IEffluent Conductivity |557.586

Client IP: INo one is connected

Check State I

Figure 4. Monitoring window for the server application.
Chemical Engineering Education


server program through the client TCP/IP socket. The server
program picks up the InputString of text across the Internet
at its TCP/IP socket, separates all of the outputs, and places
them in their respective textboxes, as shown on the lefthand
side of the image in Figure 4. In the server application, the
function PutLT takes each of the values in the textboxes on
the lefthand side of the server monitoring window and sends
them to their respective input block in Traceexe.ltc. The input
blocks receive each signal in Traceexe.ltc and send them to
their corresponding Bit Number (digital) or Interface Point
(analog) on the DT9804 interface card.

On the project Web site (),
eight groups of undergraduate students have participated in
the online operation test. After a class of introduction to the
distributed learning technology and two additional classes
on the theoretical aspects of the experiment, the instructor
demonstrated and monitored experiments using a classroom
computer connected to the Internet. When the students were
doing the measurements themselves in the computer lab one
floor above the laboratory, they were asked to provide their
inputs on the user-interface, analyze the experimental out-
comes, and answer questions posed by the instructor through
interactive dialog. Explanatory Web pages were provided to
answer most of their questions on the real instruments during
the lab session. As a result of this interactive tutoring mode,
students showed more interests in the online operation than
the local on-site operation.
During the lab session, students issued commands and pa-
rameters from the client window to the server via the TCP/IP
client socket. The experimental results were sent back by the
server and then handled and displayed in the client window.
The client program created a log containing measurements
of time, flow rates, influent conductivity, and effluent con-
ductivity. Once the client application is closed, students can
open this log to analyze the evolution of the collected tracer

iiiii TCP Seryer - WamsiiijiTrcejSudylim

0.04 Inlet
a 3
0.03 -
o 3 Outlet

0.01 :

0 10 20 30 40 50 60 70 80
Time (s)

Figure 5. Typical tracer response curves shown in
the client side. Measurements are taken at the inlet and
outlet of the tubular reactor.

response. Typical tracer response curves at the inlet and outlet
with respect to time are shown in Figure 5. In an ideal plug
flow reactor, the tracer curves collected at the exit and entrance
would be identical as thin, spikelike peaks. This experiment,
however, found that dispersion and stagnancy have significant
effects on the tracer response curves. The best flow model
was determined to be a plug flow reactor with a dead zone to
account for the stagnancy, followed by a mixed flow reactor
to account for the axial dispersion. The reactor's nonideality
must be included in order to predict reactant conversion from
given feed rates and reactant compositions.
The experiment was conducted twice, once at a remote
client computer station that was Internet-linked to the server,
and once at the server computer station directly attached to
the setup elements. The typical experimental values obtained
from the online remote control and on-site local control, as
well as the relative error between these two values, are listed
in Table 2. Although there is a time delay between the client
and server due to the Web data transfer and the instrumenta-
tion synchronization, this delay penalty is negligible when
instruments take a long time to complete the measurement.[41
Hence, the insignificant relative error leads us to conclude that
the online control and on-site control give rise to the same


residence time measurement in the tubular reactor.
Student feedback is a key consideration for improvement
of the experiment. Surveys were filled out by the students
after each lab session to evaluate the beneficial features of the
remote learning experience and the fulfillment of educational
objectives. The responses contained encouraging comments
and constructive suggestions. In general, the proposed survey
recommendations were implemented before the next student
group was invited to evaluate the lab session. Students agreed
that lab sessions became improved with more user-friendly
options and tools added to the client window. One feature the
students liked most about operating experiments remotely was
that it allowed them to perform the process at any time from
a place that was convenient for them. The other appreciated
feature was that the remote operation helped the students get
used to a real world application that was either in a remote
control room or at a remote operation facility, especially when
hazards and safety concerns were present. Some students
showed intention to run the on-site physical experiments as
the complimentary reference check since their understand-
ing of the experimental flow scheme can be enhanced with
the actual devices in front of them. Actually, this intention
could be fulfilled by incorporating the live audio and video
streaming to the remote client window so students can listen
to the sounds of the device station and view it on the Internet
while they are operating. Such sophisticated user interface
will soon be added to the current system.

This paper describes an Internet-based client-server archi-
tecture specifically designed to allow flexible management
of remote instruments. The proposed solution is portable
using the employment of the TCP/IP protocol suite, and
also extensible because of the high level of abstraction in
system implementation. This approach offers a valuable
component to remote engineering instruction that cannot be
replaced by simulation software packages. Compared to the
traditional way of teaching, due to the absence of schedule
and physical constraints, this new approach reaches students
who otherwise would not have chance to take these courses
and allows a larger and more diversified audience to access
learning opportunities. A set
S of experiments based on the

Experimental Measurements of Residence Times
Valve Opening Actual Flow Local Control Remote Control Relative
Position (%) Rate (cm/sec) (sec) (sec) Error (%)
45 2.93 22.96 23.77 -3.53
55 3.86 18.32 19.02 -3.82
65 5.11 14.57 14.97 -2.75
75 6.02 11.90 12.19 -2.44
85 6.44 11.74 11.38 3.07
95 6.67 10.87 10.67 1.84
Winter 2007

proposed technique for the
control of remote instru-
mentation has been made
available to the students
of chemical engineering
laboratory courses held in
Washington University in
St. Louis. There is the op-
portunity to use this technol-
ogy to add other experimen-

tal demonstrations or assignments to one lecture. In order to
expand the scope of the experiments and to share costs and
software development time, we are planning collaboration on
this project with other universities.


1. Ferrero, A., and V. Piuri, "A Simulation Tool for Virtual Laboratory
Experiments in a WWW Environment," IEEE Trans. Instrum. Meas.,
48,741 (1999)
2. Shin, D., E.S. Yoon, S.J. Park, and E.S. Lee, "AWeb-Based, Interactive
Virtual Laboratory System for Unit Operations and Process Systems
Engineering Education: Issues, Design, and Implementation, "Comput-
ers and Chem. Eng., 26, 319 (2002)
3. Benetazzo, L., M. Bertocco, E Ferraris, A. Ferrero, C. Offelli, M.
Parvis, and V. Piuri, "A Web-Based Distributed Virtual Educational
Laboratory, "IEEE Trans. Instrum. Meas., 49, 349 (2000)
4. Bertocco, M., F Ferraris, C. Offelli, and M. Parvis, "A Client-Server
Architecture for Distributed Measurement Systems," IEEE Trans.
Instrum. Meas., 47, 1143, (1998)
5. Arpaia, P, A. Baccigalupi, E Cennamo, and P Daponte, "A Measure-
ment Laboratory on Geographic Network for Remote Test Experi-
ments, "IEEE Trans. Instrum. Meas., 49, 992 (2000)
6. Enloe, C.L., WA. Pakula, G.A. Finney, and R.K. Haaland, "Teleopera-
tion in the Undergraduate Physics Laboratory--Teaching an Old Dog
New Tricks," IEEE Trans. Educ., 42, 174(1999)

7. Shen, H., Z. Xu, B. Dalager, V. Kristiansen, O. Strom, M.S. Shur, T.A.
Fjeldly, J.Q. Lu, and T. Ytterdal, "Conducting Laboratory Experiments
Over the Internet, "IEEE Trans. Educ., 42, 180 (1999)
8. Henry, J.,"Web-Based Laboratories: Technical and Pedagogical Con-
siderations, "AIChE Annual Meeting, Reno, NV (2001)
9. Rueda, L., and T.F Edgar, "Process Dynamics and Control Experiments
Carried Out Over the Internet, "AIChEAnnual Meeting, San Francisco
10. Colton, C.K., "Heat Exchanger Experiment at MIT," International
Conference on Engineering Education, Valencia, Spain (2003)
11. Latchman, H.A., C. Salzmann, D. Gillet, and H. Bouzekri, "Informa-
tion Technology Enhanced Learning in Distance and Conventional
Education, "IEEE Trans. Educ., 42, 247 (1999)
12. Cameron, I.T., "An Interactive Web-Based Decision Support System
for Hazardous Industry Land-Use Planning," Computers and Chem.
Eng., 24, 1057 (2000)
13. Shin, D., E.S. Yoon, S.J. Park, and E.S. Lee, "Web-Based Interactive
Virtual Laboratory System for Unit Operations and Process Systems
Engineering Education," Computers and Chem. Eng., 24, 1381
14. Aung, W, P Hicks, L. Scavarda, V. Roubicek, and C.H. Wei, "Engineer-
ing Education and Research: A Chronicle of Worldwide Innovations,"
Arlington, VA, USA: iEER (2001)
15. Hough, M., and T. Marlin, "Web-Based Interactive Learning Modules
for Process Control," Computers and Chem. Eng., 24, 1485 (2000)
16. Ferrero, A., S. Salicone, C. Bonora, and M. Parmigiani, "ReMLab: A
Java-Based Remote, Didactic Measurement Laboratory, "IEEE Trans.
Instrum. Meas., 52, 710 (2003) 1

Chemical Engineering Education

--- - ^ K.___________________________-




Villanova University * Villanova, PA 19085-1681
The teaching of statistics can be one of the most chal-
lenging topics in the engineering curriculum. Students
often find the subject matter abstract and the plethora
of equations used in analysis rather confusing. For these
reasons, an applied approach that emphasizes and reinforces
how concepts presented in the statistics course can be used
in the practice of engineering has been proposed.1" An ex-
ample is the use of the senior laboratory course to reinforce
the concepts presented in the engineering statistics course.[2]
A stronger emphasis on case studies and realistic problems
of direct interest to engineering students is also suggested
to help motivate and create a more positive attitude toward
statistics] and engineering education in general.[4]
The statistical analysis project described in this article began
as a reactor simulation for a senior design course project. It
was later integrated into the professional development course,
and, after a curriculum revision, the Applied Statistics course,
over the last five years. The novel aspects of this project are
that the students are given a budget with which to perform

their experimental study, and the experimental results are
made available to the students one day after an experiment
is requested. Although a process simulation is generating
the experimental results, the intent is to mimic a realistic
experimental study where results are not available immedi-

Kenneth Muske is the Mr. and Mrs. Robert F. Moritz Sr. Chair of Systems
Engineering and professor of chemical engineering at Villanova University,
where he has taught since 1997. He received his B.S. and M.S. from
Northwestern (1977) and his Ph.D. from the University of Texas (1990), all
in chemical engineering. Prior to teaching at Villanova, he was a technical
staff member at Los Alamos National Laboratory and worked as a process
control consultant for Setpoint, Inc. His research and teaching interests
are in the areas of process modeling, control, and optimization.
John Myers is an emeritus professor of chemical engineering at Villanova
University, where he had taught from 1963 until his retirement in 1999.
He received his B.S. (1958), M.S. (1960), and Ph.D. (1964) in chemical
engineering from the University of Kansas. His teaching interests are in
the areas of process design, transport operations, and statistics. His
research interests are in the area of process design and operations.
He also served as a consultant to local industries. He currently spends
much of his time traveling.

� Copyright ChE Division ofASEE 2007

Winter 2007

ately and there is an economic limit imposed on the amount
of information that can be obtained.

The pedagogical advantage of this approach is it requires
students to efficiently plan and adjust their experimental data
collection. A similar experimental design philosophy for a gas
chromatography experiment is described in Reference 5. It
also incorporates student data into the analysis exercise. The
integration of data sets collected by students into the teaching
of statistics as part of class projects and exercises has been
widely advocated. The benefits of this integration are the
incorporation of problem-based learning into the statistics
course,[6] and the recognition that experimental data sets
represent observations from a larger popula-
tion distribution, which may yield different
"answers" from a statistical analysis.71 An The appr
important goal of any engineering statis-
tics presentation is the appreciation that a article ..
single measurement does not represent the videe
"true" value.[8]
The approach in this article also avoids that ca
the "video game" syndrome that can occur
in process simulation exercises. Although prOcess
simulation modules can be very useful exe
teaching and learning aids in chemical en-
gineering education, they can also impart
an exhaustive iteration approach to problem
solving and a lack of appreciation for the true time scale of
real engineering processes. The addition of a cost and the
delay of simulation results in this project are intended to
address this issue.

In this project, the students determine the kinetic rate
constants of both the forward and reverse reaction for the
hydrolysis of ethylene to form ethanol.
CH, + H20 = C2,HOH
The hydrolysis is a vapor phase reaction that is catalyzed
by phosphoric acid supported on porous solid catalyst pellets.
The reaction rate for the hydrolysis can be expressed as
R (A) = k,PEP - kP, (1)
in which R(A) is the rate of formation of ethanol (i, I 1 ti ,'i in i.
k is the forward reaction rate constant
(gmol/ fmin-bar2), k is the reverse reaction
'h in this rate constant (gmol/ -min-bar), and PE, Pw,
voids the PA are the partial pressures (bar) of ethylene,
water, and alcohol.
meThe students are told that they have a
me packed-bed tubular reactor available to
cur in carry out hydrolysis reaction experiments.
They must specify the molar flow rates of
ulation the feed components, the outlet reactor pres-
Ses. sure, and the average reactor temperature for
each experiment. The molar feed rates of the
reactants (steam and ethylene) and an inert
gas (methane) may be varied by adjusting
the corresponding flow controllers. Methane is supplied to
the reactor in order to dilute the reacting species and prevent
a runaway reaction. The average reactor temperature and
reactor outlet pressure can also be varied by adjusting the

Tubular Reactor
Length = 1 m
Diameter = 0.05 m
D Void Fraction =40% %

Ge , Gw , G


Safe Operating Temperature Range: 300-400 Deg C
Operating Temperature Limits: 250-450 Deg C
Safe Operating Pressure Range: 47.5-60.5 bar
Operating Outlet Pressure Limits: 34-68 bar

Reaction Specificationn

T Reactor average temperature (Deg C)
Po Reactor outlet pressure (bar)
G Ethylene molar feed rate (gmol/min)
G, Water molar feed rate (gmol/min)
Gm Methane molar feed rate (gmol/min)

,,1,- ,Molar Flow Operating Limits: 0*-20 gmole/min
EthLleneolar Flow Operating Limits: 0*-25 gmole/min
Methane Molar Flow Operating Limits: 0*-25 gmole/min
* Flow rates below 0.01 can not be accurately controlled

Reaction Result,

P Inlet Pressure (bar)
YA Alcohol (mole fraction)
Y Water (mole fraction)
YH Ethylene+Methane (molfrac)

Figure 1. Experimental reactor system.

Experimental Costs

$100,000 TotalBudget
$1000 Reaction Experiment
$200 Replicate Experiment
$2500 Expedite Results
$9500 Replacement Reactor

Chemical Engineering Education

. a
o g
n o


respective controllers. The reactor outlet
gas stream is sampled and analyzed for
alcohol fraction and hydrocarbon frac-
tion (ethylene plus methane). Since water
cannot be analyzed, it is determined by
The students are given a feasible reac-
tor temperature range of 300 to 400 �C
and inlet pressure range of 45 to 65 bar.
Under these conditions, the reactor can
be safely operated. There is a potential, * *
however, for the reactor to detonate due
to an exothermic, runaway reaction at
higher temperature or pressure. The initial
students are informed that temperatures
beyond 400 �C and inlet pressures beyond
70 bar are dangerous and can very likely
result in detonation of the reactor. Operation of the reactor
with methane in the feed at the higher temperature and pres-
sure range is also recommended. The students must therefore
first determine safe operating conditions from initial experi-
mental trials as discussed in the sequel.

The project is carried out in two- or three-person groups.
Each student group is given a $100,000 budget to carry out
the experiments necessary to determine the reaction-rate
constants. Each experiment costs $1,000 for the initial run at
a given set of operating conditions and $200 for each replicate
run at the same conditions. The results from each experiment
are made available the day after they are requested. An ad-
ditional $2,500 cost is incurred in order to receive the results
on the same day for each expedited experiment and replicate
requested. Experiments can no longer be carried out when
there are insufficient funds to cover the cost. If the chosen
operating conditions cause the reactor to detonate, the students
are charged $9,500 for a replacement. The intent of this aspect
of the project is to illustrate that, as in an actual experimental
study, there are consequences to poor experimental design
choices. A schematic of the reaction system is presented in
Figure 1.

The students are asked to determine the Arrhenius equation
parameters, activation energy, and pre-exponential factor for
the forward and reverse rate constants. They are also asked to
verify that the rate constants follow the Arrhenius equation

k = ko exp (-Ea / RT) (2)

over the feasible reactor temperature range where ko is the
pre-exponential factor, E is the activation energy, and T is
absolute temperature. Both ko and E can be determined by
obtaining each rate constant at two or more temperatures and
using the logarithmic transformation of Eq. (2)
Winter 2007

regression line through the origin
slope = kf

Figure 2. Example initial rate data regression.

Ink = Inko -E 1 (3)

where In k is the y-intercept and -Ea/R is the slope of a linear
regression of In k as a function of 1/T. In order to determine
the forward and reverse rate constants students must carry
out two different types of experiments.
Irate data non-initial R ate ExperimentsPw
Thgure 2. Example initial rate method of measuring reaction rate constants

is used to detennine the forward reaction rate constant kf. This
Ink = Ink E- l (3)

wherique make is the followinintercept and assumption: 1) there of a liners so
little product fof edn k as a function of T. In order to determine
and,the forward and reversion of the cnstants is small ents must carry
out two different types of experim taken as constant. Using these
Initial Rate Experiments

The initial rate method assumptions with an idof measuring reaction rate constants
is used to determine the forward reaction rate constant kf. This
technique makes the following assumptions: 1) there is so
little product formed that the reverse reaction is negligible;
and, 2) the conversion of the reactants is small enough that
their concentrations may be taken as constant. Using these
initial rate method assumptions with an ideal tubular reactor
results in the following relationship for the outlet alcohol
mole fraction
yA = kfPEPWem (4)
where y, is the mole fraction of alcohol in the exit gas, kf is
the forward reaction rate constant, PE and P are the partial
pressures of the reactants, and em is the molar space time
defined as
m =V/F (5)
in which V is the void volume of the reactor and F is the molar
feed rate of gas entering the reactor.
Determination of the forward rate constant can be accom-
plished by noting that yA is directly proportional to the product
PEPw m in Eq. (4) where the proportionality constant is k,.
A plot of yA vs. PEPw m should be a straight line through the
origin with slope k,. When em increases beyond the value
where the initial rate method assumptions are valid, yA < k,
PE PW Om because the reverse reaction will begin to become
significant. Therefore, one would expect the data to begin to
deviate from a straight line when the initial rate method as-
sumptions are no longer valid, as illustrated in Figure 2.

A value for the forward rate constant can be determined
from the slope of a linear regression on the initial rate ex-
perimental data through the origin. The confidence interval
on the rate constant is obtained from the confidence interval
on the slope of the regression line.
Equilibrium Experiments
If the reactor is operated at low enough feed rates, the
reaction will reach equilibrium at the reactor outlet. The
equilibrium constant for the reaction can then be determined
from these experiments:

Kp YA kf (6)
KP = - - ' (6)
ePw YEYwPo kr
where P is the reactor outlet pressure. The reverse reaction
rate constant can be determined once the forward rate constant
and the equilibrium constant are known from Eq. (6).
Determination of the equilibrium constant can be accom-
plished by noting that yA is directly proportional to the product
YE yw Po in Eq. (6), where the proportionality constant is Kp. A
plot of yA vs. yE yw Po should be a straight line with slope Kp.
When m is below the value required for the reaction to reach
equilibrium, yA < Kp YE yw Po. Therefore, one would expect
the data to deviate from a straight line when the reaction is
not at equilibrium, as illustrated in Figure 3.
The equilibrium constant can be determined from the slope
of a linear regression on the equilibrium experimental data
through the origin. The confidence interval on the equilibrium
constant is obtained from the confidence interval on the slope
of the regression line. The reverse rate constant is calculated
from the ratio of the forward rate constant to the equilibrium
constant at a given temperature.

The students are instructed to select at least three tempera-
tures to study. At each temperature, they are encouraged to
perform exploratory experiments to determine the feed rate

range that will give measurable initial rates and the feed rate
range that results in equilibrium. Based on this information,
a series of initial rate experiments to determine the forward
rate constant and equilibrium experiments to determine the
equilibrium constant should be conducted at different feed
rates and compositions.
In order to carry out initial rate experiments, the reactor
must be operated with high feed rates that result in low outlet
alcohol concentration and low consumption of reactants. Al-
though short residence times are necessary for the assumptions
made by the initial rate method in Eq. (4) to be valid, the high
flow rates will also result in high inlet reactor pressures due to
pressure drop across the catalyst packed in the tube. Therefore,
students are encouraged to initially obtain an estimate of the
pressure drop at high flow rates. Class discussion is used to
suggest that this analysis may be safely carried out by oper-
ating the reactor without one of the reactants. The low feed
rates necessary for the equilibrium assumption in Eq. (6) to
be valid can be obtained without similar issues.
Class discussion is also used to point out possible sources
of variability in the reaction system study such as error in
laboratory analysis and experimental operating conditions.
Measuring instruments are often imprecise and/or inaccurate,
operating conditions cannot be set precisely as desired, and
factors that cannot be observed or controlled can affect the
behavior of any system under study. Therefore, any attempt
to duplicate or repeat a single set of experimental conditions
will usually produce different results. Sometimes the magni-
tude of this variation is small enough that useful conclusions
can be drawn from a single experiment. At other conditions,
however, an experiment must be repeated a number of times
to be confident that the average value is an acceptable repre-
sentation of the actual value.


The students

regression line through the origin
slope = Kp /

equilibrium data

* . * non-equilibrium data
* *
* * *

** *

* *

obtain experimental data by e-mailing the
desired reaction conditions for each experi-
ment using a specified procedure outlined
in the project description handout. The
costs of the experiments are deducted
from the student group's budget as they
are performed. The results are made avail-
able by e-mail to each group member the
morning of the following day for normal
experiments and by that afternoon for ex-
pedited experiments. The results include a
summary of the experimental costs and the
remaining budget.
A separate e-mail account using the class
number as the e-mail address is created
each year for this project. Scripts were
developed to extract the operating condi-
tions from the e-mail message, pass this
Chemical Engineering Education

Figure 3. Example equilibrium data regression.

information into the simulation and run it, create a report
containing the experimental results and budget information,
and then e-mail this report back to the student group. The
original intent was to automatically perform each of these
tasks without the intervention of the instructor. This approach,
however, was quickly abandoned. The ability of undergradu-
ate students to continuously find incorrect permutations to
the required e-mail format resulted in increasing complexity
to the data extraction script. Keeping in touch with each
group's progress and the experiments they requested was
also valuable. For these reasons, the project is administered
by manual execution of the scripts. The administration task
typically takes no more than 10 to 15 minutes each morning.
As the report deadline approaches, the time commitment does
increase slightly as a larger fraction of student groups request
experiments on a given day.

The reactor simulation is performed using the Octave
computational environment running under the Debian linux
operating system. Octave is a freely available mathematical
computation package with similar capability to MATLAB.
We note, however, that the Octave program files generated
to support this project will not run in MATLAB. Additional
information on Octave may be found at the Web site octave.org>.
The reactor is simulated using an isothermal, steady-state,
ideal plug flow reactor model. The forward and reverse rate
constant activation energy and pre-exponential factor values
are modified by the instructor each year. Literature values for
these parameters are not used in order to prevent the more
industrious student from obtaining the answer and reverse en-
gineering their analysis. The values are also changed each year
in order to prevent the less industrious student from getting
values out of a prior-year project report. We note that these
values are a function of the catalyst system used in the reactor
and would be expected to change with different catalysts.

Normally distributed random variation is added to the
specified values for reactor operating conditions. A standard
deviation of 7.5 x 10" mol/min is used for the variation added
to each of the requested molar flow rate values and 5 x 10'
bar is the standard deviation used for the variation added to
the requested outlet pressure. There is no variation added to
the requested average reactor temperature and the simulation
assumes a constant temperature at this value. The pressure
drop across the reactor is determined from the expression

P, = Po + oun (7)
where P is the inlet pressure (bar), Po is the specified outlet
pressure (bar), u is the inlet gas velocity (m/min), a = 1.25
x 104 and 3 = 1.25. These values provide a reasonable pres-
sure drop range for the molar flow rate limits. Slight changes
in these values have been made between years. Normally
Winter 2007

distributed random variation with a standard deviation on the
order of 2 x 10" is added to the ethanol mole fraction. The
standard deviation of the variation in the hydrocarbon mole
fraction is typically half that of the ethanol variation. Slight
changes in these values have been made between years. The
water mole fraction is determined by different checks made
to ensure that reported values are positive and consistent.
Determination of reactor detonation is made by comparing
the requested reactor average temperature and computed inlet

The students are informed that temperatures
beyond 400 �C and inlet pressures beyond
70 bar are dangerous and can very likely result
in detonation of the reactor.... The students
must therefore first determine safe operating
conditions from initial experimental trials ....

pressure to a table of values. Temperatures below 375 �C
or inlet pressures below 70 bar cannot result in detonation.
Temperatures above 400 �C require inlet pressures above 69
bar for detonation, temperatures above 390 �C require inlet
pressures above 72.5 bar, and so forth. These limits are chosen
to make detonation rather difficult unless one is either very
careless or intentionally wants to detonate the reactor. There
have been few unintentional reactor detonations in our experi-
ence with this project. There have been a number of groups,
however, who intentionally try to detonate the reactor with
their last experiment. Although this practice is not within the
scope of presenting a realistic experience to the students, it is
not actively discouraged because it does provide a source of
amusement and a final goal for some group members.

For each temperature selected, the students are instructed
to plot the experimental outlet alcohol mole fraction as a
function of PE P, Om and yE y, Po to determine which data
points represent initial rate conditions and which data points
represent equilibrium conditions. Deviation from the lines
shown in Figures 2 and 3 by a given data point can be caused
by experimental variation and/or violation of the assumptions
made in the corresponding derivation. Although replicate
experimental runs can help quantify the experimental variabil-
ity, they do not provide the information necessary to exactly
determine the point at which the initial rate and equilibrium
assumptions are violated. This determination requires some
judgment by the students.

A linear regression analysis on the selected initial rate and
equilibrium data points is performed using a least squares
linear fit through the origin at each temperature studied. The

A number of


intentionally try

to detonate

the reactor

with their last


Although this

practice is not

within the scope

of presenting

a realistic

experience to

the students, it

is not actively


because it does

provide a source

of amusement

and a final goal

for some group


forward rate constant and equilibrium constant are determined from the slope of the
respective lines. A 95% confidence interval on each of these values is determined
from the standard error of the slope. These calculations are typically performed by
the students using the EXCEL regression data analysis tool. The formulas may also be
found in a number of introductory statistics texts. An extensive summary of statistics
texts can be found in Reference 8 and is not replicated here.
A value for the reverse rate constant can be obtained from rearranging Eq. (6) to
yield k, = k /K . The determination of a confidence interval, however, is more prob-
lematic. The reverse rate constant is the ratio of two independent t-distributed random
variables. The result is a Cauchy distributed random variable with an undefined vari-
ance.[9] The unbounded variance arises from the fact that there is a finite probability
that the equilibrium constant can be within an arbitrarily small neighborhood of zero.
Further discussion of this aspect of the project is presented in the section on discus-
sion topics.
A linear regression analysis based on Eq. (3) can be performed on both the forward
and reverse rate constants to determine the activation energy and the log of the pre-
exponential factor. This linear regression is also typically performed by students using
the EXCEL regression data analysis tool. The students are asked to explain any rate
constant values that they believe are inconsistent with the others and excluded from
the regression. The activation energy can be determined from the slope using the
relationship Ea = -mR, where m is the linear regression slope, and a 95% confidence
interval can be determined from the confidence interval of the slope by scaling with
the gas constant. The pre-exponential factor can be determined from the exponential
of the intercept.
The students are asked to determine an estimate of the error variance for the labora-
tory ethanol analysis from the variance of residuals for each initial rate constant and
equilibrium constant linear regression. The result is two error-variance estimates for
each temperature studied. They are asked to discuss any differences between the esti-
mated variances and whether the error in the alcohol analysis depends on the amount
present in the sample. A 95% confidence interval on the analysis error is determined
from the standard error computed from a pooled variance.

Students report their results in a short group memo to the instructor. The memo
must contain a description of how the group arrived at their results, and enough detail
for someone to replicate their results. An appendix to the memo should contain all of
the data that was obtained. Plots of all the initial rate and equilibrium data with the
regression line and an indication of which points were used in the regression must be
included for each temperature selected. An Arrhenius plot for the forward and reverse
rate constants with the regression line and an indication of any rate constant values
that were not used in the regression must also be included.
Each group is scheduled for a 10-minute appointment with the instructor where
only the instructor and the group members are present. The students turn in the memo,
present their results, and answer any questions about their experimental plan and
statistical analysis. The intent of this oral presentation is to provide an opportunity
for the students to experience a technical interaction with a supervisor that many will
encounter early in their careers as practicing engineers.

The project described in this article brings up a number of topics for discussion
concerning the application of the statistical analysis techniques presented in the Ap-

Chemical Engineering Education

plied Statistics course. The first topic typically brought up in
discussion is the method used to determine valid initial rate
and equilibrium experimental data. Although many student
groups use the "eyeball" method to perform this determina-
tion, a more rigorous approach is to perform the regression
with and without a given data point and look at the effect on
the slope, confidence interval, and correlation coefficient.
For points that are questionable, replicate experimental data
should be used to help determine whether the deviation is due
to experimental error alone.

A second topic for discussion is the basis for the linear
regressions used in this project. The students are reminded
that the regression equations given in their statistics text, and
carried out by EXCEL, assume that there is no error in the
independent variable. This assumption is clearly violated in
the rate and equilibrium constant regressions due to error
in the outlet composition measurements and the Arrhenius
expression regression due to error in the average reactor
temperature. Although an estimate of the magnitude of in-
dependent variable error can be obtained from the ethanol
analysis error variance, a formal treatment of linear regression
in this case is outside of the scope of the one-semester Applied
Statistics course. It is anticipated that student groups would
acknowledge that the regression assumption was violated.
Very few student groups, however, realize this point without
being prompted during the group oral presentation or class
discussion. A very valuable contribution from this aspect of
the project is to reinforce to the students that they must con-
sider the basis and limitations of a statistics formula before
they start performing any calculations.
Some student groups attempt to determine a reverse rate
constant confidence interval by dividing the maximum error
of the forward and equilibrium constants. A less suspect ap-
proach adopted by many student groups is to determine the
confidence interval by approximating the variance from the
forward rate and equilibrium constant variances as follows

S2 22 k
s2 kr 2 &kr 2
s- kf kf OK KP

1 kr s (8)
K2 kf KKp 8

where the partial derivatives are obtained from the rear-
rangement of Eq. (6), and s2kf, and s2, are obtained from
the standard error of the slope from corresponding linear
regressions. This variance is used to compute the standard
error and a confidence interval is obtained from a t-distribu-
tion. A confidence interval for the pre-exponential factor is
obtained by most student groups from the exponential of
the 95% confidence interval of the intercept. Some groups
determine the variance of the pre-exponential factor from
that of the intercept from

2 =Oln(ko)2 2
Sko 0k In(ko

1 2
S 1 n(ko)

and then compute the standard error and confidence interval
using this variance and a t-distribution. These approaches are
not correct. Confidence intervals on the reverse rate constant
and pre-exponential factor cannot be determined because the
parameter variance is undetermined. This aspect of the proj-
ect attempts to reinforce the concept presented early in the
statistics course that nonlinear transformations of normal or
t-distributed random variables no longer retain their original
distribution. Although it is fair to criticize the practice of
asking for values that cannot be computed by the students,
they may very well find themselves in this position later in
their careers and should have some experience in realizing
this point.
A further area of discussion on this topic is how one could
obtain a confidence interval for the reverse rate constant and
whether there is a more accurate method to determine its
value. The students are prompted to consider a revision of
the experimental plan that involves performing initial rate
experiments using ethanol as the feed. In this case, the reverse
rate constant can be determined directly from a single set of
experiments in the same manner as the forward rate constant.


The student groups are given about five weeks toward
the end of the semester to complete this project. They are
reminded in class during this period that it takes time to
obtain data and they should not wait for the last minute to
begin collecting data. Most student groups have successfully
determined forward and reverse rate constants for at least
three temperatures and have obtained reasonable values for
the activation energy and pre-exponential factor. Very few
groups have been unable to determine these values. The most
typical reasons are the group started their data collection too
late in the semester to obtain enough data and/or they were
very inefficient in their experimental plan and expended their
budget. Grading of the project in these cases is based on their
pattern of experimental data requests. Groups that started early
and appeared to have a plan but didn't quite get enough good
data are treated in a much more forgiving manner than groups
that waited for the last minute to request all of their data with
little or no planning.

Groups have been formed both by students' own selection
and by assignment of the instructor. There have been fewer
cases of incomplete or poorly executed projects with the
assigned groups, in our experience. Groups are instructed
not to discuss any aspect of the assignment with anyone out-
side of their group, including the exchange of experimental
data. Although it is difficult to enforce complete compliance
with this policy, analysis of requested experiments has not
revealed any obvious signs of copying experimental designs
between groups or the use of data that was not requested by
a group. We note that no two groups have ever obtained the

Winter 2007

same values for the Arrhenius parameters or used exactly the
same number of experiments in a given semester. We have not
performed this analysis between different semesters.


The experimental design and statistical analysis project
documented in this article has been developed to provide a
realistic experience to students. Based on comments contained
in course surveys, students have found the project to be in-
teresting and worthwhile. A number of students have made
positive comments on the realistic nature of the experience.
Although not incorporated into the scope of this project, ad-
ditional studies - such as an analysis of variance to determine
the sources of variability in the experimental data-can be
included within the framework discussed in this article. This
project has also provided valuable documentation of the stu-
dents' ability to design, conduct, analyze, and interpret experi-
ments for Criterion 3b of the current ABET criteria.1101


A curriculum revision grant to the Villanova University
Chemical Engineering Department from Air Products and
Chemical Co. that supported the development of this project is
gratefully acknowledged. We would also like to acknowledge
the helpful advice of Dr. John Eaton on the development of the
Octave simulation model software and the statistical analysis

discussions with Profs. Dorothy Skaf of Villanova University
and Babatunde Ogunnaike of the University of Delaware.

1. Nelson, P, and T. Wallenius, "Improving the Undergraduate Statistical
Education of Engineers, "in G. Cobb, "Reconsidering Statistics Educa-
tion: A National Science Foundation Conference, "J. Stat. Educ., 1(1)
2. Prudich, M., D. Ridgway, and V. Young, "Integration of Statistics
Throughout the Undergraduate Curriculum: Use of the Senior Chemi-
cal Engineering Unit Operations Laboratory as an End-of-Program
Statistics Assessment Course, "Proceedings of the 2003 ASEE Annual
Conference (2003)
3. Romero, R., A. Ferrer, C. Capilla, L. Zunica, S. Balasch, J. Serra, and R.
Alcover, "Teaching Statistics to Engineers: An Innovative Pedagogical
Experience," J. Stat. Educ., 3(1) (1995)
4. Mustoe, L., andA. Croft, "Motivating Engineering Students by Using
Modern Case Studies," Int. J. Eng. Educ., 15(6) (1999)
5. Ludlow, D., K. Schulz, and J. Erjavic, "Teaching Statistical Experi-
mental Design Using a Laboratory Experiment," J. Eng. Educ., 84(4)
6. Mackisack, M., "What is the Use of Experiments Conducted by Sta-
tistics Students?"J. Stat. Ed., 1(2) (1994)
7. Vaughn, T., "Teaching Statistical Concepts with Student-Specific Data
Sets, "J. Stat. Ed., 11(1) (2003)
8. Fahidy, M., "An Undergraduate Course in Applied Probability and
Statistic," Chem. Eng. Ed., 36(2) (2002)
9. Evans, M., N. Hastings, and B. Peacock, Statistical Distributions, 3rd
Ed., Wiley, New York (2000)
10. ABET, Criteria for Accrediting Engineering Programs, Engineering
Accreditation Commission, (2004) 1

Chemical Engineering Education

--- - ^ K.___________________________-



Middle East Technical University * Ankara, Turkey 06531

Forced convection inside circular pipes under fully
developed conditions is one of the main subjects
covered in both undergraduate- and graduate-level
heat transfer courses. Two types of boundary conditions are
usually considered, i.e., constant wall heat flux and constant
wall temperature. In engineering calculations, heat transfer
correlations are expressed in terms of the Nusselt number and
such expressions require the solution of the energy equation
given as
SOT k r OT 1
pCpv -- r-
Oz r r dr

in which the velocity distribution under laminar flow condi-
tions is given by

vz = 2(vz) 1- (2)

When the heat flux at the wall is constant, the temperature
gradient in the axial direction, 8T /8z, is constant. This makes

the solution of Eq. (1) rather simple since the left side is depen-
dent only on r. Integration of Eq. (1) twice yields the Nusselt
number equal to 48/11. This approach is presented in almost
all textbooks on heat transfer and/or transport phenomena.
In the case of constant wall temperature, however, the
solution of Eq. (1) requires advanced mathematical skills in
partial differential equations.1 As a result of this mathematical
complexity, the value of the Nusselt number is given as 3.66
in textbooks without presenting the analysis. Incropera and

Ismail Tosun received his B.S. and M.S.
degrees from the Middle East Technical
University, and a Ph.D. degree from the
University of Akron, all in chemical engineer-
ing. He is the author of the book Modelling
in Transport Phenomena-A Conceptual
Approach (Elsevier, 2002). His research
is directed to the unification of solid-liquid
separation processes using the multiphase
equations of change.

i ___________i__

� Copyright ChE Division of ASEE 2007

Winter 2007

DeWitt,[2] for example, stated that:
". .. the solution may be obtained by an iterative proce-
dure, which involves making successive approximations to
the temperature profile. The resulting profile is not described
by a simple algebraic expression, but the resulting Nusselt
number may be shown as Nu = 3.66."
The method of Stodola and Vianello3, 4] is an approximate
technique used for a quick estimation of the first eigenvalue.
The purpose of this paper is to show students how to apply
this technique in the calculation of the Nusselt number for
forced convection in a circular pipe when the wall temperature
is constant. From my experience in teaching graduate-level
Transport Phenomena and Heat Transfer courses, the method
is well received by students.
Consider the laminar flow of an incompressible New-
tonian fluid in a circular pipe of radius R under the action of
a pressure gradient. The fluid is at a uniform temperature of
T for z < 0. For z > 0, the wall temperature is kept constant
at T (T > T ) and we want to develop a correlation for heat
transfer in terms of the Nusselt number under thermally fully
developed conditions.
As engineers, we are interested in the variation of the bulk
(or, mean) fluid temperature, Tb, rather than the local tempera-
ture, T. The bulk fluid temperature is defined by

fb , fR vzTrdrdO
b " 2O RO
f2 fR vz r drdO

1 R2(vzo f R vzTrdrd (3)

Since both v. and T are independent of 6, Eq. (3) simplifies
Tb =-- )fvzTrdr (4)
R "(vz
The governing equation for the bulk temperature can be
obtained by multiplying Eq. (1) by r dr and integrating from
r = 0 tor =R, i.e.,

pC R vz OTrdr
o 9 Oz

rR1 0f G T (T
k -- r- rdr
Or r Or r

Since v z v(z), the integral on the left side of Eq. (5) can be
rearranged, with the help of Eq. (3), as

pCP fv -rdr= pCPf vT) rdr
o z zOz
d R
= pCP vzTr dr
dzJo o(

= pCP O (6)
2 dz

On the other hand, the integral on the right side of Eq. (5)
takes the form
R OT 9OT r=R
kR -- r- T rdr = k r- (7)
Jor Or Or Or r = 0
The heat flux at the wall, qw, is defined by

qw = k - (8)
Orr r= R
so that Eq. (7) becomes
p R 1 ) (T
k1 -- - rdr Rq (9)
Jr r Orr )
Substitution of Eqs. (6) and (9) into Eq. (5) results in the
governing equation for the bulk temperature as
dTb 2 qw (10)
dz pCpR(vz)

The requirement for a thermally fully developed flow is
expressed as

0 _ T-TT j= 0 (11)
Oz Tw - Tb
Note that the thermally fully developed condition also implies
that the local heat transfer coefficient, h, is a constant.
When the wall temperature, Tw, is constant, Eq. (11) re-
duces to

OT Tw - T dTb
z T - Tb) dz

Substitution of Eq. (10) into Eq. (12) results in

OT T 2q (13)
z T- T,) pCpR(vz)

The Nusselt number is defined by

Nu = (14)
The heat flux at the wall is also expressed in terms of the
Newton's law of cooling as
q, = h(T - Tb) (15)
so that the Nusselt number takes the form

Nu qw/(Tw -Tb) 2R (16)
k (16)

Chemical Engineering Education

Elimination of the term qw between Eqs. (13) and (16) leads
OT Nu(T -T)k (17)
Oz pCpR2(vz)
Substitution of Eqs. (2) and (17) into Eq. (1) yields

2Nu 1- R (T - T) = -- [r (18)
e b ry r Or)
The boundary conditions associated with Eq. (18) are

atz =0

atr = 0

atr= R


In terms of the following dimensionless quantities
T-Tb T -T (20)
9=1- - = (20)
Tw -T Tw -T
r (21)
the governing equation together with the boundary conditions
take the form

The first approximation to X1, I1(1), is given by

Sfbw(x) fl(x)yl (x)dx
X11 b 2-= (27)
fa w(x)[fl(x)]2dx
4. Repeat step (2) with a second trial function y,(x) defined

y2(x)= f(x)


5. Solve the resulting differential equation and express the
solution in the form
y(x)= Xf(x) (29)
The second approximation to 1, 1 (2), is given by
)(12) fbw(x)f2(x)y2(x)dx
1= (30)
f w(x)[f,(x)2 dx
6. Continue the process until the desired convergence is
For the problem at hand, comparison of Eq. (22) with Eq.
(25) gives
y= x= p= X=Nu w=2(1- E2) (31)

1 d dO
2Nu(1- ) = --
( dd

at = 0

at= =l


It should be kept in mind that the dimensionless tempera-
ture, 6, is dependent on only the dimensionless radial coor-
dinate, �, for the thermally fully developed condition. Eq.
(22) can be easily solved for Nu by the method of Stodola
and Vianello.

The method of Stodola and Vianello[3 4] is a successive ap-
proximation technique to estimate the lowest eigenvalue ?1
in the boundary value problem of the form

d dy
- p(x)\ x
dx dx


with appropriate homogeneous boundary conditions at x = a
and x = b. The procedure is as follows:
1. Assume a trial function for y, (x) which satisfies the
boundary conditions x = a and x = b.
2. On the right side of Eq. (25), replace y(x) by y,(x).
3. Solve the resulting differential equation and express
the solution in the form

A reasonable first guess for 6 which satisfies the boundary
(22) conditions is

((23) 0) 1=- (32)
(23)Substitution of Eq. (32) into the leftside of Eq. (22) gives
Substitution of Eq. (32) into the left-side of Eq. (22) gives

d -2Nu(-2 +s) (33)

The solution of Eq. (33) using the boundary conditions given
by Eqs. (23) and (24) is

0 11- 182 9� - 2 j


The first approximation to the Nusselt number is calculated
from Eq. (27) as

Nu(1) f1� (1- 2)2 f()d�
fNu I(= - -2)fl2 ()d
Jo (1- ()f (0)d


Substitution of f1 (� ) from Eq. (34) into Eq. (35) and evalu-
ation of the integrals gives
Nu(1 = 3.663 (36)
The trial function for the second approximation is
11- 18_ 2 + 94 - 26
0, (E = (37)

(26) Substitution of Eq. (37) into the left-side of Eq. (22) gives

y(x)= Xf,(x)

Winter 2007


d 1d Nu
- d - N(1- 293 + 272 -11+ 2(9) (38)
d� d� 18
The solution of Eq. (38) using the boundary conditions given by Eqs.
(23) and (24) is

= Nu( 2457 - 4400k2 + 29004 - 1200,6 + 275$8 - 3210 (3

Therefore, the second approximation to the Nusselt number is given

Nu2 = (1- 2)f2 ()02 () d
f I(1- _ 2)f22 () de


Substitution of f,(&) and 62(&) from Eqs. (39) and (37), respectively,
into Eq. (40) and evaluation of the integrals gives

Nu(2) = 3.657 (41)
which is equal to the exact value calculated by Graetz and Nusselt.

The solution of eigenvalue problems by the method of Stodola
and Vianello gives accurate results and convergence is very rapid.
Although the integrals seem formidable, they can be easily evaluated
using engineering calculation software such as MATHEMATICA or
MATHCAD. The method is easy to follow and students have no dif-
ficulty in understanding the presented material. The transformation
of the governing differential equation, Eq. (1), to the form to which
the method of Stodola and Vianello is applied, Eq. (18) or Eq. (22), is
also very helpful for students in grasping the concept of area averag-
ing and the difference between local and bulk temperatures, as well
as their functional dependence on coordinate directions.

IP Heat capacity, J kg-K
D Pipe diameter, m
k Thermal conductivity, W/m-K
Nu Nusselt number, dimensionless
qw Heat flux at the wall, W/m2
R Pipe radius, m
r Radial coordinate, m
T Temperature, K
Tb Bulk temperature, K
Tw Wall temperature, K
v Axial velocity, m/s
(vz) Axial average velocity, m/s
z Axial coordinate, m
Greek symbols
6 Dimensionless temperature
& Dimensionless radial coordinate
9 Density, kg m3

1. Sellars, J.R., M. Tribus, and J.S. Klein, "Heat Transfer to
Laminar Flow in a Round Tube or Flat Conduit-The Graetz
Problem Extended, " Trans. ASME, 78, 441 (1956)
2. Incropera, EP, and D.P DeWitt, Fundamentals of Heat and
Mass Transfer, 5th Ed., Wiley, New York (2002)
3. Bird, R.B., R.C. Armstrong, and 0. Hassager, Dynamics of
Polymeric Fluids, Volume 1: Fluid Dynamics, 2nd Ed., Wiley,
New York (1987)
4. Hildebrand, EB., Advanced Calculus for Applications, 2nd Ed.,
Prentice-Hall, Englewood Cliffs, NJ (1976) [

Chemical Engineering Education


Future of Chemical Engineering:



Rowan University * Glassboro, NJ 08028
C creating a working knowledge of biological principles Patricia Mosto has extensive environmental science experience. She has
in students and developing their ability to apply been actively involved with field and laboratory projects related to water
engineering principles to biological systems (and quality and pollution issues for the last 30 years. She has worked with the
Departments of Water and Power and Sanitation in Los Angeles for 10 years.
vice versa) is recognized nationwide as a goal for chemical In her 14 years at Rowan, she has supervised more than 50 independent
engineering programs.[15] The same can be said from a global undergraduate projects, taking many students to national and international
conferences. She is author of over 100 publications.
perspective.[6] There is currently a significant movement to
change chemical engineering department names to reflect Mariano Savelski has seven years of industrial experience in design and
in biofocused engineering. Bioengineering manufacturing. He has received a Lindback Foundation Award to continue
faculty expertiseinbio-focused engineering Bioeng ng his research in the area of industrial wastewater minimization, as well as
is very broad and inherently interdisciplinary. The need for a U.S. EPA Award to investigate zero water discharge cycles in manu-
bioengineers is on the rise. By 2010, there is projected to be facturing and chemical plants. He has been recognized as a rising star in
chemical engineering and participated as a panelist in the 2001 Galaxy of
a 31.4% increase in employment positions in bioengineering Stars at the ASEE meeting in Albuquerque. He has been actively involved
fields.[8' Moreover, most engineering jobs listed in the "Fast in undergraduate research through Rowan Engineering's clinic.
Company 25 Top Jobs" are bio-related.[9] To meet the needs of Gregory B. Hechthas extensive research experience in prokaryotic genet-
the global job market today's chemical engineering students ics and molecular biology. With Dr. Mosto, he has developed a new course
for chemical engineering students, Biological Systems and Applications. He
must receive a solid background in biology. The conventional is the creator and coordinator of the Rowan University Student Research
approach is to add a standard biology course, and many Symposium, an annual forum at which Rowan students from all of the SMET
schools do offer biology courses at the senior or graduate disciplines present the results of their independent research.
level.[10 11] The integration of biology in the undergraduate Stephanie H. Farrell received her B.S. in 1986 from the University of Penn-
chemical engineering curriculum, however-although dif- sylvania, her M.S. in 1992 from Stevens Institute of Technology, and her
Ph.D. in 1996 from NJIT. Prior to joining Rowan in 1998, she was a faculty
ficult in an already overloaded curriculum-provides a more member at Louisiana Tech University. Her research expertise is in the field
holistic and rewarding learning experience, of drug delivery and controlled release, and she is currently focusing efforts
on developing laboratory experiments related to membrane separations,
biochemical engineering, and biomedical systems for students.
� Copyright ChE Division of ASEE 2007
Winter 2007 4.

Due to the interdisciplinary nature of the field, this holistic
way of teaching integrates both biology and engineering. The
biology provides knowledge and skills dealing with biological
concepts as the building blocks for engineering design and
process. It may also create a whole-system perspective neces-
sary for innovation and creativity. The engineering provides
access to existing technologies with an emphasis on the design
process itself. At Rowan University, we have developed such
an integrated, collaborative approach between engineering
and biology faculty to introduce chemical engineering stu-
dents to the application of engineering principles in biologi-
cal systems throughout their four-year curriculum. Through
specially designed courses and active learning modules that
can be easily adapted to any course, students are exposed to
the newest biological trends for chemical engineering. The
implementation of this philosophy exposes students to key
areas of collaboration between biologists and chemical en-
gineers at early stages in their undergraduate education, and
continues systematically during the upper years. This strategy
develops a cumulative knowledge of biological principles
in students, enabling faculty to build increasing detail and
technical content into problems and projects that address the
interface between biology and engineering. This application
allows students to work in interdisciplinary teams, think in
a more global fashion, create innovative ideas, and enhance
their entrepreneurship and communication skills.

Revisions to the chemical engineering curriculum at Rowan
University include: several laboratory modules and projects
at the freshman and sophomore levels; a novel, required
Biological Systems and Applications course designed to in-
troduce students to a variety of biological principles relevant
to chemical engineering1231; vertical integration of experi-
ments and applications of bio-related engineering analysis
in core engineering courses; collaborative research projects
involving biologists and chemical engineers in their junior
and senior years; team-taught senior chemical engineering
elective courses with strong biological components; and a
bioengineering concentration for those graduating with a
cadre of bio-related courses.

As the only four-year engineering college in Southern New
Jersey, Rowan Engineering is deeply committed to being a
major technological resource for the area, preparing students
for engineering careers in regionally important industries such
as biomedical, biotechnology, pharmaceutical, and food. The
abundance of such industry in New Jersey and nationwide
creates a steady demand for well prepared engineering gradu-
ates. Our collaborative approach to integrating biology and
chemical engineering helps prepare students for careers in
food, biotechnology, and pharmaceutical industries.

This paper will discuss the implementation, impact, and
benefits of our approach, with emphasis on the core courses
and junior- and senior-level engineering experiences. A de-

tailed description of the integration of biological principles
into the lower levels has been published previously."121

Generally speaking, the Freshman Clinic sequence corre-
sponds to Introduction to Engineering courses in many other
universities, though in unique format. It consists of two parts.
In the fall semester we teach basic engineering skills (such
as problem solving and teamwork fundamentals) and ethics
that will be essential to students' success (or even survival) in
engineering school and in their future engineering careers. In
the spring semester students are exposed to an intense study
of engineering design through reverse engineering (or "dissec-
tion") and competitive assessment of consumer products.[13 14]
Comparable products are reverse engineered to gain under-
standing of the mechanisms by which they work.
In the Freshman Clinic we immediately establish a hands-
on, active-learning environment in which students are intro-
duced to a wide range of engineering principles applied to
both biomedical and biochemical systems.1" 15 18]
A strategy for introducing biological concepts throughout
a traditional engineering curriculum using examples, demon-
strations, and experiments has been presented by Maynard and
Razatos.[191 Their approach provides graduating engineers with
the skills to handle nontraditional problems and to address
emerging areas of research and development. We use a similar
approach in integrating biological concepts throughout our
core chemical engineering courses at Rowan. An important
feature of our implementation method is the emphasis on
vertical integration of bio-related course materials and labora-
tory experiments throughout core courses. Vertical integration
enhances educational quality by integrating concepts, skills,
models, and data throughout all levels of the curriculum,
building upon not only the work done in the previous labora-
tories of the same course but also those of previous courses.
Re-using experiments in freshman, core, and elective courses,
as well as in undergraduate research projects, makes efficient
use of laboratory equipment and space. This truly integrated
learning experience enhances student learning, concept reten-
tion, and motivation. 20 221
The Freshman Engineering Clinic biomedical engineer-
ing project mentioned previously in this paper is used here
as an example to illustrate the vertical integration of topics
throughout the curriculum. Through eight hands-on modules,
students in the freshman course are introduced to a variety
of multidisciplinary biomedical topics. Each topic is then
explored in greater depth in the appropriate core courses of
the chemical engineering curriculum. Table 1 shows the topi-
cal content of the eight hands-on modules taught during the
freshman year (first column), with the associated measure-
ments, calculations, and engineering principles (columns 2,3,
and 4). The engineering courses into which the experiments,
analysis, and concepts are integrated appear in the right-most
Chemical Engineering Education

column of the table. While the vertical integration of the
courses is multidisciplinary and involves other engineering
and science disciplines, this table shows only the information
that is directly related to the vertical integration into chemical
engineering courses.

To meet the anticipated growing demand for biology-liter-
ate engineers, faculty from biological sciences and chemical
engineering worked closely together to develop a lab-inten-
sive course open only to sophomore chemical engineering
majors. A detailed description of the Biological Systems and
Applications (BS&A) topical content and laboratory exercises
has been described previously, along with an assessment of
the effectiveness of the course.[23]
Concurrent with the Biological Systems and Applications
course, students take Sophomore Clinic I and II, a multi-
disciplinary engineering design and practice two-semester

course sequence providing them the necessary technical
communication tools. The students work in teams of three
to five for the entire semester. The lecture and laboratory
sessions are structured so that parallel activities support
the eventual completion of the project. In the semesterlong
project student teams design and create a microbial fuel cell
(MFC) that powers a Lego Mindstorms robot. The design of
microbial fuel cells provides an ideal application for many
concepts taught in the BS&A course.[23] In conjunction with
the design project, the first semester focuses on total quality
management and writing, and the second semester focuses
on public speaking.


As part of the clinic sequence at Rowan Engineering, stu-
dents participate in sponsored research projects during their
junior and senior years. Each semester, students work in
multidisciplinary teams as part of a two-credit course. Project

Winter 2007

Biomedical Engineering Modules: Measurements, Calculations,
Engineering Principles, and Vertical Integration of Project Modules into Chemical Engineering Courses

Measurements Calculations Engineering Principles Vertical Integration
Respiration 0,, CO2 concentration Gas volumes Material balances Mass & Energy Balances
Air flow rate Moles of gas PVT relationships Biomedical Processes
Rate of gas consumption (elective)
and production
Metabolism Food intake Energy expenditure Material balances Mass & Energy Balances
Body surface area Energy balances Biomedical Processes
Dimensional homogeniety
Pulmonary Lung volume PV work Mass transfer/separations Thermodynamics
System Air pressure Efficiency PV work Mass Energy Balances
Air flow rate Rate of heat transfer Efficiency Fluid Flow
Blood O0 % saturation Dissolved O0 Energy balance Separations
concentration Gas solubility/Henry's Biomedical Processes
Resistance Poisieulle's
Cardiovascular Heart rate Blood flow rate Mass balance in flow Fluid Mechanics
System Blood pressure system
Fluid flow Bernoulli
Pumps-power and
Work/Power Force Work Work Dynamics (ME, ECE)
Distance Power Energy Kinesiology (HS)
Recovery time Efficiency Power
Mechanics of For bone and cartilage: Stiffness Stress Materials Science
Materials Force (tension and Dampening Strain
compression) Forces
Deformation (tension Deformations
and compression)

funding is provided through either government or industrial
grants or sponsorships. Projects span a wide variety of emerg-
ing disciplines, depending on faculty expertise and availability
of funding. The number of projects that involve integration of
biology with chemical engineering has increased dramatically
during the seven years the Junior/Senior Clinic has existed.
Their preparation during the Biological Systems & Appli-
cation course allows students to tackle these bio-oriented
projects and succeed in their upper-class work.
At the conclusion of four semesters of Junior/Senior Clinic
activities, students are expected to:
* Demonstrate expanded knowledge of the general prac-
tices and the profession of engineering . /h. *, 1 immer-
sion in engineering projects of moderate complexity.
* Demonstrate an ability to work effectively in a multidis-
ciplinary team.
* Demonstrate acquisition of new technology skills.
* Demonstrate understanding of business and entrepre-
neurial skills.
* Demonstrate effective use of project and personnel
management techniques.
* Integrate engineering professionalism and ethics in
their work as it relates to the context of engineering in
* Demonstrate improved communication skills including
written, oral, and multimedia.
* Use information obtained from sources that cross geo-
political and language barriers.

Currently only 2% of U.S. energy needs are met by renew-
able resources. The National Renewable Resources Labora-
tory (NREL), however, projects that biomass resources can
eventually provide more than 50% of transportation fuel,
reducing dependence on foreign sources of energy, alleviating
air pollution problems, and increasing employment oppor-
tunities. Bioethanol is one biofuel that has been receiving a
great deal of attention in recent years. One factor suppressing
wider use of bioethanol is the costs associated with produc-
tion. In North America, most bioethanol is made from the
fermentation of corn. This process sets aside the stalks and
leaves of the corn plant referred to as corn stover. It has been
estimated that if the corn stover available from current crop
yields could be fermented efficiently, bioethanol production
in North America could be tripled."30 Because of its cellulose
and hemicellulose content, however, corn stover is more
difficult to ferment than corn itself and is considered to be a
waste product of corn farming. In particular, the preparation
of the fermentation feedstock and the subsequent increase
in ethanol concentrations can be toxic to the fermenting
microorganisms. The overall objective of this project is to

create and characterize new strains of the bacterium Esch-
erichia coli with the potential to sidestep these issues and, as
a result, produce greater yields of ethanol from corn stover.
For this project, teams of student researchers are assembled
as a cohort of two biology and two chemical engineering
majors, and each cohort works with a team of four professors
(two from biology and two from chemical engineering). The
student cohorts select a particular toxicological problem to
investigate over a two-year period. The sum of each cohort's
project is broken down into modules with specific objectives
that include extensive biological and engineering literature
search and review, isolation of novel toxin-resistant deriva-
tives of known ethanologenic microbes, quantification of the
toxicological properties of the new strains, pilot fermentation
studies to demonstrate the effectiveness of the new strains,
and presentations of their results at national microbiology
and chemical engineering conferences. This module ap-
proach and the cohort composition allows an emphasis on
multidisciplinary learning. The experiments conducted by
the students address applied microbiology, toxicology, fer-
mentation technology, engineering design, economics, and
professional communication. A conscientious effort is made to
ensure that all students in the cohort participate in all phases
of the experimental design and execution, including determin-
ing the effects of altering process variables (e.g., feedstock
composition), isolation, and characterization of the biological
catalyst with the desired properties, assessing the impact of
these activities on the process conditions of the downstream
operations and the overall economic feasibility of the system,
and disseminating the results at professional venues.

Haematococcus pluvials is one of the largest algal produc-
ers of astaxanthin, a carotenoid that is commonly used as a
feed supplement in the salmon farming industry to give the
salmon their pinkish hue. Astaxanthin is the ideal component
to color the salmon because it is a stable natural product and
is naturally retained by the fish's flesh. It has already been
established[241 that extreme light conditions yield a higher
production of astaxanthin in H. pluvials, however exact light-
to-dark time periods for optimum astaxanthin production are
unknown. The goal of this project was to determine the proper
lighting conditions for optimum astaxanthin production by
H. pluvialis so a pilot scale plant for large-scale production
could be constructed. Two students, one from biology and
one from chemical engineering, work over the course of a
year with two professors (one from biology and one from
chemical engineering). The students grow H. pluvialis in an
environmental chamber at different light/dark cycles (16/8,
20/4, 24/0) and constant temperature (26 �C) to determine
the best light-to-dark ratio for maximum astaxanthin produc-
tion. Chlorophyll a, ash-dry biomass, and a cell count were
obtained daily for each of the growth conditions to establish
the optimum growth curves for H. pluvialis. Correlations
Chemical Engineering Education

between growth and astaxanthin production were studied,
and a continuous bioreactor for pilot scale production for
H. pluvials was designed, constructed, and tested. The 16/8
light-to-dark ratio was used, and it was possible to grow the
algae in one compartment and use gravity feed to a separate
compartment where the algae were stressed (e.g., longer
light cycle, carbon dioxide bubbled into reactor) to enhance
astaxanthin production. Several aspects of the reactor were
modified for use in Chilean salmon farms in a large-scale
algae production facility. This year's Junior/Senior Clinic
will complete the design of the reactor.

The Junior and Senior Engineering Clinic projects described
above are just a few examples of collaborative, multidisci-
plinary projects that integrate biological and engineering prin-
ciples. Additional clinic projects have investigated problems
related to drug delivery, food preservation, pharmaceutical
separations, and artificial organs.
The clinic has proven to be a very effective vehicle for
development of educational experiments and course con-
tent. The biomedical, drug delivery, and food engineering
modules that are integrated throughout the curriculum were
developed via the clinic. In a typical project, students would
be responsible for collecting background material, building
the experimental apparatus, developing the experimental
procedure and methods of data analysis, writing a detailed
laboratory handout for students, and providing an instructor's
manual for a module on a given topic.

Food Engineering Course
Rowan Engineering is committed to being a major techno-
logical resource for the area, preparing students for engineer-
ing careers in regionally important industries such as food
processing. The state has major manufacturing operations of
top companies such as The Campbell Soup Co., Coca Cola,
Anheuser-Busch, General Mills, and Kellogg's. The immedi-
ate Vineland area is the hub of Southern New Jersey's food
processing industry, home to about 30 companies employing
3,000 people and producing $700 million in shipments. The
abundance of food processing companies in New Jersey de-
mands a steady pipeline of well-prepared engineering gradu-
ates. Rowan Engineering students respond to the regional
emphasis on food processing with a tremendous interest in
the industry. In their senior exit interviews, an overwhelm-
ing number of graduating seniors strongly indicated a need
for more exposure to food-oriented projects and courses. To
respond to student demand and regional industrial needs,
chemical engineering faculty have secured support in recent
years for undergraduate clinic research projects. Food ex-
periments have been introduced to all engineering students
in the Freshman Engineering Clinic (a multidisciplinary,
Winter 2007

introductory course required of all freshmen) and a new
Food Engineering elective course was designed for chemical
engineering students.
This course provides students with the necessary background
in food science, food chemistry, unit operations relevant to
food industry (rarely taught in traditional chemical engineer-
ing curricula), and finally an approach to food preservation
designed and taught by biological science faculty.
Biomedical Engineering Course
The discipline of biomedical engineering has emerged from
informal collaborations between engineers, physicians, and
life scientists. While relatively new, it is the fastest-growing
engineering discipline at most universities.[20l Chemical engi-
neers play an important and expanding role in this burgeoning
field because core chemical engineering concepts are critical
to solving medical problems such as the design of artificial
organs and drug-delivery devices.
This course introduces students to applications of chemical
engineering fundamentals and biomedical systems. Students
analyze and design biomedical processes through the appli-
cation of advanced principles in mass transfer, heat transfer,
fluid flow and chemical reactions, pharmacokinetic models,
the circulatory system, transport across cell membranes, and
human and artificial organs. Several laboratory experiments
are conducted to explore the circulatory system, respiration,
metabolism, and cardiopulmonary dynamics.
It should be noted that many of the basic biomedical
concepts have been vertically integrated throughout the cur-
riculum, beginning with freshman biomedical modules re-
introduced in relevant core courses. The specific focus of this
course permits these topics and experiments to be explored in
greater depth with a more significant emphasis on the associ-
ated physiology and other biological concepts.
Drug Delivery Course
Controlled-release systems are designed to provide delivery
of a biologically active agent (e.g., a drug or pesticide) at a
predetermined rate for an extended period of time. Controlled
release offers several advantages over traditional methods of
formulation and administration such as: maintenance of ef-
fective concentrations for a sustained period, less total agent
required, cost effectiveness, convenience, and compliance.
This course on controlled-release systems introduces students
to chemical engineering fundamentals applied in controlled-
release systems. Basic principles of materials, mass transfer,
heat transfer, fluid flow, and chemical reactions are used to
analyze and design controlled-release systems. Applications to
pharmaceutical, agricultural, and food industries are explored,
with a primary focus on drug delivery systems. Several labo-
ratory experiments are conducted to explore drug stability,
membrane-based transdermal patches, controlled-release
tablets, erodible and dissolution-based systems, and osmotic
pumps.[26] Drug delivery topics represent another example of

vertical integration of experiments and examples throughout
the curriculum. Freshmen are first introduced to drug delivery
in the freshman year, and drug delivery examples are revis-
ited in core courses such as Transport Phenomena and Mass
Transfer. In the senior-level elective on controlled release,
students explore drug delivery systems in greater depth, with
more emphasis on topics such as distinguishing rate-control-
ling mechanisms and pharmacokinetic considerations.

The combination of modules at the freshman and sopho-
more level, the Biological Systems and Applications course
specifically designed for chemical engineers, the research
projects as part of junior and senior clinics, the elective se-
nior courses in Food Engineering, Biomedical Engineering
and Drug Delivery, and the Concentration in Bioengeneering
all help prepare students for a future career in research and
industry. Located in Southern New Jersey, Rowan University,
through its Junior/Senior Clinic, has successfully completed
a wide range of projects generated and sponsored by local
industries and agencies. These include private companies
(e.g., Biothane, US Filter, Lockheed Martin, Johnson Matthey,
General Mills, ExxonMobil) and research foundations (e.g.,
Engineering Information Foundation, Water Environment
Research Foundation). These industrial partnerships benefit
both the faculty and the students. 27 Students are more likely
to obtain internships as a result of these experiences, and en-
gineering faculty with expertise that reflects this bio-intensive
regional interest strengthen their industry interactions and
receive research support. This research interest is reflected in
the types of clinic projects offered in the Junior/Senior Clinic

1998 1999 2000 2001 2002 2003 2004 2005 2006

Figure 1. Number of bio-oriented abstracts and total abstracts submitted
by engineering students at Rowan University's STEM Symposium.

course, such as bioethanol production, astaxanthin production,
drug delivery, and food engineering. Working cooperatively
with local industry has also enabled students to obtain valu-
able entrepreneurship experience in supporting small- and
medium-size businesses. As part of clinic projects, students
may propose their own ideas and gain funding through the Na-
tional Collegiate Inventors and Innovators Alliance (NCIIA)
Venture Capital Fund. This fund is managed by a faculty
member and specifically earmarked for the development of
original inventions by multidisciplinary student teams within
the Junior and Senior Clinics.[28]

Students often cite a potential career in biochemical engi-
neering as a motivator for pursuing a chemical engineering
degree. This interest in the interplay between biology and
engineering is apparent in the demand by students for bio-
oriented research projects at the junior and senior levels. One
measure of student interest in bio-related projects is their
participation in Rowan University's student symposium in
the Science, Technology, Engineering, and Math (STEM)
Symposium. As shown in Figure 1, the percentage of bio-
related engineering projects that have been presented at the
symposium has increased dramatically. In 1998, only one
engineering abstract at the symposium had biology content.
By 2004, the number of posters with engineering students
pursuing biology-related projects was similar to the number
of nonbiology engineering posters. Importantly, Figure 1
demonstrates that 2004 was not a peak but the realization of
a new status quo, since subsequent years have had similar
numbers of bioengineering presentations. In many cases, the
lab component of the BS&A course has directly benefited
students working on research projects at Rowan University.
As the beginning cadre of students who
have been exposed to these innovations
in the curriculum progresses, we expect to
develop new engineering courses on mo-
lecular biotechnology or bioengineering
S ,,... that will be part of the new bioengineering
-i t concentration approved this year within
t %the College of Engineering.

Preparing students early in their college
career through a specially designed course
and bio-related modules during their
Freshman and Sophomore Clinics yields
excellent results on their bioengineering
clinic project and courses in their junior
and senior year. Additionally, students are
able to learn more material and applica-
tions in the time that is traditionally spent
in an introduction to biological principles.
Also, the Junior/Senior Clinics fostered a
strong research environment, evidenced
by the percentage of students pursuing
graduate degrees as shown in Figure 2.[29]
Chemical Engineering Education

20 - O Biology Abstracts w/ Engineenng Students
8 Engineenng Abstracts
* Engineenng Abstracts with Biology Conten
S10 -



The impact of the clinic model has been very positive in foster-
ing a spirit of inquiry and engaging students in cutting-edge
research as undergraduates.
As a final measure of impact on students, career paths of
chemical engineering graduates are considered. The AIChE
Placement Survey for Recent Graduates from domestic
institutions indicates that 22.5% of chemical engineering
graduates found work in biotechnology, pharmaceutical, and
food industries. A survey of Rowan Chemical Engineering
graduates reveals that more than 27% of chemical engineering
graduates found employment in these industries.

The implementation of these innovations into the curricu-
lum was relatively smooth, particularly considering that it has
required cooperation across not just separate departments but
also separate colleges within the university. We believe that
several factors were crucial to this success.
Foremost, the culture on the Rowan campus during the
implementation process was focused on de-emphasizing the
protection of "turf" by the academic departments and moving
towards interdisciplinary activity. Importantly, relations be-
tween the Departments of Chemical Engineering and Biologi-
cal Sciences were collegial at the start of the implementation,
as were the interactions between the deans of the College of
Engineering and the College of Liberal Arts and Sciences.
Moreover, both the engineering and the biology personnel
viewed the curricular development as a mutually beneficial
process. While the curricular development described here has
had an obvious benefit for the chemical en-
gineering department, it has also resulted in
dividends for the Department of Biological 45.00%
Sciences. No less than four biology faculty
have been involved in numerous collab- 40.00%
orative research projects, some of which
received external funding. Even better, the o 35.00%
addition of biology to the curriculum has (n
provided additional research opportunities - 30.00%
for biology majors.

Incremental implementation was also im-
portant. Incorporation of biological content
and application into the curriculum required
resources from both departments, which to
some degree necessitated a stepwise ap-
proach. Initial steps involved the establish-
ment of biology projects in the Freshman
and Sophomore Clinics and the creation of
the sophomore Biological Systems & Ap-
plications course. Subsequent changes in
the curriculum at the junior and senior levels
would not have been successful without the
prior addition of both content and experien-
tial knowledge at the lower levels.
Winter 2007

The future of chemical engineering is in nano- and bio-
technology. This curriculum, with its integrative biological
components, is at the front of future education.

Funding for the development and integration of the biomed-
ical and drug delivery modules was provided by grants from
the National Science Foundation, Division of Undergraduate
Education: DUE-CCLI 0088437 and DUE CCLI 0126902,
respectively. NSF REU EEC -( 1353744 provided additional
support for the development of drug delivery experiments.

American Institute of Chemical Engineers, 2001-2002 Initial Placement
of Chemical Engineering Graduates
2. Baum, R.M., "The Engineering Approach to Molecular Biology,"
Chem. and Eng. News, 76(13) (1998)
3. Breslow, R., "Into the Future," Chem. and Eng. News, 78(47) (2000)
4. Rawls, R.L., "Biochem Meets Engineering," Chem. and Eng. News,
77(35) (1999)
5. Westmoreland, PR., "Chemistry and Life Sciences in a New Vision of
Chemical Engineering," in Annual Meeting of the American Institute
of Chemical Engineers, Los Angeles (2000)
6. Oberholz, A., "Chemicals in 2010-Systems Solutions for the Cus-
tomer, " CHISA Conference, 1492 (2004)
7. AIChE Annual Meeting, San Francisco (2003)
8. U.S. Department of Labor, Bureau of Labor Statistics, i ...i... ,1 ,,.1
Outlook: 2000- 2010, "Monthly Labor Review (2000)
9. < ii)l. I.- i. .. ... .... ..- i l1/bestjobs06.html>
10. Lauffenburger, D.A., "A Course in Cellular Bioengineering," Chem.
Eng. Ed., 23(4) (1989)
11. Oerther, D.B., "Introducing Molecular Biology to Environmental

2000 2001 2002 2003 2004 2005

Figure 2. Percentage of students pursuing graduate degrees.28'

2 25.00%
. 20.00%

. 15.00%

| 10.00%


Engineers Through Development of a New Course,"( i..... i .Ed.,
36(4) (2002)
12. Hollar, K.A., S. Farrell, G. Hecht, and P Mosto, "Integrating Biology
and Chemical Engineering at the Lower Levels," Chem. Eng. Ed.,
38(2) (2004)
13. Farrell, S., "A Laboratory Project to Design and Implement a Process
for the Production of Beer, " Proceedings of the American Society of
Engineering Education Conf. (1999)
14. Jahan, K., "WaterTreatment in Reverse," Proceedings of theAmerican
Society of Engineering Education Conference (1999)
15. Farrell, S., R.P Hesketh, and M.J. Savelski, "A Respiration Experiment
to Introduce Chemical Engineering Principles,"( ..... in i . 38(3)
16. Farrell, S., and R.P Hesketh, "An Introduction to Drug Delivery for
Chemical Engineers," Chem. Eng. Ed., 36(3) (2002)
17. Farrell, S., J.A. Newell, and M.J. Savelski, "Introducing Chemical
Engineering Students to Product Design through the Investigation of
Commercial Beer," Chem. Eng. Ed., 36(2) (2002)
18. Farrell, S., R.P Hesketh, J.A. Newell, and C.S. Slater, "Introducing
Freshmen to Reverse Process Engineering and Design through Inves-
tigation of the Brewing Process," I.J.E.E. 17(6) (2001)
19. Maynard, J., and A. Razatos, "The Evolution of Engineering: Incorpo-
rating Biology into Traditional Engineering Curriculum, "Proceedings
of the ASEE Annual Conference, Session 2313 (1999)
20. McDonald, D., A. Mahajan, and M.E. Walworth, NSF EHR 9751372
21. McDonald, D., K. Schmaltz, M. Walworth, andA. Mahajan, "The De-
velopment of an Innovative Undergraduate Laboratory that Emphasizes
Vertical Integration in Multiple Engineering Curricula," Proceedings

of the ASEE Annual Conference, Session 2526 (1999)
22. Mahajan, A., M. Walworth, D. McDonald, and K. Schmaltz, "The
Integrated Systems Engineering Laboratory-An InnovativeApproach
to Vertical Integration using Modern Instrumentation, "Proceedings of
the ASEE Annual Conference, Session 2259 (1999)
23. Hecht, G.B., P Mosto, and C.S. Slater, "Effectiveness of an Applied
Microbiology Course Specifically Designed for Chemical Engineering
Majors, "Microbiology Education (2002)
24. Kobayaski, M., T. Kakizono, N. Nishio, and S. Nagai. "Effect of Light
Intensity and Light Quality onAstaxanthin Formation in a GreenAlage,
Haematococcus pluvialis, "J. Fermentation and Bioengineering, 74(1)
25. "First Leadership Awards Made: Hopkins and UCSD get $30 Million
Total," The Whitaker Foundation, Biomedical Engineering News
26. Farrell, S., R.P Hesketh, M.J. Savelski, and C.S. Slater, "Fundamentals,
Design and Applications of Drug Delivery Systems," ASEE Annual
Conference, Session 1313 (2003)
27. Dorland, D., and P. Mosto, 'The Engineering Clinics at Rowan Univer-
sity: A Unique Experience, "Proceeding of the International Congress
of Chemical and Processing Engineering (2006)
28. Marchese, A., J. Schmalzel, K.M. Weaver, "Creating an Entrepreneurial
Culture at a Startup Engineering Program, "Proceedings of the Ameri-
can Society of Engineering Education Conference (2004)
29. Sukumaran, B., K. Jahan, D. Dorland, J. Everett, J. Kadlowec, Z.
Gephardt, and S. Chin "Engineering Clinics: An Integration of Re-
search into the Undergraduate Engineering Curriculum, " Council on
Undergraduate Research Quarterly, (3) (2006)
30. 1

Chemical Engineering Education

Random Thoughts...



Education Designs, Inc.
North Carolina State University

If you're like most faculty members, you began your
academic career knowing very little about what you'd be
doing for a living. You knew about working on a research
project someone else had defined and gotten funded, but not
about starting and managing a research program, planning and
delivering courses, and dealing with the hundreds of technical
and management problems that always crop up in research
and teaching. No one told you much about those things after
you showed up either, so you had to figure it all out yourself
by trial-and-error.
This bizarre approach to career development has unfortu-
nate consequences. Roughly 95% of new faculty members
take an average of four to five years to meet or exceed their
institution's expectations for research and teaching.1, 2] The
remaining 5%, however-the ones Robert Boice1ll calls
"quick starters"-manage to do it in their first two years.
Considering the enormous investment institutions make in
each faculty member they hire, moving more of the new
ones into the quick starter category would clearly be good for
everyone - the new faculty, their institutions, and the students
they will teach and mentor.
Converting new faculty members into quick starters is not
impossible-it's not even difficult. You just give them early
guidance on how to teach well, do good research, and balance
the competing demands of teaching, research, service, and
personal life, and supplement this orientation with one-on-one
mentoring by skilled senior colleagues.
A program containing those elements has been in place
since 2000 in the N.C. State University College of Engineer-
ing. We offer it as an example of what can be done-and in
our opinion, what should be done-to help new engineering
faculty make the transition to their new careers quickly and
successfully. In this column, we briefly outline the program
(Brent, et al.[3] provide more details) and summarize the les-
sons we have learned from our experience with it.
Winter 2007

The centerpiece of the NCSU program is a four-day orien-
tation workshop held in mid-August. It covers grantsmanship,
recruiting and working with graduate students, designing
courses and getting them off to a good start, effective lecturing
and active learning, advising, time management, and dealing
with a variety of crises faculty members commonly encounter.
All presentations are highly interactive, and the presenters in-
clude some of the best teachers and researchers on the faculty
as well as key administrators and support staff. The workshop
was first given in 2000 to new engineering faculty, and since
2001 it has been given jointly to new faculty in the Colleges
of Engineering and Physical and Mathematical Sciences.

Richard M. Felder is Hoechst Celanese
Professor Emeritus of Chemical Engineering
at North Carolina State University. He is co-
author of Elementary Principles of Chemical
Processes (Wiley, 2005) and numerous
articles on chemical process engineering
and engineering and science education,
and regularly presents workshops on ef-
fective college teaching at campuses and
conferences around the world. Many of his
publications can be seen at edu/felder-public>.
SRebecca Brent is an education consultant
specializing in faculty development for ef-
ffective university teaching, classroom and
computer-based simulations in teacher
education, and K-12 staff development in
language arts and classroom management.
She codirects the ASEE National Effective
Teaching Institute and has published articles
on a variety of topics including writing in un-
dergraduate courses, cooperative learning,
Public school reform, and effective university

� Copyright ChE Division of ASEE 2007

The orientation workshop is followed by a series of hour-
long sessions during the academic year that reinforce work-
shop material and help maintain a sense of community among
the participants. Topics addressed include troubleshooting
teaching, dealing with funding agencies, and writing effective
proposals for CAREER Awards. (Workshop alumni have an
excellent record of landing them.) Another component of the
support program is mentoring. In 2000, all departments iden-
tified specific ways the department heads and senior faculty
would provide support to their new hires, and formal mentor-
ing programs have been initiated in several departments.[3]
The response of the new faculty has been overwhelmingly
positive. The participants to date have given the orienta-
tion workshop 99 overall ratings of "excellent," 12 "good,"
and no "average," "fair," or "poor" ratings. Past workshop
participants have given significantly higher ratings than
nonparticipants to their career orientations, and preliminary
assessments indicate that they have outperformed the non-
participants in terms of both funded research activity and
teaching evaluations. The program has maintained a high level
of administrative support and has become a strong selling
point for recruiting new faculty.

We have the following suggestions for schools planning
their own new-faculty support programs.
* Keep the program at the school/college level rather than
making it campus-wide.
Many universities have teaching centers that provide new
faculty orientation, but since the organizers have to address
faculty in all disciplines, they generally limit the program con-
tent to such things as campus resources and employee benefits.
As important as those topics may be, such programs don't do
much to convert new faculty into quick starters. When orien-
tation is designed specifically for faculty in engineering and
related disciplines, presenters can use research and teaching
examples that are clearly relevant to the participants-and
the greater the perceived relevance of presented material, the
greater its likely impact on the recipients.
* Get strong and visible support from the dean and depart-
ment heads.
If the director of a teaching center or the associate dean for
academics invites new faculty members to attend a four-day
workshop two weeks before the start of their first semester,
few are likely to show up, while if the dean and department
heads strongly encourage attendance and share positive evalu-
ations from past workshop participants, most new faculty
will attend.

* Provide guidance on both research and teaching and
discuss how to balance them.
Most new faculty are nervous about meeting expectations
for research productivity. Providing guidance on how to do
it is an excellent way to persuade them that the workshop is
worth their time. Presenters should also emphasize strate-
gies for making teaching efficient as well as effective and
for maintaining a balance of teaching, research, service, and
personal life consistent with the institution's expectations and
the faculty members' health and sanity.
* Keep the presentations practical and interactive.
A workshop that is mainly a parade of talking heads is
generally not worth the time it takes to prepare and present
it. If a designated presenter doesn't know how to design and
deliver an effective interactive presentation, someone else
who does should provide some coaching.
* Treat the participants well.
The new faculty should feel welcomed into the academic
community, and treating them well is one way to make that
happen. Hold the workshop in a convenient, comfortable
location and don't skimp on the budget for meals and breaks.
Provide useful resources in a well-organized notebook. Post
lists of good local restaurants, parks and playgrounds, cultural
attractions, and automobile repair shops. End the workshop
with a celebratory reception and invite all the department
heads and mentors to attend and interact with the participants.
Make sure mentoring in teaching and research is provided by
skilled and supportive colleagues who know something about
how to mentor.[4]
In summary, if the goal is to convert new faculty members
into quick starters-productive in research and effective in
teaching in their first two years-and the orientation that most
of us got (i.e., none) is all that's provided, there is a one-in-
twenty chance of succeeding. The strategies we've proposed
should improve the odds considerably.

1. R. Boice, Advice for New Faculty Members, Needham Heights, MA:
Allyn & Bacon (2000)
2. R.M. Felder and R. Brent, "The New Faculty Member,"( ... I .
Education, 32(3), 206-207 (1998), public/Columns/Boice.html>
3. R. Brent, R.M. Felder, and S.A. Rajala, "Preparing New Faculty
Members to be Successful: A No-Brainer and Yet a Radical Concept,"
Proceedings of the 2006 Annual ASEE Conference, Washington,
DC: ASEE (2006), ASEEO6(NewFaculty).pdf>
4. R.M. Felder, 'Teaching Teachers to Teach: The Case for Mentoring,
Chem. Engr. Education, 27(3), 176-177 (1993), edulfelder-public/(-..... I, ...,,, ,, , .... ,, > [

Chemical Engineering Education

All of the Random Thoughts columns are now available on the World Wide Web at
http://www.ncsu.edu/effective_teaching and at http://che.ufl.edu/~cee/

--- - ^ K.___________________________-



into Chemical Engineering Education

Texas Tech University * Lubbock, TX 79409
Six Sigma is a buzz term in today's technology and
business worlds. In organizations like Motorola, GE,
DuPont, 3M, IBM, Dow Chemical, and PPG, Six
Sigma means a measure of quality that strives for perfection.El
Statistically, it means reducing the process variation so that
+six standard deviations lie between the mean and the nearest
specification limit. Under the Six Sigma control, the defect
probability is 3.4 per million.1, 2] Six Sigma methodology has
been successfully applied to manufacturing (especially chemi-
cal and related manufacturing), to research and development,
and to business and financial services.
Six Sigma methodology combines elements from several
quality movements with advanced statistical methodology. It
is a comprehensive tool combining business concepts with
technical and leadership skills, and thus it is suitable for pro-
fessionals at all levels: managers, engineers, and scientists.
Recently, there has been great interest in initiating Six Sigma
training in college education. This paper reports the success
of incorporating Six Sigma methodology into a traditional
chemical engineering course, Engineering Experimentation,
at Texas Tech University.

CHE 3343/4372, Engineering Experimentation, is a tradi-
tional undergraduate elective course in the chemical engineer-
ing curriculum at Texas Tech University. The original catalog
listing is "strategy in experimentation; planning efficient
experiments; analysis of data, interpretation, and presenta-
tion results." The course provided an excellent opportunity to
incorporate Six Sigma methodology training into traditional
engineering education. In practice, the instructor starts the
course with an introduction of the fundamentals of Six Sigma
methodology, emphasizing the D.M.A.I.C. process that refers

� Copyright ChE Division of ASEE 2007

Winter 2007

Lenore L. Dai is an assistant professor
of chemical engineering at Texas Tech
University. She received her B.S. from
Beijing University of Chemical Technology
and her Ph.D. from the University of Illinois.
Her research fields of interest include
solid-stabilized emulsions, self-assembly
of solid particulates, dynamics of solid
particles at liquid-liquid interfaces, and
polymers and composites.

to Define (D), Measure (MI), Analyze (A), Improve (I), and
Control (C).E31 The course is then organized to discuss various
methodologies and tools in each process stage. For example,
moving range chart/individual chart and range chart/X-bar
chart are heavily emphasized to evaluate measurement sys-
tems in the process stage of Measure (M). The tools in the
process stage of Analyze (A) overlap with various classical
topics in Engineering Experimentation, including: design of
experiments (DOE) and analysis (focused on two-level full,
half, and highly fractionated factorial designs and analyses),
residual and model adequacy analyses, regression model, and
confidence levels. Apart from technical content, it is worth-
while to note that a small fraction of Six Sigma management
and business concepts are also addressed, mainly in the Define
stage. For example, we have discussed a S.M.A.R.T. goal
(refers to a goal that is specific, measurable, agreed upon,
realistic, and time bounded), thought map (a road map that
is composed of different paths of questions), Six Sigma team
development, and effective meeting management. A summary
of the different topics discussed in each stage of the course is
shown in Figure 1. As a more specific example, the instructor
included a request to write a S.M.A.R.T. goal for this course
in the first homework assignment. The students, working in
a group format, answer questions about the course includ-
ing: what is to be accomplished (S, specific), what level of
improvement is needed (M, measurable), what do we agree
upon as a team (A, agreed upon), whether the goal can be
accomplished by the given available resources (R, realistic),
and what the expected dates for major milestones are (T,
time bounded). The full D.M.A.I.C. process is then practiced
through a formal Catapult Project, discussed later, accounting
for 15% of the final grade.

Six Sigma D. M. A. I. C. Process

Six Sigma Tools Discussed in Ch E 334314372
[ Defin Six Sigma Concept, T I M E Problem Statement, S M A R T Goal,
Thought Map, Process Map, Six Sigma Team and Management, Creativit

Run Chart, Moving Range Chart/Individual Chart, Range Chart, Range
Mea ChartX-bar Chart, Common Cause/Special Cause Model, Measurement
System Evaluation, Factor Relationship Diagram, Dot-frequency Diagram
Engineering Experimentation Method
Analyze Full Factorial Design and Analysis, Half Factorial Design and Analysis,
Highly Fractioned Design and Analysis, Residual and Model Adequacy,
J Regression Models, Normal Probability Plot, Other Miscellaneous
Statistical Concepts
S Practice in the Catapult Project and some special homework assignment


Figure 1. The Six Sigma D.M.A.I.C. process and
different tools discussed in CHE 3343/4372.

The homework assignments in CHE 3343/4372 include
the problems in the textbook Design and Analysis of Experi-
ments [4 and special problems generated by the instructor. For
example, the instructor provided raw data of several projects
in CHE 4232, Unit Operations Laboratory (permitted by the
class and the instructor Professor T. Wiesner), and requested
students perform new analyses using the tools learned in
CHE 3343/4372. Such assignments give students opportuni-
ties to work on practical problems related to other chemical
engineering subjects and, more importantly, allow them to
practice the Six Sigma methodology by solving practical
chemical engineering problems. In addition, the instructor
typically has several nontraditional homework assignments,
such as a card-drop exercise related to variation and creativity,
a paper airplane mini-project using a 22 full-factorial design to
study the influence of airplane weight and launching angle on
landing distance, and another card-drop exercise to conduct a
23 full-factorial design to study the influence of card weight,
surface area, and releasing height on target landing.
"Tell me, I'll forget; show me, I'll remember; involve me,
I will understand.""' 6] Without doubt, designing and practic-
ing are the heart of engineering majors. This is an important
element in CHE 3343/4372, Engineering Experimentation.
A formal Catapult Project assignment, which includes an
individual project report, a group presentation, and a group
competition, has been assigned for the last four successive
years and counts 15% toward the total grade. Catapults are
used by more than 200 companies as a training aid in Six
Sigma methodology training. A snapshot of the catapult used
in CHE 3343/4372 is shown in Figure 2. The project includes
four major elements. First, the students are assigned to
work in project teams (three to four students per team)
to investigate the performance of their catapults includ-
ing evaluating the measurement system and performing
factorial experiments to determine the major influencing
factorss. Second, each student works independently to
y analyze the collected raw experimental data and submit
a formal individual project report. Third, the project
team regathers and finalizes the developed model for
performance prediction and makes a formal project pre-
sentation to the entire class. Lastly, the team will use its
developed model for a project competition. During the
project competition, the instructor will place the target
at a random location within a defined target area and
each team needs to launch the ball within three minutes
with the goal of hitting the target. Figure 3a shows a
brief map of the setup in the project competition and
Figure 3b is a snapshot of a ball approaching the target
in a 2004 class competition.
The Catapult Project has given the students a unique
opportunity to practice the Six Sigma D.M.A.I.C.
Chemical Engineering Education

Figure 2. A sample catapult.

process. In the Define stage, the students practice various
concepts taught in class such as defining a S.M.A.R.T. goal,
organizing a thought map, and managing a project team. Dur-
ing the stages of Measure and Analyze, the students evaluate
the measurement system and perform two-level full, half,
and/or highly fractionated factorial design experiments and
analyses to determine the major influencing factorss. In ad-
dition, they will develop a regression model�41 quantitatively
relating the distance as a function of setting parameters such
as launching angle, type of ball, rubber band position, and stop
pin position. An example of a regression model developed
from a 23 full factorial design is:
Distance = P( + (1(parameter 1) + P2(parameter 2) +
(3(parameter 3) + (12 (parameter Ixparameter 2) +
P13 (parameter Ixparameter 3) + (23 (parameter 2x
parameter 3) + P123 (parameter Ixparameter 2x
parameter 3) + error (1)
where P( is the average response from the design and (3, (3 ,
and P(jk are calculated from the main effects of single pa-
rameters, two-way interactions, and three-way interactions,
respectively. Eq. (1) is the regression model that involves
all parameters and interactions in a 23 full-factorial design.
For practicality, the students have choices of including only
significant factors. The model adequacy will be evaluated
by various residue analyses. Finally, the students move to
the Improve and Control stage to optimize and apply the
developed regression model. For example, during the project
competition, each project team will measure the distance
where the instructor randomly locates the target (within the
target area) and use the model to decide the settings for dif-
ferent parameters. The accuracy and robustness of the model
will directly determine whether the ball can hit the target or
how close the ball is landing to the target.
It is worthwhile to note the Catapult Project also gives stu-
dents an opportunity to integrate business decision making to
Winter 2007

engineering practice, as each team is allowed a maximum of
45 shots with no deduction of scores during the entire project.
Upon completing the project, the students practice applying
Six Sigma methodology to solve a real-life problem as well
as obtaining the experience of improving the performance of
the catapult while maintaining a profitable business.

Other than traditional classroom lectures, the course also
provides two or three training sections of the JMPIn statistical
software. The software is a statistical program that is widely
used in Six Sigma methodology training and at companies
such as Dow Chemical, Procter & Gamble, HP, and PPG.
The software allows students to solve complicated statistical
problems. For example, we have used the JMP In software
to generate a contour plot to view all the possible combina-
tions for desirable properties from the model developed in
the factorial design.

Another learning impact of CHE 3343/4372, Engineering
Experimentation, is on creativity. Most chemical engineering
education focuses on problem solving based on well-estab-
lished principles, placing less emphasis on creativity. Hueter
states that modern people's "creative abilities increase in ele-
mentary school up to eight years old and then steadily decrease
with further education, including college education."16 71 The
importance of creativity in engineering can be summarized
as follows: "Engineering is an art as well as a science, and
good engineering depends upon leaps of imagination as well
as painstaking care."1 8] Creativity is also heavily emphasized
in Six Sigma methodology.[9] The project, as well as a few

Figure 3. (a) A map illustrating the setup for the Catapult
Project competition; (b) a ball is approaching the target in
an actual competition in the class of 2004.

a. Launching

of the homework assignments (paper airplane competition,
card drop exercises, etc.), provide students opportunities not
only to practice the multidisciplinary methodology but also to
maximize their potential to be creative during the exercises.

The course is among the most popular electives in the
chemical engineering curriculum at Texas Tech University.
In the spring semesters of 2003-2006, the enrollment was 16,
26, 13, and 14, respectively. The course has received excel-
lent student evaluation, with an average rating of 4.9/5.0,
5.0/5.0, 5.0/5.0, and 5.0/5.0 out of the 16 university-level
questionnaires [scores rank from 1 (poor) to 5 (excellent)]
on the instructor and course. Multiple students have said this
class was their "favorite class" and the "best experience in a
college course." Specific comments related to the Six Sigma
training and work experience include:
* !./a ,, this class has given me confidence in my ability
to attack and solve problems at my new job this sum-
mer. "
* "I think this class was one of the most beneficial
courses that I have taken."
* "Really enjoyed this class being directly applicable to
my work today."
* "I'm glad that the department decided to give this
course, with industry changing year to year. This class
will be extremely useful when we go to work!"
* "Great course. It should be offered every year. It helped
me get my job."

We have successfully incorporated Six Sigma methodol-
ogy training into a traditional chemical engineering course,
CHE 3343/4372, Engineering Experimentation, at Texas

Tech University. The course is structured along the Six Sigma
D.M.A.I.C. process and different technical and nontechnical
tools have been discussed in each stage of the process. Some
of the nontraditional aspects in this course include industrial
need, special homework assignments, the Catapult Project,
the JMP In statistical software training, and emphasis on
creativity. In addition, students have also obtained hands-on
experience to practice Six Sigma methodology and a unique
and integrative experience to practice engineering and busi-
ness concepts simultaneously.

The author would like to thank Professor T. Wiesner for
his invaluable encouragement and discussion. In addition,
the author is grateful to the support from the Texas Tech Fac-
ulty Incentive Grant Award (2003) and the National Science
Foundation (CTS-0500323).

1. Stamatis, D.H., Six Sigma and Beyond, CRC Press LLC (2002)
2. Statistical Six Sigma Definition, content/c010101a.asp>, (2006)
3. Rath & Strong's Six Sigma Pocket Guide, Rath & Strong Management
Consultants, Lexington (2002)
4. Montgomery, D.C., Design andAnalysis ofExperiments, 5th Ed., John
Wiley & Sons, Inc. (2001)
5. Eastlake, C.N., "Tell Me, I Will Forget; Show Me, I'll Remember;
Involve Me, I'll Understand (The Tangible Benefit of Labs in the
Undergraduate Curriculum), "Proceedings ASEE Annual Conference,
Washington, (1986)
6. Hueter, J.M. "Innovation and Creativity: A Critical Linkage, "Proceed-
ings ASEE Annual Conference, Washington, 1634 (1990)
7. Wankat, PC., and E S. Oreovicz, Teaching Engineering, McGraw-Hill
8. Forman, S.C., The Civilized Engineer, St. Martin's Press, New York
9. Pyzdek, T., The SixSigmaHandbook-A Complete Guide for Greenbelts,
Blackbelts, and Managers at All Levels, McGraw-Hill (2001) 1

Chemical Engineering Education

book review
--- - ^ K.___________________________-

Process Dynamics and Control, 2nd Ed.
by Dale Seborg, Tom Edgar, and Duncan Mellichamp
Wiley (2003) $138.95

Reviewed by
Derrick K. Rollins, Sr.
Iowa State University of Science and Technology
First, I want to applaud the authors for making a substantial,
well-thought-out revision to their textbook. I have used the
book to teach my introductory process control course but had
not really read the additional material in the new chapters until
this review. I was very impressed with the depth and breadth
of the material. I am amazed that the authors were able to
eliminate so much material and yet not dilute the critical
topics that are important to a first course. I know that it was
a struggle to decide what to eliminate and what to keep, just
due to the fact that three personalities were involved. They
did an excellent job.
Since the authors did eliminate so much material, espe-
cially on digital control, I see the two versions being more
like Volume 1 and Volume 1.5 (not quite two volumes) and
working to complement each other in advanced courses in
process control. I see potentially three semester courses (at
least this is the way I would do it) from this text. The first
one is a general process control course for all undergraduate
chemical engineering students covering Chapters 1-9, 11-12,
and 15-16; a course in advanced methods covering Chapters
13-14, 18-21 (bringing in material from the first edition);
and an application toward plantwide control and plant design
covering Chapters 10, 22-24, and all the Appendices.
For those working in the process control field the book is
a good textbook as well as a good reference manual. Faculty
that are not, however, yet are teaching process control might
find the text intimidating and too complex. I asked a faculty
member in my department who fits this category and is cur-
rently using the text and that was his feeling. I kept this in mind
while reviewing the text and I could understand his feeling of
insecurity with it. One way the text could be improved is to
revisit the chapters I mentioned for an introductory course and
work to rewrite it in such a way that faculty in this category
could feel more comfortable with the material.
My comments on specific chapters are as follows. Chapter
1 needs more problems. I am disappointed that they took
the block diagram out of this chapter. I have used it to tell
students where we are going and why we need Chapters 2-7
and how each block represents certain chapters that we will
Winter 2007

tie back together in Chapter 11 (what used to be Chapter 10).
Chapter 2 is essentially the same as before just with some
new problems (I particularly like the additional application
on bioprocesses and the exercises). Chapter 3 is essentially
the same but the authors should have left Exercises 3.16 and
3.20 in this edition. These were two of my most popular
problems for homework. Chapter 4 also did not change much
but could be made shorter by giving a general method using
Section 4.3 material, which covers all cases. I like the addi-
tion of state-space formulation. Chapter 5 has been basically
untouched, which I applaud, but it does have more good
problems-something faculty always appreciate. Section
6.3.1 is a good addition to Chapter 6 and is explained well.
I have always appreciated this chapter and I am glad to see
it is even better. The problems are good, especially the ones
reflecting new material and bio-systems engineering. For
Chapter 7, I feel that all the emphasis on graphical methods
should be removed and replaced with regression. In illustrat-
ing regression techniques I think it is more important to show
how software packages would do this rather than to give the
mathematical equations on how they are done. Although they
give Matlab and Excel examples they do not show, step by
step, how this is exactly done. I think professors and students
would appreciate this detail.
Example 7.4 needs to be revised or removed. Who would
fit Models 3 and 4 to that response? In addition, a better ARX
or ARMAX model should give a better fit by the fact that
Model 1 fit so well.
Chapter 8 is basically the same but this chapter has always
needed, as it does now, more problems. Chapter 9 is done
well but needs more explanation of hardware and more
Other textbooks are much stronger in this material such as
Riggs (2001). I just skipped the material in Chapter 10 and
went straight to Chapter 11. It is good material but out of step
with how I do my course. It is important for the material on
plantwide control and design. I do not like the way the mate-
rial in the new Chapter 11 has combined Chapters 10 and 11
from the first edition. I like to keep stability analysis separate.
There are, however, plenty of good exercises in this chapter.
Chapter 12 is done well and has excellent problems. Chapters
13 and 14 are a good condensation of the three chapters on
frequency response from the first edition. This material needs
to remain but not be overemphasized, in my opinion. Chapter
15 is an excellent chapter with good problems.
For Chapter 16, the addition of Fuzzy Logic Control is an
excellent improvement but I couldn't find any exercises on
this topic. In Chapter 17 I am glad that they left the mate-
rial on filtering in this edition. I know that it was difficult to
remove much of the material on z-transforms and sample
data-control systems, but the first addition could supplement

@ Copyright ChE Division of ASEE 200;

these eliminations if necessary. It would help to actually have
an example for obtaining the poles and zeros. Also, what hap-
pened to COUz in Eq. 17-46? The Co is there but where is U ?
I appreciate the addition of Section 17.6 and I am glad they
did not go into a lot of detail [it would be hard to match the
material of Ogunnaike and Ray (1994) on the topic]. There
are a lot of good problems in the exercises.
I did not find many errors or typos in this edition, which is
commendable considering the amount of new material and re-
organization. In Chapter 18, however, "n!" on page 477 should
be "n." Also, on page 479, "hidden" is mistakenly printed as,
"hidd en." On page 492, just below Eq. 18-58, "4" should be
"w." I commend the authors for adding the SVD material and
updating this chapter. It may be the most important chapter
for the control design engineer in terms of theory.
I am glad that they shortened the material in Chapter 19
since optimization is a course in itself and only an overview
is critical to any process control course. Chapter 20 is a sub-
stantial and critical improvement over the first edition. All the
basic fundamentals and concepts of model predictive control
(MPC) appear to be present. At least it gives a good overview

and introduction on the subject. Although I have not taught
from this chapter yet, the exercises appear to be excellent.
The authors did an excellent job on Chapter 21. They did
it just right and the critical material is here in just the right
amount. These topics include the following: X bar chart, S
chart, Cusum, EWMA, Cpk, Six Sigma, and multivariate
MPC. They need, however, to point out which ones detect a
"mean shift" vs. a "variance shift." For example, there is no
statement in this regard for the X bar chart. I suggest that they
add the use of Minitab in this chapter as they did Mathlab and
Simulink for chapters exploiting their use. Finally, Chapters
22-24 appear to be done quite well and I look forward to using
them in future courses.
Overall, the authors' have made a timely and significant
improvement to this textbook by bringing it up to date
with current practices and needs, and enhancing its use as
a textbook in process control for undergraduate as well as
graduate students. I have used the earlier book since 1991,
and with the improvements they have made in the second
edition, this text will be useful in the courses I teach for
many years to come. 7

Chemical Engineering Education

M]!1n class and home problems



With Mathematica in the Undergraduate Curriculum

National Institute of Applied Sciences and Technology
Anon-Newtonian fluid has a viscosity that changes with
the applied shear force. These fluids are characterized
by measuring or computing several theological prop-
erties, such as the viscosity and the first and second normal
stresses. Rheometers are used, under oscillatory shear flow
or extensional flow, to obtain experimental values of these
theological properties while kinetic theory calculations using
dumbbells allow the prediction of these theological properties.
For a Newtonian fluid (such as water), the viscosity is indepen-
dent of how fast you are stirring it. For a non-Newtonian fluid
the viscosity is dependent. It gets easier or harder to stir faster
for different types of non-Newtonian fluids. By adding corn
starch to water, one obtains a non-Newtonian fluid. Applying
agitation with a spoon makes the fluid behave like a solid.
Thus, the shear-thickening property of this non-Newtonian
fluid becomes apparent. When agitation is stopped and the
fluid is allowed to rest for a certain period of time, it recovers
its liquid-like behavior.
Non-Newtonian fluids display many peculiar phenomena
that can serve as the basis for multiple "fun" experiments
students can perform in the laboratory. These include dye
Winter 2007

*Tunis, Tunisia
swelling, rod climbing, and suspensions of particles behavior
while moving in non-Newtonian vs. Newtonian fluids. Stu-
dents can determine the terminal fall velocity and rotation
direction of a single settling particle as well as wall effects
and interaction between particles. Problems involving non-
Newtonian fluid flow are ubiquitous in modern industry, such
as in polymer processing plants. The study of body fluids such
as blood, which is non-Newtonian, has important applications
in biomedical engineering. In the present paper, we show
how one can use the mathematical software Mathematica to

Housam Binous is a full-time faculty member
at the National Institute of Applied Sciences
and Technology in Tunis. He earned a Dip-
lome d'ingenieur in biotechnology from the
Ecole des Mines de Paris and a Ph.D. in
chemical engineering from the University
of California at Davis. His research interests
include the application of computers in
chemical engineering.

� Copyright ChE Division of ASEE 2007

The object of this column is to enhance our readers' collections of interesting and novel prob-
lems in chemical engineering. Problems that can be used to motivate the student by presenting
a particular principle in class, 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 elucidate difficult concepts.
Manuscripts should not exceed 14 double-spaced pages and should be accompanied by the origi-
nals of any figures or photographs. Please submit them to Professor James O. Wilkes (e-mail:
wilkes@umich.edu), Chemical Engineering Department, University of Michigan, Ann Arbor,
MI 48109-2136.

solve some simple non-Newtonian fluid problems. The most
relevant Mathematica commands'11 are inserted in the text
and can be found in any introductory book such as Math-
ematica, A System for Doing Mathematics by Computer by
Stephen Wolfram.[2] We start by reminding the reader of the
few simple constitutive equations for the power-law, Carreau,
and Bingham fluids. Then, we give the velocity profile for
the horizontal flow of power-law and Carreau fluids in a pipe
and an annulus. The velocity profile for the fall of a Bingham
liquid film is obtained in the next section. We also derive
volumetric flow rate expressions for pipe flow of Bingham
and power-law fluids. In the last part of the paper, we make a
model determination using previously found volumetric flow
rate expressions and representative data.

For Newtonian fluids, the shear stress, T, is proportional to
the strain rate, T
S= rl (1)
where the viscosity, l, the proportionality factor, is constant.
The situation is different for non-Newtonian fluids, and the
viscosity is a function of the strain rate:
T =q 0 ) (2)
Different constitutive equations, giving rise to various models
of non-Newtonian fluids, have been proposed in order to ex-
press the viscosity as function of the strain rate. Inpower-law
fluids, the following relation is satisfied:

TI = r.-1 (3)
Dilatant fluids correspond to the case where the exponent
in Eq. (3) is positive (n > 1) while pseudo-plastic fluids are
obtained when n < 1. We see that viscosity decreases with
strain rate for n < 1, which is the case for pseudo-plastic
fluids, also called shear-thinning fluids. On the other hand,
dilatant fluids are shear-thickening. If n = 1, one recovers the
Newtonian fluid behavior.
The Carreau model describes fluids for which the viscosity
presents a plateau at low and high shear rates separated by a
shear-thinning region:
1I- i 1
S1 n2 (4)

where 0o is the zero-shear viscosity and 1_ is the infinite-
shear viscosity.
Finally, the Bingham model is defined as follows:

Atlow shear rates :

Athigh shear rates:

1 2
-(T:T)< T , =0

1 2
-(T: T)> To,

T= tI'q+1 (6)

Problem Statement
Find the velocity profiles for the laminar flow of power-
law and Carreau fluids in a pipe, shown in Figure 1. Use the
following values for the pressure difference AP, the exponent
n, the Newtonian fluid viscosity 1, the consistency index K,
the infinite-shear viscosity ] , the zero-shear viscosity %,
the relaxation parameter X, the pipe length L, and radius R,
whose units appear under "Nomenclature" at the end of this
AP =100; L = 50; and R = 0.02
Newtonian fluid: T = 8.9 X 104.
Dilatant fluid: n = 3.39 and K = 106.
Pseudo-plastic fluid: n = 0.4 and K = 5 X 103.
Carreau fluid: n= 0.5, X= 0.2, 0o= 1.72 X 103, and Tl= 0.
This problem is treated using Polymath, a numerical com-
putational package,[31 in Problem Solving in Chemical Engi-
neering with Numerical Methods by Cutlip and Shacham.[4]
The governing equation is the z-component of the equation
of motion in cylindrical coordinates:
Id _dvz P
d � dv (7)
r dr dr L

Eq. (7) is subject to the following split boundary condi-

At r = 0:
At r= R:

z =0
v= 0

These kinds of mathematical problems often require the use of
a particular numerical approach called the shooting technique.
This method consists of guessing different values of v. at r = 0,
solving the differential equation, and checking that the no-
slip boundary condition at r = R is satisfied. An analytical
solution is possible for power-law fluids and details about
its derivation can be found in Fluid Mechanicsfor Chemical
Engineers by Wilkes:51]
AP Iln (R1+/n -r l/in
z (r)= 2 , (10)


AP=P - P2=100



Figure 1. Flow of Carreau and power-law fluids
in a pipe.
Chemical Engineering Education

For the Carreau fluid, one must use a numerical approach
since no analytical solution is available.
For the power-law fluids, the following Mathematica com-
mands are used to find the velocity:
system[QJ = { D[r Tj[r],{r,1}] == AP/Lr,
D[v[r],{r,1}] == If[ Trz[r]0, -( T[r]/K )A(1/n),
(-T_[r]/K)A(1/n) ], Tr[10^-5] == 0, v[10^-5] == Q };
myODEsoln[Q_] := NDSolve[system[Q], { vz, TJ, {r,
10^-5, R}]
yend[Q_?NumericQ] := Flatten[(v,[r] / myODEsoln[t])
/. rR]
bc = FindRoot[yend[Q] == 0, {I, 0, 0.5} ][[1,2]];

The graphical capability of Mathematica allows the student
to plot the velocity profile without having to use different
software. Figure 2 shows the velocity profile for the New-
tonian, dilatant, Carreau, and pseudo-plastic cases using the



0.15 Carreau

0.1 Pseudo-plastic


0 0.005 0.01 0.015 0.02
Figure 2. Velocity profiles of dilatant, pseudo-plastic,
Carreau, and Newtonian fluids in a pipe.

AP=P, - P2=100

pltl=Plot[v[r] /. soil, {r, 0.00001, R}, PlotStyle - RGB-
Color[0, 0, 1]]

These profiles are obtained under equal volumetric flow
conditions. The velocity near the wall is higher for Carreau
and pseudo-plastic fluids than for Newtonian and dilatant
fluids. This results in higher heat transfer rates due to a higher
convection. The approach to solve split boundary problems
using Mathematica is more systematic than the one proposed
by Cutlip and Shacham[4] using Polymath, despite a steeper
initial learning curve for students. In fact, it automatically
finds the velocity at the center of the pipe by verifying the
no-slip boundary condition and using the Mathematica com-
mand FindRoot.

Problem Statement
Find the velocity profiles for the laminar flow of power-law
and Carreau fluids in an annulus, shown in Figure 3. Use the
following values, where R and R, are the inner and outer radii,
and all other symbols have already been defined:
AP= 100; L= 50; R,=0.02 and R =0.05
Newtonian fluid: T = 8.9 X 104.
Dilatant fluid: n = 1.2 and K = 4.7 X 104.
Pseudo-plastic fluid: n = 0.5 and K = 4.5 X 103.
Carreau fluid: n = 0.5, X = 0.2, 0o= 2.04 X 10 and T,= 0.
Cutlip and Shacham[4] have solved this example using
Polymath. The governing equation is again the z-component
of the equation of motion in cylindrical coordinates:

1 d dv A
!rdr -dr ) - (11)
r dr dr L

Eq. (11) is subject to the following split boundary condi-

At r = R:
At r = R,

v =0
v =0

To solve this problem, we make use of the shooting technique
in a similar fashion as the previous example. This method
works by guessing different values of Tz at r = R , solving the
differential equation, and checking that the no-slip boundary
condition at r = R, is satisfied. An analytical solution[41 is
available for the Newtonian fluid case:

vz(r)= ['PR2
4 4IL

r 2nR22- R ln(r /R2) (14)
ln(R, / R1)

There is no analytical solution for dilatant, pseudo-plastic, and
Carreau fluids, so one must resort to a numerical method.

Figure 3. Flow of Carreau and power-law fluids
in an annulus.
Winter 2007

For the power-law fluids, the following Mathematica com-
mand is used to find the velocity as a function of r:
system[Q] = { D[r T [r],{r,l}] == AP/L r,
D[v[r],{r,1}] == If[ T[r]>0, -(Tr[r]/K )A(l/n),
(-T_[r]/K)A(l/n)], T_[R1] == Q, z[R1] == 0 };
myODEsoln[Q_] := NDSolve[system[Q], {vz T), {r, R1,
yend[Q_?NumericQ] := Flatten[(v[r] /. myODEsoln[Q])
/.r R2]
bc = FindRoot[yend[Q] == 0, {0, -2, 2} ][[1,2]];

One can plot the velocity profile, shown in Figure 4, for the
Newtonian, dilatant, Carreau, and pseudo-plastic cases using
the Mathematica commands:
pltl=Plot[vx[r] /. soil, {r, 0.00001, R}, PlotStyle - RGB-
Color[0, 0, 1]]

These profiles are obtained under equal volumetric flow
conditions. The velocity profiles found for all four fluids are
not symmetric. In fact, they reach a maximum value close
to the radial position, given by r = 0.033, slightly less than
halfway from R and R2.

Problem Statement
Find the velocity profile for the vertical laminar flow of a
Bingham fluid down the wall depicted in Figure 5. Values
of the gravitational acceleration, g, the density, p, the yield
stress, T0, the zero-shear viscosity, 10, the film thickness, b,
are given by:
g=9.81; p=950; T =5; 10 =0.15 and 8=0.005

Cutlip and Shacham 41 have presented a solution of this
example using Polymath. The governing equation is the
z-component of the equation of motion in rectangular co-

d-= pg (15)
Eq. (15) is subject to the following split boundary condi-

At x= 0:

Tx =0

At x = 6: v = 0 (17)
We make the same treatment as the first two problems by
applying the shooting technique:
system[Q_] = {D[Tj[x],{x,l}] == Q g,
D[v[x],{x,l}] == If[Abs[T][x]] If[Tj[x] > To, (To-Tj[x] )/o -(To + Tj[x] ) / No ]]
J[O] == 0, v[O0] == Q };
myODEsoln[QJ := NDSolve[system[Q], {v , TZ}, {x, 0,
yend[Q_?NumericQ] := Flatten[(v[r] /. myODEsoln[b])
/. r 6]
bc = FindRoot[yend[Q] == 0, {J, 0, 0.5} ][[1,2]];

For the Newtonian case, an analytical expression for the veloc-
ity, vz, as a function of position, x, can be easily derived:

v= ' 1- xI2 (18)

In Figure 6, we show the velocity profile for the Newtonian
and the Bingham fluids. This plot is obtained by using the
Mathematica commands:

0. / 'irrt.iu\

0.F P.tuiliii-llic t

0. 1


0.02 0.025 0.03 0.035 0.04 0.045 0.05
Figure 4. Velocity profiles of dilatant,
pseudo-plastic, Carreau, and Newtonian
fluids in an annulus.

Figure 5.
Vertical flow
of a Bingham
fluid in a
liquid film.


0.2 Newtonian

0 0.001 0.002 0.003 0.004 0.005
Figure 6. Velocity profiles of Bingham and Newtonian
fluids in a liquid film.
Chemical Engineering Education

pltl=Plot[v[x] /. soil, {x, 0, 6}, PlotStyle - RGBColor[0,
0, 1]]

A comparison of the velocity profile obtained using the ana-
lytical solution for the Newtonian fluid and the velocity profile
corresponding to the Bingham fluid shows that the latter is flat
near the surface of the liquid film. In fact, we have a nonzero
velocity gradient only when Txz> T. This behavior is typical
of Bingham fluids.

Problem Statement
Derive expressions of volumetric flow rates for pipe flow of
Bingham and power-law fluids using symbolic computations
with Mathematica.
Power-law fluid case
First, we find the expression of the shear stress, T, as a func-
tion of the radial position, r:
sol3 = DSolve[D[r T [r],{r,l}] == - AP/L r, T [r], r]
T[r] = soll[[l, 1, 2]]/.C[1] - 0

We get the following result:
Trz A- Pr (19)
Then, we determine the velocity distribution using the sym-
bolic command, Dsolve,
sol4 = DSolve[D[vz[r],{r,1}] == -( -rz[r]/K )A(l/n), [r], r]

2 -1/"n nR ( )PR
v[r] = sol4[[1, 1, 2]]/. C[l] xL
Finally, the symbolic command, Integrate, is used,
Q = Integrate[2 Pi r v[r], {r, 0, R}]

Volumetric Flow Rate vs. Pressure Gradient

AP/L (Pa/m) 105 X Q (m3/s)
10000 5.37
20000 26.4
30000 68.9
40000 129
50000 235
60000 336
70000 487
80000 713
90000 912
100000 1100
Winter 2007

and we get the following expression for the volumetric flow

2 n R RAP /n

Q=L (20)
1 +3n
Bingham fluid case
Just like the treatment above, we start by finding the ex-
pression of the shear stress, Trz, as a function of the radial
position, r:
soil = DSolve[D[r Tj[r],{r,l}] == - AP/L r, Tr[r], r]
Tr[r] = soll[[l, 1, 2]]/.C[1] - 0
We get the following result:
, Pr (19)
In the first part of the derivation, we determine the velocity
distribution between r = (2T0L)/AP and r = R using boundary
condition vz(R) = 0 and the symbolic command, Dsolve:
sol2 = DSolve[D[v[r],{r,l}] == ( Tz[r]+T0)/qi, z[r],r]
vz[r] = sol2[[1, 1, 2]] /. C[l] i ((AP R2)/(4 q L))
- (RT0)/qi

The symbolic command, Integrate, is used to obtain the ex-
pression of the volumetric flow rate between r = (2ToL)/AP
and r = R,
Q1 = Integrate[2 Pi r v[r], {r, -2 T0 L/AP, R}]

In the second part of the derivation, we determine the con-
stant velocity, v0, between r = 0 and r = (2T L)/AP using the
following symbolic command:
v = 1/4t (-AP/L) (r2-R^2) /4 + To /i (r-R) /. r 2 T0 AP

This is nothing more than expressing the continuity of the
velocity at r = (2T0L)/AP. In fact, we have written that vo=
Vz((2T0L)/AP) in the above Mathematica statement.
The symbolic command, Integrate, is used to obtain the
expression of the volumetric flow rate between r= 0 and r =
Q2 = Integrate[2 Pi r v0, {r, 0, 2ToL/AP}]

and we get the following expression for the overall volumetric
flow rate,

7R4 AP 7R3T0 2Tr0L3
Q= p3 (21)
8vlL 3iq 3 AP3

Problem Statement
Wilkes5s] provides representative values of the volumetric
flow rate vs. the applied pressure gradient for horizontal flow
in a pipe. These values are reproduced in Table 1. The pipe
radius is equal to R = 0.01m. Use these representative values,

in conjunction with the analytical expression of the volumetric
flow rates determined in the previous section, to compute the
parameters of the constitutive equation.
First, we compute the following sum:
S= (Qep _ h 2 (22)

where Q rep and Qth are the representative value and analytical
expression of the volumetric flow rate. Then, we use the built-
in command of Mathematica, FindMinimum, to determine
the values of n and K for the power-law model, and to and
1] for the Bingham model that minimize the objective func-
tion, J. The approach used here is the least squares method.
For the power-law model, we find n = 0.437 and K = 6.708,
while for the Bingham model the result is to = 77.55 and q =
0.0326. The value of the sum given by Eq. (22) is 9.89X 106
for the Bingham model and 2.67X 107 for the power-law
model. Thus, we conclude that the power-law model fits the
representative data better.


We presented the solution of four non-Newtonian fluid
mechanics problems using Mathematica. The velocity profile
is obtained for the horizontal flow of power-law and Carreau
fluids in pipes and annuli, and for the vertical laminar flow
of a Bingham fluid. These problems have split boundary
conditions and were solved using the shooting techniques.
Analytical expressions of volumetric flow rates for pipe flow
of the Bingham and power-law fluids were derived using
Mathematica. The parameters of the constitutive equation
of non-Newtonian fluids were obtained from representative
data of flow rates measured under different applied pressure
gradients in a horizontal pipe. These problems are simple
enough to constitute an excellent introduction to the field
of non-Newtonian fluid mechanics. Students at the National
Institute of Applied Sciences in Tunis performed well de-
spite no previous knowledge of Mathematica. Mathematica

notebooks are available from author upon request or at the
information center[1]


g gravitational acceleration ( m/s2)
Q volumetric flow rate (m3/s)
L pipe length (m)
n power-law exponent
AP pressure difference (Pa)
R pipe radius (m)
R1,R2 annulus radii (m)
r radial position (m)
v velocity (m/s)
z axial position (m)
K power-law consistency index (N " in1:
8 film thickness (m)
X relaxation parameter (s)
Y1 viscosity (kg in i:)
10 zero-shear viscosity (kg in i:
1r] infinite-shear viscosity (kg in i:)
p density (kg in i
T0 yield stress (kg/m-s)
Trz shear stress (kg/m s)

1. results= 1;search_person_id= 1536>
2. Wolfram, S., Mathematica, A Systemfor Doing Mathematics by Com-
puter, Addison-Wesley, Redwood City, CA (1988)
3. 4. Cutlip, M.B., and M. Shacham, Problem Solving in Chemical Engi-
neering with Numerical Methods, Prentice Hall, Upper Saddle River,
NJ (1999)
5. Wilkes, J.O., Fluid Mechanicsfor Chemical Engineers, Prentice Hall,
Upper Saddle River, NJ (1999) 1

Chemical Engineering Education

M, laboratoryy



in the Unit Operations Laboratory

Oklahoma State University * Stillwater, OK 74078
Brigham Young University * Provo, UT 84602
The recent boom in the biomedical/biochemical in-
dustry has necessitated the introduction of biological
components into the chemical engineering curriculum.
According to the U.S. Department of Labor, the job market for
biomedical engineers is projected to increase 31.4% through
2012.11] In 1990, less than 4,000 students were enrolled in
undergraduate biomedical/biochemical programs; in 2002
there were more than 10,000 students enrolled.[2] In the next
five years, it is estimated that two to three times more students
per year will take biomedical/biochemical courses.
To enhance biomedical/biochemical engineering oppor-
tunities in chemical engineering, experiments involving
enzymatic degradation of cellulose and dialysis of creatinine
were introduced at Oklahoma State University (OSU) in the
Unit Operations Laboratory (UOL). These projects enhance
the instruction students receive in optional Introduction to
Biomedical Engineering and Introduction to Bioprocess
Engineering courses. In the UOL, students work in teams of
Winter 2007

Sundararajan V. Madihally is an assistant
professor in the School of Chemical Engi-
neering at Oklahoma State University. He
received his B.E. in Chemical Engineering
from Bangalore University and his Ph.D.
in chemical engineering from Wayne State
University. He held a research fellow position
at Massachusetts General Hospital/Harvard
Medical School/Shriners Hospital for Chil-
dren. His research interests include tissue
regeneration and the development of thera-
pies for traumatic conditions.
Randy S. Lewis is a professor of chemical
engineering at Brigham Young University.
He received his B.S. and Ph.D. degrees in
chemical engineering from Brigham Young
University and Massachusetts Institute
of Technology, respectively. He recently
served as chair of the Career and Education
Operating Council of AIChE. His research
interests include biomaterial development
and the utilization of renewable resources
for the production of chemicals.

� Copyright ChE Division ofASEE 2007

three participating in three five-week projects during each
semester. While assigning projects, bio-related ones are al-
located preferentially to students enrolled or committed to
biomedical and/or bioprocess courses.
The dialysis experiment demonstrates the fundamental
concepts of a hemodialysis device using creatinine as the
target agent for removal. Creatinine (\IW 113) is one of
several waste products produced in a human that must be
removed by the kidney. Although some dialysis experiments
have previously been demonstrated in the chemical engineer-
ing curriculum using salt solutions with short experimental
times,[3] the use of creatinine has several advantages. These
advantages include its larger relative size to other waste
products and its use, along with urea, as a marker for effec-
tive dialysis treatment. 4] The larger creatinine size leads to a
longer removal time in comparison to other waste products.
The waste product with the longest removal time is often
used in determining dialysis treatment time. Thus, the use of
creatinine leads to a more realistic dialysis experiment-even
with the drawback of longer dialysis time. This work presents
a dialysis model that demonstrates the assessment of model
assumptions. It will detail the dialysis project statement de-
livered to the student team, the experimental protocol, the
dialysis model, experimental results, and student feedback
and assessment. A benefit of incorporating the dialysis project
is that the student can integrate a number of concepts such
as material balances/modeling (i.e., blood and dialysate bal-
ances with assumptions), transport issues (i.e., evaluation of
transport coefficients), model validation of assumptions, and
solving differential equations (i.e., using Polymath) toward a
bioengineering project that allows the student to expand the
scope of his/her chemical engineering education.

A biomedical engineering company makes many biomedical
devices, one of which is a hollow-fiber separatorfor dialysis
machines. Hospitals use dialysis to process the blood of
patients whose kidneys do not effectively remove toxins and
excess water from the blood. The machine has many features
that you will not need to use. We are interested in the hol-
low-fiber membrane separation unit that dialyzes the blood.
Please develop and validate an unsteady-state model for
predicting the temporal profile , .. ', ioit. creatinine removal,
one of several toxic metabolites, from "blood" (represented
by water in this experiment). Using your model, determine
the effects on dialysis treatment time, blood recirculation flow
rate, and transmembrane pressure (AP) for removing 90%
of creatinine. Metabolites and electrolytes ,fi.A. t the osmotic
pressure, which affects the transport of water across the
membrane. This osmotic pressure effect must be included in
your model to determine the amount of water removed from
(or added to) the "blood" during the dialysis treatment.

As part of the model, you need to generate experimental
data from the hollow-fiber membrane separator to obtain an
overall transfer coefficient for creatinine and then you must
validate your model. The hollow-fiber membrane unit is simi-
lar to a shell-and-tube heat exchanger. Basically, blood flows
;ll,.. ,. 1 the inside of the hollow fibers (tubes), and dialysate
(composed of salts similar to normal blood concentrations)
flows ;,ii . .i, il,.. shell-side. Unwanted toxins and other excess
metabolites and electrolytes in the blood difflt e ,Ir.. .. 1, the
fiber walls into the dialysate. The patient's blood is continu-
ously o1. il.air. wilhin hin hel body. The blood compartment
of the body can be assumed as a Continuous Stirred Tank
(CST) in which the partially purified blood coming from the
kidney dialysis unit is returned to the body.

The experiment consists of a hemodialysis unit connected
to a bucket of water containing creatinine (representing the
patient's blood), as shown in Figure 1. A schematic of the

Figure 1. Dialysis unit, dialyzer, and a continuously
stirred tank containing creatinine that represents the
patient's blood.

Chemical Engineering Education

experimental system is shown in Figure 2. Although hemo-
dialysis units are expensive, many dialysis centers regularly
replace their units on an annual or bi-annual basis. Since
there is often a cost for disposal of the units, the supplier of
hemodialysis units at a local dialysis center was contacted and
the supplier donated 10 units to OSU at no charge. It is likely
that such donations can be obtained from other hemodialysis
unit suppliers in the same fashion. Manometers were placed
at all inlets and outlets of the dialyzer to measure pressure
drops from one end of the dialyzer to the other end as well
as to measure transmembrane pressure differences. Three to
four liters of a solution (denoted "blood") containing up to
4.1 mM creatinine were used to simulate the patient's initial
blood concentration. The blood was continuously mixed using
a magnetic stirrer. The blood was pumped to the tube side of
the dialyzer at rates varying between 300-500 ml/min, con-
trolled by the dialysis unit, and blood volume changes were
monitored at regular intervals by weighing the bucket on an
electronic scale. Water (denoted "dialysate") was continuously
added at rates between 815-865 ml/min to the shell side of the
dialyzer and, upon exiting, emptied into a waste sink.
Experiments were conducted over a two-hour period. In the
open-loop experiment, the blood flowed such that the blood
entering the dialyzer from the bucket contained a constant
creatinine concentration and the blood exiting the dialyzer
was sent to a waste bucket. This experiment was performed to
determine the creatinine mass transfer parameter, Kc, neces-
sary to predict the creatinine concentration with time in the
closed-loop experiment. In the closed-loop experiment, the
blood continuously circulated such that the blood volume and
creatinine concentration decreased with time. The changing
creatinine concentration was used to compare model predic-
tions with experimental results. During the experiments, the
inlet and outlet pressures across the dialyzer and the volume
of the blood were recorded every five minutes (via measuring
the weight of the blood). Samples (0.3-0.5 mL) were collected
for analysis of the creatinine concentration.
To analyze the creatinine concentration, one part sample
was mixed with three parts of a solution containing a 10:1 ratio
of 0.14% picric acid and sodium hydroxide. Note that picric
acid is hazardous, highly explosive, and should only be used
under the careful guidance of an instructor. In this analysis,
creatinine reacts with alkaline picrate to form a reddish-yel-
low solution from which the absorbance can be detected in
a spectrophotometer at 490 nm.J51 The absorbances were
converted to concentrations via a linear calibration curve.
Spectrophotometers are a common component in many labo-
ratories and thus there is a possibility that arrangements could
be made to use existing spectrophotometers for the creatinine
analysis. Inexpensive spectrophotometers may be purchased,
however, for as little as $1,500. Other calorimetric methods
also exist for assaying creatinine, although the methods are
more expensive.[6]
Winter 2007

Figure 2. Schematic of Figure 1 showing the dialysis unit
and countercurrent flow in the dialyzer. The creatinine
concentration (C) parameters and volume (VcST) of the
continuously stirred tank used in Eqs. (2) and (3) and (7)
through (9) are shown.

To meet the objectives of the project statement, students
needed to develop a mathematical model for the process.
The blood in the bucket was modeled as a CST with a given
volume (VCST). During the closed-loop experiment, the CST
creatinine concentration (CT ) changes with time accord-
ing to:
dCCST QB out
-C (( B.out _ CST
dt VCST (C c
where QBout is the volumetric outlet flow rate of blood through
the dialyzer and C Bout is the "blood" creatinine concentration
exiting the dialyzer (and entering the CST). Since the CST
volume changes with time, the creatinine material balance
and the total mass balance (assuming constant density) were
combined to obtain Eq. (1). The total mass balance, assuming
constant density, is represented by:
dVCST Bout B in
S= QBout _ QBn (2)
where QB."n is the volumetric inlet flow rate of blood through
the dialyzer. To solve Eqs. (1) and (2) to predict VCST and
Cs with time (as part of the project statement objective),
it is important to know how C B out and QB.out are related to

dialyzer inlet conditions. QB'" is constant and set by the
dialysis machine.
Material balances around the dialyzer identify the inter-
relationship between C B�t" and Q out, and demonstrate how
these parameters can be used in Eqs. (1) and (2) to predict
VCST and CST with time. Assuming countercurrent flow be-
tween the blood and dialysate, the material balances around
the dialyzer are:
QBn _ QBout = QDout _ QD,,n = a PAM (3)

dC' Kc
d = (CD - CB) (4)
dA Q
dCD gAPP + K (
dA QD (C C (5)
Eq. (3) represents that total water loss from the blood side
into the dialysate side following a single pass, resulting from
the average transmembrane pressure difference (AP) between
the blood and dialysate. QB and QD are the blood-side and
dialysate-side volumetric flow rates. APis often constant and
can be approximated as AP = PB av_p avg where PB and PD are
the average pressures of blood and dialysate, respectively. The
convective transport coefficient for water across the dialyzer
membrane is represented by a, and AM is the total transport
area of the membrane. For the dialyzer
used in this study, AM= 1.5 m2 (CL T150L,
Terumo Medical Corporation, Tokyo, This wo
Japan). Eq. (4) is the material balance a dialysis
for creatinine in the blood side ( C) of
the dialyzer where Kc is the mass transfer demonstrates
coefficient of creatinine (units of length of model assi
per time) describing diffusive transport of
creatinine across the membrane relative detail the d
to convective flow. Eq. (5) is the material statement d
balance for creatinine in the dialysate side
(C') of the dialyzer. The differential student
membrane transport area with integration experiment
proceeding from the blood inlet to the
blood outlet is represented by dA.
The assumptions in the development of experime
Eqs. (3) to (5) include: 1) pseudo-steady and studt
state material balances, 2) plug flow, 3)
constant AP, and 4) the Staverman reflec- and as
tion coefficients (o) for all solutes have
a value of zero. Note that o has a value
ranging from 0, denoting solute flows unimpeded through the
membrane, to 1, in which solute is completely reflected by
the membrane and only diffusion occurs.[4] When o=0 for all
solutes, there is no osmotic driving force for water transport.
For this study, the small MW of creatinine (113 Da) relative
to the average pore size of the dialyzer (8,000-10,000 Da)
leads to oz0 for creatinine.[4] For analysis involving co-current
dialysate flow, the term QDout Q D"n in Eq. (3) is replaced with
QD in_ QDout, and the differential sign in Eq. (5) is negative.

The parameters a and Kc must be assessed to use the model.
To obtain a, the closed-loop experiment is performed in the
absence of solutes (or the presence of any solute as long as
o=0 for the solute) by measuring AP and the changing VCST.
Combining Eq. (3) with Eq. (2) suggests that a plot of VCST vs.
time yields a slope of -i, AP \ ,. Thus, the measurement of AP
with the known value of AM enables the calculation of a.
Kc can easily be obtained when aAP< aAPAM is very small compared to QB. n and QD.,n for Eq.
(3), such that QB.ou"' QB.n and QDout" QDn (i.e., QD and QB
are constant) according to Eq. (3). For these assumptions,
integration of Eqs. (4) and (5) results in CoB�"= C s (1-E)
where E is

E 1- exp[-NT(1+ z)]
(1+ z)
E exp [NT(1- z)] -1
exp[N(1- z)] - z

co-current flow

countercurrent flow

and NT and z are KcAM/QB," and QBn"/QDin, respectively.'[7 The
parameter Kc can be obtained from the open-ended experi-
ment with a known z by measuring CCsT (also equivalent to
CBIn) and CoB.o�t, solving for E, solving for NT from either
Eq. (6) or (7), and then solving for Kc at
the given QB." and A . The validity of the
resents assumptions can be assessed by comparing
aAP with K, and aAPAM with QB. and
del that QD,n . The validity of assumptions should
assessment always be checked when applying models.
Thus, the dialysis project is an excellent
tool for allowing students to demonstrate
sis project the validation of assumptions.
ered to the Once values of a and Kc are known, VCST
m, the and C'w can be predicted with time for
fixed dialysate and blood inlet flow rates
protocol, (i.e., a given z value) by utilizing CB,' ut
model, CCST (-E) and QBout__ QB.n. AP\, in the
integration of Eqs. (1) and (2). Although
results, beyond the scope of this article, solving
feedback Eqs. (3) to (5) simultaneously [with Eq.
(3) in differential form] and then applying
ment. the solutions to Eqs. (1) and (2) allows a
more rigorous approach for predicting VCST
and CCs with time. The rigorous approach
eliminates the need for the assumptions that aAP is much
smaller than Kc, and QB and QD are constant as demonstrated
above. In many dialysis models, the assumption of negli-
gible aAP is always assumed.[47] The rigorous approach
is advantageous for students to use when the negligible
aAP assumption is not valid, providing an additional op-
portunity for students to compare the rigorous solution to
the simplified solution to understand the error associated
with assumptions.
Chemical Engineering Education

rk p



| ^ VCST=-0.001t +3.96



0 20 40 60 80 100

Figure 3. Volume of the "blood" (VcST) as a function of time (t).
Creatinine was initially present at 4.1 mM with QB," = 500 ml/min.
The line represents the solution to Eqs. (2) and (7),
where the slope is used to evaluate K'.

In addition, the models can be further expanded to include mul-
tiple species and the utilization of species in which o # 0. Thus,
the dialysis project has great potential for many applications
involving model development, validation of assumptions, and
comparison with experimental results. Exposing students to
the various levels of model development helps them learn
how to simplify models using certain assumptions.
For example, blood contains proteins, salts, urea, and other
metabolites. The proteins are too large to transport through
the dialyzer (o = 1 and K for protein is negligible), such that
the proteins contribute to an osmotic pressure. If only proteins
(P) and creatinine (C) were present, Eqs. (3) to (5) would be
expanded to:

-dQB -dQD cxA - RTCB
dA dA
dCB Kc D
dA QB (Cc c)





I1 ll

Kc +K' D

The RTC B term in Eq. (8) is the osmotic pressure contribu-
tion due to proteins. Eq. (9) is the protein material balance
on the blood side that demonstrates the protein concentration
can change as a result of water flow through the membrane
(note that protein is not in the dialysate, so a protein balance in
the dialysate is not needed). The solutions to these equations
Winter 2007

[combined with Eqs. (1) and (2)] can be solved to
provide VCST and CCsT predictions with time. The
value of RTCB is typically 28 mmHg, and dialysis
is often performed where AP is on the order of 200
mmHg.[4] Thus, K' is positive leading to a decreas-
ing QB along the length of the dialyzer. According
to Eqs. (8) and (9), the protein concentration will
increase due to the loss of water and the creatinine
concentration in the blood will decrease. The model
can be used to assess the degree to which increases
and decreases occur. Unlike the example given in
this article using creatinine alone where K' (or aAP)
was assumed negligible, K' is required to assess the
changes in protein concentration on the blood side
(as a result of water loss) according to Eq. (9). A
valuable exercise for students would be to derive
Eqs. (8) through (11) and show how the equations
can be solved with Eqs. (1) and (2) to predict time
profiles of VCST and C .sT

Figure 3 shows a plot of VCST VS. time for a closed-
loop experiment in which creatinine was initially

present at 4.1 mM and QB,' = 500 ml/min. With a negative
slope of one ml/min (i.e., 1 ml/min of water transports from
the blood to the dialysate) and A, = 1.5 m2 (CL T150L,
Terumo Medical Corporation, Tokyo, Japan), aAP is 6.7 X
105 cm/min (aAPAM = 1 ml/min). According to Eq. (3), QB.out
/ QB." = 0.998. Thus, QB is essentially constant such that the
assumption of aAPA << QB.'" is valid. Since QD was greater
than QB, the assumption of aAPA << QD,1 is also valid for
this particular experiment. Although AP was not measured
at QB." = 500 ml/min, AP z 26 mmHg was observed at QB. n
= 300 ml/min, leading to an approximation of a = 2.6 x 106
cm min' mmHg 1.
When evaluating Kc, the open-loop experiment was per-
formed at QB.n = 300 ml/min and QDn = 817 ml/min with
CS = 3.22 mM. The analytical measurements for this work
were sensitive enough to distinguish differences between
Cs and C" . The measured value of C" was 2.03 +
0.03 mM, leading to C .." = 0.63 CcST and E = 0.37. For
countercurrent flow with z = 0.37, Eq. (7) yielded a value
of 0.5 for NT. Thus, Kc is 0.01 cm/min, which validates the
assumption that aAP << K,. This value of Kc is similar to
values observed for other hemodialyzers. 71
Once model parameters were obtained and the assump-
tions were validated, Eqs. (1) and (2) (with CB .." =(CST
(1-E) and QB.out = sQB .,. AP \ ,) were solved simultaneously
using Polymath[81 to predict VCST and CcST with time. The
predictions were compared to experimental results from a
closed-loop experiment in which 2.54 mM of creatinine was
initially present in 4 liters with QB.n' = 300 ml/min and QD n'
= 865 ml/min (z=0.347). From Eq. (7) u illi \ =1.5 m2 and

A benefit of incorporating the dialysis project is that the student can integrate a number of concepts
such as material balances/modeling (i.e., blood and dialysate balances with assumptions),
transport issues (i.e., evaluation of transport coefficients), model validation of assumptions,
and solving differential equations (i.e., using Polymath) toward a bioengineering project that allows
the student to expand the scope of his/her chemical engineering education.

Kc = 0.01 cm/min (NT=0.5), E=0.37. Figure 4 shows that
the model results for CoCST, with aAP ranging from 0 (con-
stant flow assumption) to 20 X 105 cm/min (representing a
3-fold increase in aAP from the experimental aAP value),
are the same. Increased water transport can occur via either
increasing AP or adding a constituent that contributes to the
osmotic pressure (where o # 0). The model predictions are
in general agreement with the experimental results although
there is a small discrepancy. The time to remove 90% of the
creatinine is 80 minutes. In all cases, varying aAP did not
affect the C ST profile (as expected with aAP << K,). After
100 minutes, however, the predicted VCST was 4000 ml, 3900
ml, and 3700 ml for aAP=0, aAP=6.7 X 105 cm/min, and
aAP=20 X 105 cm/min, respectively. As is evident, aAP is
critical for predicting water loss but does not affect predic-
tions. Figure 4 also shows predictions for aAP=0 with QB.
= 500 ml/min (E=0.24). As seen, the CcST profile does not
drastically change and the time to remove 90% of the creati-
nine only decreases to 76 minutes.

Some of the comments from students included "this experi-
ment trained us with [nontraditional] equipment" and "I liked
seeing how close a model would actually fit experimental
data." There were several indirect assessments. Students
liked this project and explored more than they were asked
to do in the project statement. For example, they explained
the difference in various types of dialysis processes, pro-
vided statistics about each type, examined the relevance of
creatinine in clinical settings, explored the importance of
osmolarity, and developed an understanding for the need
of electrolytes in the dialysate. Whlil . pi i. . iing their find-
ings, they named their patient, talked about poor "Charlie"
needing to sit for two hours while the dialysis was taking
place, and worried about how creatinine generated in the
body during dialysis would affect the creatinine removal
process. To account for creatinine generation during dialy-
sis, a constant generation of creatinine could be introduced
into the experiment. It was also observed that as students
were told that they could present their results at the regional
and national American Institute of Chemical Engineers
(AIChE) student conferences, students were more willing
to spend extra time on the project. One student presented
his team's results at the 2004 Mid-America Regional

AIChE Conference and won second place.
One difficulty observed during the implementation of the
dialysis project was that students tended to focus on the
analysis (i.e., measurements), and focused less on the mod-
eling aspects. For instance, students spent ample time on
the creatinine analysis. One problem that was encountered,
however, was that students were not used to using micropi-
pettes and error in the calibration curve could be dominant
if proper volumes were not dispensed every time. Thus, it is
important to train students in using equipment that is not often
associated with traditional unit operations experiments. With
regards to modeling, the students did not always explore the
resources for model development. It would be beneficial for
the instructor to direct students toward resources containing
information about model development. The main point is
that students involved in nontraditional experiments should
have some training or guidance (such as identification of key
resources) to help them achieve their objectives.

Through the incorporation of creatinine dialysis, under-
graduates can integrate a number of concepts such as material
balances/modeling (i.e., blood and dialysate balances with


* Experimental
- E=0.37, K'=0 to 20 x 10-5 cm/min
- E=0.24, K'=0 cm/min

90% removal
----.----. 90 rem ovY al ----- ---- -

0 20 40 60 80 100

Figure 4. Creatinine concentration (initially 2.54 mM) in the
continuously stirred tank (CST originally at 4000 ml) as a
function of time for QB,'" = 300 ml/min and QDin = 865 ml/min.
The model is the solution to Eqs. (2) and (3) with CCB,-" = Ccsr
(1-E) and QB,ot = QBin- K'A, A is 1.5 m2.
Chemical Engineering Education

assumptions), transport issues (i.e., evaluation of Kc), model
validation (i.e., validating assumptions), and solving differen-
tial equations (i.e., using Polymath) towards a bioengineering
project that allows the student to expand the scope of his/her
chemical engineering education. The students enjoyed the
exposure to "nontraditional" experiments and this project
provided them an opportunity to connect bioengineering ex-
periments to material learned in the classroom. Deriving the
differential equations from the continuity equation requires the
student to draw on his/her math and engineering knowledge. It
is important for students to assess the validity of assumptions
when applying experimental results to model equations, and
this project allowed for such opportunities.

The authors would like to thank the following students
in the School of Chemical Engineering at Oklahoma State
University who conducted the experiments: Paul Engel,
Kimberly Garrison, Kimberly Northy, and Jason Powell. In

addition, we thank the Stillwater Dialysis Center for helping
secure the dialysis machines and cartridges.

1. (2004)
2. (2004)
3. Cavanagh, D.P, and L.H. Herbertson, "Effective Laboratory Exercises
for an Introduction to Biomedical Engineering Course," Proceedings
of the ASEE Annual Conference and Exposition, Salt Lake City, UT.
4. Foumier, R.L., Basic Transport Phenomena in Biomedical Engineering,
Taylor and Francis, Philadelphia (1999)
5. Narayanan, S., and H.D. Appleton, "Creatinine: A Review, " Clin. Chem.
26, 1119, (1980)
6. Bell, T., Z. Hou, Y. Luo, M. Drew, E. Chapoteau, B. Czech, A. Kumar,
"Detection of Creatinine by a Designed Receptor, "Science, 269(5224),
671, (1995)
7. Cooney, D.O., Biomedical Engineering Principles, Marcel Dekker,
New York (1976)
8. Cutlip, M.B., and M. Shacham, Problem Solving in Chemical Engineer-
ing with Numerical Methods, Prentice Hall PTR, Upper Saddle River,
NJ (1999) 1

Winter 2007

72 Chemical Engineering Education

on the

University of Florida Home Page
© 2004 - 2011 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Powered by SobekCM