chemical engineering education
... of the University of Cape Town, South Africa
Solid-Liquid And Liquid-Liquid Mixing Laboratory for Chemical Engineering Undergraduates (p. 1011
a Barar Pour, Benoit Norca, Fradette, Legros, Tanguy
0 Random Thoughts: How to Prepare New Courses While Keeping Your Sanity (p. 121)
c Brent, Felder
u A Student-Centered Approach to Teaching Material and Energy Balances: 1. Course Design (p. 93)
a "' Bullard. Felder
V Teaching Transport Phenomena Around a Cup of Coffee (p. 137)
w A Population Balance Based Design Problem in a Particle Science and Technology Course (p. 88)
2 Ehrman, Castellanos, Dwivedi, Diemer
. E Conceptests for a Thermodynamics Course (p. 107)
D - Teaching Process Engineering Using an Ice Cream Maker (p. 131)
0 a Kaletun.:, Duemmel. Gecik
u 2 Using StudentTechnical Conferences to Build Multidisciplinary Teamwork Skills (p. 81'
c Controller Performance AssessmentThrough Stiction in Control Valves in a Process Control Class (p. 123)
Cec Srinivasan, Rengaswamy, Harris
SChemical Engineers Go to the Movies (Stimulating Problems forthe Contemporary Undergraduate Student) (p. 11 5)
The Development and Deployment of a Virtual Unit Ops Laboratory (p. 1-144
Vaidyanath. Williams, Hilliard. Wiesner
r SI j teaching tips
This one-pagecolumn will present practical teaching tips in sufficientdetail thatChE educators can
adopt the tip.Thefocus should be on the teaching method, not content. With no tables or figures
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.
EXPLAINING THE CONVECTIVE TERM
IN THE NAVIER-STOKES EQUATIONS
DANIEL A. CROWL AND JASON M. KEITH
Michigan Technological University * Houghton, MI 49931
M any undergraduatesstrugglewiththeshell balance
approach to deriving conservation equations in
transport phenomena, in particulartheconvective
term (pv v) of the Navier-Stokes momentum equation. For
simplicityand brevitywewillfocus on thex-componentina uni-
directional flow system in rectangular coordinates given by:
S(pv,) +a (pv,) aT aP
( += -- -+ _ (1)
at ax ax ax
Students have difficulty comprehending what is done in
the leading textbooks,111 where it is stated that the rate of
momentum entering the element is pv v.The shell balance
approach is typicallyfollowed with the student question"Why
is that term rho-v-v?" Providing the typical answer that"the
units work out" is not an explanation.
DESCRIPTION OF THE METHOD
The following physical derivation for the bulkflow or con-
vective term is easier for students to understand. We begin
our analysis by considering a stream of elements, of width
w, flowing into a control volume x y z, as illustrated in
Figure 1 .These elements bring with them momentum due to
their mass and velocity.
The mass ofeach element is pV = pw y z.Themomen-
tum is given by mvx, which is pvxw y z.The rate at which
n elements (they are in a continuously moving stream) come
into the control volume is n / t= v /w.Thus, the rate of
momentum brought into the control volume is equal to the
productofthemomentumofeach element multiplied bytheir
rate: pv w y zv /w = pvxy y z. Subtracting this value
from the momentum going out of the control volume, divid-
ing by the control volume, and taking the limit as x- 0,
gives a (pvv).
Copyright ChE Division of ASEE 2007
GENERALIZATION OF THE METHOD
This methodcan begeneralized. Forinstance,theenergyenter-
ingthevolumeelementofwidth w is pCpTw y z.Following
the above procedure gives rise to the thermal energy convec-
tion term, a (pCp VT).
Also, using thespecies mass Cw y zenteringthecontrol
volume gives rise to the mass convection term, a (vcA ).
Physical intuition is used to derive the convective terms in
chemical engineering transport equations. Using this method
will allow students to better understand the nature ofconvec-
tive transport governing many applications.
1. Welty, J.R., C.E. Wicks, R.E. Wilson, G. Rorrer, "Fundamentals of
Momentum, Heat, and Mass Transfer," 4th Ed., Wiley, New York
2. McCabe, W.L., J.C. Smith, P. Harriott, "Unit Operations of Chemical
Engineering,"6th Ed., McGraw-Hill, New York (2001)
3. Bird, R.B., W.E. Stewart, E.N. Lightfoot, "Transport Phenomena," 2nd
Ed., Wiley, New York (2002)
4. Geankoplis, C.J., "Transport Processes and Separation Process Prin-
ciples,"4th Ed., Prentice Hall, Upper Saddle River, NJ (2003) p
A /7 /A
w W W Ax
Figure 1. Conceptual cartoon of bulk flow into a control
volume (thicker lines).
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
EDITORIAL AND BUSINESS ADDRESS:
Chemical Engineering Education
Department of Chemical Engineering
University of Florida * Gainesville, FL 32611
PHONE and FAX : 352-392-0861
Phillip C. Wankat
James O. Wilkes, U. Michigan
LEARNING IN INDUSTRY EDITOR
William J. Koros, Georgia Institute of Technology
John P. O'Connell
University of Virginia
* PAST CHAIRMAN *
E. Dendy Sloan,Jr.
Colorado School of Mines
University of Colorado
Thomas F. Edgar
University of Texas at Austin
Richard M. Felder
North Carolina State University
H. Scott Fogler
University of Michigan
Carol K. Hall
North Carolina State University
University of Toledo
Ronald W. Rousseau
Georgia Institute of Technology
C. Stewart Slater
Donald R. Woods
Vol. 41, No. 2, Spring 2007
Chemical Engineering Education
Volume 41 Number 2 Spring 2007
74 Duncan Fraser of the University of Cape Town, South Africa
Jennifer M. Case
> RANDOM THOUGHTS
121 How to Prepare New Courses While Keeping Your Sanity
Rebecca Brent, Richard M. Felder
88 A Population Balance Based Design Problem in a Particle Science and
Sheryl H. Ehrman, Patricia Castellanos, Vivek Dwivedi,
R. Bertrum Diemer
107 Conceptests for a Thermodynamics Course
John L. Falconer
81 Using Student Technical Conferences to Build Multidisciplinary
David L. Silverstein
93 A Student-Centered Approach to Teaching Material and Energy
Balances: 1.Course Design
Lisa G. Bullard, Richard M. Felder
123 Controller Performance Assessment Through Stiction in Control Valves
in a Process Control Class
t.,,,,11,, i,, .;..;.., ,, t.,J,,i .1,1,, ii,,i Rengaswamy, Sandra Harris
101 Solid-Liquid And Liquid-Liquid Mixing Laboratory for Chemical Engi-
Sanaz Barar Pour, Gregory Benoit Norca, Louis Fradette,
Robert Legros, Philippe A. Tanguy
131 Teaching Process Engineering Using an Ice Cream Maker
Goniil Kaletunc, Kevin Duemmel, Christopher Gecik
144 The Development and Deployment of a Virtual Unit Ops Laboratory
Sreeram Vaidyanath, Jason Williams, Marcus Hilliard,
> CLASS AND HOME PROBLEMS
115 Chemical Engineers Go to the Movies (Stimulating Problems for the
Contemporary Undergraduate Student)
Jimmy L. Smart
137 Teaching Transport Phenomena Around a Cup of Coffee
Jean Stiphane Condoret
92 Call for papers
S inside front cover Teaching Tip
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 ofaddress should be sent to CEE, Chemical Engineering Department, University
of Florida, Gainesville, FL 32611-6005. Copyright � 2005 by the Chemical Engineering Division, American Society for
Engineering Education. The statements and opinions expressed in this periodical are those of the writers and not necessarily
those of the ChE Division, ASEE, which body assumes no responsibilityfor them. Defective copies replaced if notified within
120 days ofpublication. Writefor information on subscription costs andfor back copy costs and availability. 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).
---- U s_____________________________________
of the University of Cape Town, South Africa
JENNIFER M. CASE
Department of Chemical Engineering, University of Cape Town
A s a young Ph.D. 9j g
student in chemical
engineering at the ,
University of Cape Town A
(UCT), Duncan Fraser was
asked on a number of oc-
casions to cover courses for
lecturers who were on sab-
batical. This was maybe not
the most ideal arrangement
with regard to completion
of the thesis (he says it took
"forever") but the teaching
bug bit him well and truly,
and this experience was the
start of a long career devoted
to all aspects of the teaching
of chemical engineering.
Duncan has taught nearly ev-
ery course in the curriculum,
has introduced numerous in-
novations in the classroom,
has overseen major curricu-
lum change, has researched
the learning of students, and
has worked at institutional,
national, and international
levels to improve the quality
of undergraduate education.
On top of this he has pursued
technical research in process
synthesis that has recently
been awarded with an impressive "B-rating"* fror
tional Research Foundation in South Africa.
Richard Felder, Hoechst Celanese Professor Er
Chemical Engineering at North Carolina State Univ
the following recollections: "I first encountered Du
* B-rated researchers are those who "enjoy considerable ii
recognition by their peers for the high quality and imn
recent research outputs."
Duncan with his youngest student, grandson James.
eral decades ago at an AIChE conference education session,
where he spoke about some work he was doing with black
students from educationally disadvantaged backgrounds. I
was struck by several things as I listened. First, the work he
described was remarkably innovative and clever-as good
as any educational work I knew about going on anywhere in
the world. In addition, he was doing it in South Africa and
� Copyright ChE Division of ASEE 2007
Chemical Engineering Education
in the era of apartheid, which spoke volumes to me about
both his principles and his courage. I introduced myself to
him afterwards and that began a lasting friendship." High
praise indeed from one of the legends in chemical engineer-
How did the uncertain Ph.D. stand-in lecturer develop
into the internationally renowned engineering educator?
Duncan identifies two early experiences that took his innate
enthusiasm for teaching to a new level. First, he attended a
series of teaching workshops run by the UCT Medical School
in the early '80s. '"To go away for two whole days and just
to think about your teaching-that was a
rare experience," he says. The second key
experience was a sabbatical at Imperial
College in London during 1984/85. Here D
he was delighted to work with kindred inspirati
spirits in the innovative group of Stephen working
Richardson (now head of the Department
of Chemical Engineering), Geoff Mait- He says tl
land (now chair of Energy Engineering), the stude
and John Perkins (now vice president and
dean at the University of Manchester). are brig
He particularly remembers the manner in but he a
which they always gave detailed construc- adv
tive feedback to students, and how he took
this practice back home into his teaching exj
The mid-1980s in South Africa was a
difficult time. The struggle against apart-
heid was in full swing and black schools were very troubled
places. At this time, the government began to relax the racial
restrictions on higher education institutions and the first black
African students entered chemical engineering at UCT. (Un-
der the apartheid system UCT had in 1959 been designated
a "whites-only" university, and only small numbers of other
racial groups had been permitted to study engineering there,
but no black African students). These students were coming
from schooling backgrounds that were very poorly resource
and often quite dysfunctional, being at the heart of the struggle
and subject to boycotts and protests.
From his early student days Duncan had a passion for a
nonracial society emanating from his church involvement,
and he was keen that these students should succeed despite
their poor school backgrounds. Cyril O'Connor, then head
of the department and now dean of the Faculty of Engineer-
ing and the Built Environment at UCT, picks up the story:
"Duncan recognized these challenges sooner than most and
vigorously and enthusiastically began developing new ap-
proaches to teaching chemical engineering to these students."
Duncan first tried a system of extra sessions, but found that
this further added to the high workload and had no impact
on student success. He also tried increasing the number of
postgraduate teaching assistants for the course, but this had
equally little impact. He then came across the literature on
Minority Engineering Programs in the United States and
found out about collaborative study groups. Duncan says that
he thought to himself, "I've got to try this!" He had never
spoken to anyone else about this idea, and started out with
considerable fear and trepidation. He was going to be doing
something rather different in the classroom and negotiated
with the class for the go-ahead. They said they were happy
to give it a try. Duncan now says that the first collaborative
learning session was "one of the most exciting things I have
ever done." Students worked the whole afternoon without a
break, and Duncan noted a real "buzz"
in the class. He then continued with this
innovation through the remainder of
greatest the course and was delighted to note a
homes from marked effect on success rates. He notes
h students. that the innovation seemed to improve
the success rates of all students, but that
at least half of black African students the most.
in the class Cyril O'Connor notes: "There were
than him, two remarkable outcomes to Duncan's
research. First, it soon became clear
ist has the that the methods he was promoting also
age of represented a far better way to teach
engineering to all students irrespec-
ence. tive of their educational background.
Secondly, it also became clear that the
unique circumstances in which we found
ourselves at UCT with such a diverse
student population provided an opportunity for us to develop
new approaches that would have a universal appeal, as has
happened with Duncan's research."
Duncan was soon called on to advise on the design of all
engineering programs at UCT, and he played a key role in a
wholesale curriculum overhaul in the early 1990s. This in-
volved the introduction of an engineering course to the first
year of the program, which had previously had the traditional
format of a purely science-based introduction. Duncan found
it challenging to develop an engineering course at first-year
level, and after the first lecture he found he had to throw his
meticulous plans out of the window and rethink ways of
reaching students at their level. He devised an innovative
series of experiments to introduce students to the fundamen-
tals of the discipline. His philosophy here involved taking
students from the known to the unknown. They therefore
studied heat transfer using coffee cups as well as sophisticated
digital thermometers. Reaction kinetics was approached by
boiling potatoes and measuring the reaction progress using
vernier calipers then modeling the results using simple DEs
and spreadsheets. Mass transfer was seen in the dissolution
of lollipops in a stirred beaker of water. Duncan had seen an
Vol. 41, No. 2, Spring 2007
Duncan (back row left)
with members of the
hiking group he led.
Jenni Case, is in the
front row with her son.
article about a mass trans-
fer experiment in Chemi-
cal Engineering Educa-
tion, but it involved loose
candies that needed to be
dried and weighed, and he
was looking for something
simpler to measure and
also to model. He was out
shopping with his young
daughters one day and they' - -
had asked him to buy some
lollipops, at which point he
realized this was exactly the thing he needed. Much to their
dismay he took one of the lollipops to try out how long it
would take to dissolve in water, and he found that it suited
his purpose perfectly.
Duncan played an important role in mentoring new aca-
demic staff members in the Department of Chemical Engi-
neering at UCT. Professor Sue Harrison, former head of the
department, remembers the early days: "As a young novice
lecturer, directly out of a Ph.D. degree, I was extremely for-
tunate to share my first teaching assignment in material and
energy balances with Duncan Fraser. He educated me from
day one in pacing and varying my lecture speed and complex-
ity, on the importance of learning 92 first and family names
of culturally diverse students within the first two afternoons
of term, and on the elements of the students' baptism by fire
in problem solving. I could not have had a better teacher, or
educator of the educator!" Later on, Sue only agreed to take
on the head of department position with the support of Duncan
as Director of Undergraduate Studies. She says, "Not only is
he the member of staff with the most extensive knowledge
of the curriculum, the learning process, the student body,
and the academic administration, he is also one of the most
dedicated. I have had the pleasure of working with him in
both the introduction of the revised design-based curriculum
and in ensuring a robust outcomes-based curriculum-both
rigorously tackled. Probably the greatest compliment to him
as an educator is that, in a department where every academic
is a serious researcher, these developments were achieved
through his championing with the buy-in of the entire staff.
This inspiration to continue to marry research and teaching
enthusiasm in the same academic group is key to the achieve-
ments of our department."
Duncan's reading of the education literature also sparked
an interest in conducting educational research to address the
challenges faced in improving chemical engineering educa-
tion at UCT. The Chemical Engineering Department, with
its strong research profile and its tradition of dedication to
undergraduate education, plus solid industrial support, was
the perfect place for launching this new enterprise involving
research on education. A major donation from Caltex (now
Chevron) in 1993 allowed Duncan and colleagues to establish
a lectureship in the department to focus specifically on these
issues. I met Duncan when I came into this post in 1996 and
we have worked collaboratively on many projects since then.
A major highlight came in 2002 when a paper we had jointly
published in Chemical Engineering Education was awarded
the Corcoran Award for best paper of the year.
Another key initiative that helped build an engineering
education community at UCT was the establishment in 1996
of the Centre for Research in Engineering Education (CREE),
which Duncan spearheaded together with Jeff Jawitz. It is now
a full decade down the line and CREE has evolved into an
active network of tertiary science and engineering education
researchers in the Western Cape province (see
uct.ac.za> for more details). A recent UCT review of CREE
commended it highly for the quality of research output and
the presence of a national and international profile for the field.
There is an exhaustive list of events that Duncan has helped co-
ordinate to raise the profile of engineering education, including
seminars, workshops, conferences, and the like. A particularly
Chemical Engineering Education
Above left, Duncan and his wife, Jenny, on the
Cam. Above, Duncan with Richard Felder at Cape
Point in 1996. Left, colleagues Jenni Case and Dun-
can Fraser with 2003final-year students Rosanna
Martin and Bryan Maytham, exhibiting a poster on
their education research project.
memorable series of workshops has been run by Richard Felder together
with Rebecca Brent, and these have made a huge impression, particularly
on young engineering academics in South Africa.
It has not always been easy going. At times Duncan struggled to
get recognition for his work and felt discouraged. He says that it is
his faith in God that has kept him going. He is also sympathetic to
the challenges that young academics face in following their passions
and building a scholarly focus to their teaching.
Close to retirement now, Duncan shows no signs of slowing down. In
2006 he was centrally involved in setting up a new initiative that took
second-year chemical engineering students on an extended industrial field
trip involving active engagement in the plant environment. This event
was particularly well received by students, and was only made possible
by a donation from Xstrata, a mining company, who also facilitated the
appointment of a second lecturer to focus on education development.
The incumbent of this post, Linda Kotta, a young chemical engineer
with industrial experience and a master's degree in education, has been
the latest excellent addition to the team. Duncan has also recently been
appointed to the position of assistant dean in the Faculty of Engineering
and the Built Environment at UCT, tasked with addressing the throughput
of quality engineers across all the programs in the faculty.
Vol. 41, No. 2, Spring 2007
FAMILIES AND FUN
Duncan was born in Cape Town in 1946, but most
unusual for a Capetonian, he spent his childhood and
youth in a range of places across the Southern African
continent. This was because his father was a Method-
ist minister, and these early experiences established in
Duncan a love of wide open spaces, a fascination with
birds and wildlife, a passion for the country, and an en-
joyment of traveling long distances on dirt roads (which
fortunately his children also got used to in many family
holidays on the road!). Duncan also built on this initial
immersion in a spiritually focused home to become a
committed Christian, and throughout his life he has
been an active leader in his church and other Christian
organizations. Growing up in a church manse he got
used to having a very open and hospitable home, and
together with his wife, Jenny, he has now replicated a
sense of that in their own home in Cape Town, which
has a constant stream of visitors and lodgers.
While at junior high school in a relatively rural back-
water of South Africa, he came across a pamphlet on
industrial chemistry, and decided that it sounded like
a career choice that he would like. As it turned out the
pamphlet was possibly quite dated as the industrial
chemistry course at UCT had already changed into the
discipline of chemical engineering. Fortunately, how-
ever, by the end of senior school his teachers had also
become aware of this development, and so he arrived at
UCT in 1964 to study chemical engineering. Although
he had always been the top student at his school he
failed some of his first class tests in that first year, and
The Fraser clan on sabbatical in the
UK in 1985...
this experience is still with him today when
he encourages first-year students, many of
whom have similar experiences. Soon, how-
ever, he found that he loved the discipline,
feeling that Ii .a, me!" He graduated in the
regular four years with first-class honors, a
As noted earlier, the Ph.D. was something
else--a thesis focusing on turbulent air flow,
analyzing velocity fluctuations in three
dimensions and building the circuitry to
sample data, something that Duncan wryly
notes would be trivial with today's technol-
ogy. One very important development dur-
ing his Ph.D. studies was an acquaintance
with a young doctor named Jenny Hurley.
They met while playing mixed doubles squash at a Student
Christian Association conference. Prior to this point he had
come to the sad conclusion that women and Ph.D. studies were
a poor combination, but meeting Jenny quickly changed his
mind. Six months after that first meeting they were married.
The wedding took place on Jenny's parent's farm in Zimba-
bwe, at a turbulent time in that country's history, happening
as it did exactly a week after the start of the guerilla war that
occupied most of the '70s. Duncan and Jenny have retained
a strong emotional commitment to Zimbabwe during the
changing fortunes of that country's political and economic
Jenny finished her final year of medical studies during
their first year of marriage, and when she started working
the following year she had to support Duncan, as he had run
out of funds with his lengthy Ph.D. pursuit. Jenny became
pregnant with their first child, Debbie, while working around
108 hours per week in cardiology. Duncan remembers 1976 as
the year when he started his first job, they had a small baby,
and he was writing up his dissertation after hours. He says
"I wouldn't want to live through that year again!" He is also
reminded of the joy of being a new parent, however, and still
remembers his utter disappointment and surprise when Debbie
didn't win the "baby of the year" competition for which he
had enthusiastically entered one of her baby photos.
He was delighted when, after three years in industry, UCT
invited him to apply for a lecturing position. Jenny says that
the day Duncan started working at UCT he was a different per-
son. Duncan knows that it is the stimulation and the academic
freedom that has kept him in the same job for nearly 30 years.
He also values the three years he spent in industry, however, and
continually draws on this experience in his academic work.
The remainder of the Fraser brood arrived in relatively
quick succession, and Duncan has played a very important
role in the lives of all four of these interesting young people.
He particularly enjoyed watching them as students and re-
calling his own happy memories of varsity life. The children
all studied at UCT, are now all married (the last wedding in
the family took place in April 2007) and busy producing the
next generation of Frasers. Although Duncan didn't manage
to convince any of them to study chemical engineering, their
careers reflect the same interesting mix of people and technol-
ogy that Duncan has made his speciality. Daughter Debbie
works in developing small businesses, son David is a systems
analyst developing software for the process industries, son
Andrew is a mechatronic engineer and daughter Anni is an
occupational therapist working in a psychiatric hospital in
Johannesburg. Duncan says that he is grateful for what has
been such a "rich" family life, including many holidays to
game reserves and at the seaside, even if he has maybe not
amassed riches in the conventional financial sense.
Duncan still remembers going on sabbatical with four
children aged between one and eight years old. They arrived
at Heathrow airport with more than 20 pieces of luggage be-
tween them and a kindly porter took charge and put Duncan
on the escalator first with, piece by piece, the luggage and the
Duncan's technical research is in process synthesis, and the
amazing thing is that the stimulation to go in this direction also
came from his very formative sabbatical at Imperial College.
Chemical Engineering Education
Initially this research was aimed at improving energy recovery
in chemical processes. In 1988 he produced a pinch analysis
study of the Caltex refinery where he had cut his teeth as a
graduate engineer 10 years earlier. To this day this publication
remains the only refinery-wide study in the literature. In 1989
Duncan developed the concept of a minimum flux to replace
the minimum approach temperature that was traditionally
used as a design parameter for heat exchanger networks. This
enables one to replace a set of stream-dependent minimum
approach temperatures with a single minimum flux, which
reduces the optimization of such systems from a multivariable
one to a single-variable one.
Some of Duncan's most exciting technical work arose from
his work with a particularly gifted Ph.D. student, Nick Hallale.
Building on the work that had been done over 20 years in the
development of heat exchanger networks, they were able in
two years to develop a parallel approach for mass exchanger
networks. This has made it possible to optimize the design of
mass exchange networks ahead of detailed design, as well as to
design systems that meet the optima established. This means
that design alternatives can be compared without having to
do detailed design. Duncan and Nick also demonstrated that
their approach leads to designs that are better than or equal
to those generated by mathematical programming techniques,
for less effort. Currently Duncan is working on developing
approaches to the analysis and design of joint heat and mass
exchange systems -work of crucial importance in developing
more sustainable chemical process systems.
At a deep philosophical level there are interesting links
between Duncan's technical research and his engineering edu-
cation work. Process synthesis involves getting the structure
... and on the occasion of his
eldest daughter Debbie's wedding
of processes right before optimizing
individual pieces of equipment. In
education, Duncan has found that you
get limited returns if you only make
S changes in one course; you need to
work on the whole structure in order
to move forward.
Duncan is a very happy occupier
f of the cramped quarters that one is
subjected to in economy-class air
travel, and has clocked up impres-
sive mileage around the globe in
building academic contacts in both
chemical engineering and engineer-
ing education. Over the last few years
Duncan has found himself "out of town" for approximately
two months each year, and despite the jet lag, lost luggage,
missed flights, and general exhaustion he still seems to have
an appetite for more! I have had the lucky opportunity of ac-
companying Duncan on a few trips to education conferences,
and have discovered what impels Duncan to cross time zones
again and again: Travel is for Duncan a precious time away
from undergraduate students, postgraduate students, small
children, big children, and the remainder of the huge commu-
nity that all depend on him for help. Duncan himself says that
he likes the stimulation of meeting new people on his travels,
and that he also likes to take himself out of his comfort zone,
something that the legendary Scott Fogler says is essential to
enhancing your creative abilities. It also helps not having to
keep the proverbial 10 balls in the air, but rather being able
to focus on one thing at a time. Walking the Yorkshire moors
or strolling along a beach in Rio are the times when Duncan's
mind crosses all the boundaries that restrict it during normal
life. He comes up with all sorts of inspiring ideas, and I have
found it to be quite a treat to be dragged along on a walk by
Duncan at the end of a conference day.
On the technical front, he had a fruitful formal collaboration
over a number of years with the late Professor Zsolt Fonyo
from the Technical University of Budapest, who passed away
in 2005. On one such trip I happened to accompany Duncan
following an engineering education conference we had just
attended in Poland. In the few minutes we had available before
catching a train from Krakow, I headed off to buy food, while
Duncan headed in another direction to draw money. We had
arranged to meet on the advertised train platform, but a subtle
complication entered when the Polish train authorities decided
Vol. 41, No. 2, Spring 2007
at the last minute to change the platform. I happened to notice
the advertised correction, but Duncan's mind was presumably
solving a difficult equation or planning a cross-continental
curriculum reform, and so he settled down nicely at the old
platform. As the train pulled out of the station I realized that
Duncan was not on the train with me, and also that I had both
train tickets while he had all our money! I would also need
to somehow make myself known to Zsolt, who was meeting
us in Budapest on the other end, as he was expecting to see
Duncan. I therefore arrived with a handwritten note stating
that I was "Duncan Fraser's colleague" and fortunately Zsolt
found me, but when he asked me 'Where is Duncan?' I had to
reply that I didn't know-hopefully somewhere making his
way across Eastern Europe to Budapest! Zsolt knew Duncan
to be famously eccentric but this news had him in stitches.
Fortunately Duncan managed to find another train heading to
Budapest and we were all reunited much later that evening,
but it took Duncan a while to live down this incident.
Another productive collaboration took him to India to work
with Professor Uday Shenoy from the Indian Institute of Tech-
nology, Bombay. Together they developed a novel approach
to sizing mass exchange units that avoids the discontinuity
of the conventional method. One of the bizarre features of
modem air travel struck him on returning from one of his trips
to India-having dinner in Mumbai one evening, and being
back in Cape Town for breakfast the next morning!
Duncan is great at crossing borders, not only in a geo-
graphical sense, but also academically. Although he identifies
fervently with the chemical engineering discipline, he has
developed enormously fruitful collaborations with academics
in other disciplines. And with Duncan being Duncan, these
collaborators have often ended up as close friends. One life-
long friend was the late Professor Brian Hahn, a mathematics
professor. Brian and Duncan met as postgraduate students
when they started a Christian newspaper on the UCT campus.
Many years later they developed an innovative course to in-
troduce first-year students to chemical engineering, with Brian
contributing his amazing expertise in the teaching of math-
ematical modeling. Many of these innovations were worked
out on the slopes of Table Mountain, with Brian and Duncan
using their shared love of running to get some free space to
think and discuss ideas together. Students would be somewhat
surprised to see their lecturer running past the cafeteria in
jogging shorts over lunchtime but soon got used to it.
Another generative friendship-cum-academic collaboration
has been with Professor Cedric Linder, who holds professor-
ships at Uppsala University in Sweden and the University of
the Western Cape in Cape Town. Duncan and Cedric met while
walking their dogs in a local Cape Town park, and discov-
ered that there was considerable synergy between Duncan's
educational approach and that which Cedric was trialing
with his physics students. Cedric later introduced Duncan
to the arcane mysteries of phenomenography, an approach
to studying student learning that they have both put to good
use in researching their courses. Their research collaboration
was facilitated by a grant from the Swedish-South African
Intergovernmental S&T Co-operation Programme.
Duncan's greatest inspiration comes from working with
students. He says that at least half the students in the class
are brighter than him, but he at least has the advantage of
experience. He enjoys being challenged and stretched by
these young people. On a lighter note, their sense of humor
also keeps him on his toes. One day while running a design
project where he did what he thought was a superb role play of
a difficult and demanding boss, he was somewhat tickled when
halfway through the afternoon one of the students requested
"Can we have Professor Fraser back please?"
WIDENING THE NETWORK
Duncan feels that SouthAfrica, with its unique mix of first-
world and third-world contexts, has a particular role to play
in encouraging the development of engineering education in
Africa. He has built up an impressive network of engineering
educators across the continent, partly built from his contacts as
an external examiner in Tanzania, Kenya, Ghana, and Zambia,
and also through his active participation in engineering edu-
cation conferences in South Africa, Nigeria, and Cameroon.
He has been centrally involved in efforts to formalize these
networks into an association, and has been recently selected
as secretary general of the new African Engineering Educa-
Dr. Russel Jones, chair of the World Federation of Engi-
neering Organization's Committee on Capacity Building,
explains Duncan's modus operandi: "In his quiet but effec-
tive way, Duncan has led reform of engineering education
both at his home institution and well beyond throughout
sub-Saharan Africa. He works by providing a role model for
faculty colleagues, and by demonstrating what can be done by
a dedicated and talented individual faculty member." Cedric
Linder says the following of Duncan: "At a personal level
he is strong-willed, but with a gentle open-minded manner
that has been particularly useful in helping open doors to
pursue new directions in higher education transformation."
This is the key aspect of Duncan's work. Although he net-
works across the globe, his work is solidly rooted in his own
classroom experience. Right now if you pop into Duncan's
famously cluttered office on the second floor of the Chemical
Engineering Building at UCT, you will see him hard at work
preparing a holiday class for students who are having a second
shot at mastering the chemical engineering basics. Come a bit
later and you will see a stream of students outside his door
asking for advice on anything from solving mass balances to
obtaining a bursary for their studies to coping with a death
in the family. Russel Jones describes him as an "educator's
educator," someone who has built a career out of a passion-a
rare privilege indeed. 7
Chemical Engineering Education
STUDENT TECHNICAL CONFERENCES
To Build Multidisciplinary Teamwork Skills
DAVID L. SILVERSTEIN
University of Kentucky * Paducah, KY 42001
Accredited chemical engineering programs in the
United States continue to face the issue of how to
assess "soft skill" outcomes in their curriculum,
including the ability to function on multidisciplinary teams,
communicate effectively, and engage in lifelong learning.il' Of
these three, perhaps the most obvious to address is the com-
munication outcome. The other two require a little more effort,
not only to achieve the outcome but to define what it means.
The lifelong-learning criterion seems most often interpreted
to mean "give students the ability to learn independently,"
meaning make them go to the library and teach themselves.
Others extend this concept, suggesting that not only should
they be able to locate information, but they should be able to
learn from their peers. Supporters of collaborative learning
strongly endorse this concept.
Programs also need to address the "multidisciplinary
teams" criterion, which first requires a definition of what a
multidisciplinary team is supposed to be. In some programs,
multidisciplinary refers to students with different degree ma-
jors collaborating on a single project. This requires a course
involving such students, or some other method of bringing this
diverse group t g tiL ~1 1'' Obviously, this can be challenging
at most institutions, since the requirement must be fulfilled in a
required course in the chemical engineering curriculum. Oth-
ers consider a team project that gives each student a distinct
role, function, or discipline to apply as fulfilling requirements
for the outcome.[8 This is more readily accomplished and is
the method that appears to be most commonly adopted. In
both cases, teamwork training is recommended. Not only is
this outcome important for ABET purposes, but industry also
considers teaming skills as critical.9, 10]
� Copyright ChE Division of ASEE 2007
Vol. 41, No. 2, Spring 2007
Additionally, the AIChE Annual Meeting often causes can-
cellation of a week's worth of chemical engineering classes
during the fall term. Many students will also participate in
the National Student Conference the weekend before, miss-
ing another day or two of classes. Instructors, already under
time constraints in most courses, often attempt to redeem
the otherwise lost time by assigning extra homework,
reading, or short-term projects to keep students engaged
during the week.
With both the need to address difficult ABET Engineer-
ing Criteria outcomes and lost class time in mind, a novel
student project was created to develop student skills while
taking advantage of student participation in conferences. The
task also engages those who do not attend such conferences.
Chemical engineering students at the University of Kentucky
Extended Campus Program in Paducah, Ky. ,11 were assigned
this project in the fall semesters of 2003 and 2004.[121
The key feature of this project is that students are placed
in teams that span courses across years of the curriculum. In
other words, sophomores, juniors, and seniors are placed on
David L. Silverstein is an associate profes-
sor of chemical and materials engineering
at the University of Kentucky College of
Engineering Extended Campus Programs in
Paducah. He received his B.S.Ch.E. from the
University of Alabama in Tuscaloosa, Ala.;
his M.S. and Ph.D in chemical engineering
from Vanderbilt University in Nashville; and
has been a registered P.E since 2002. Sil-
verstein is the 2004 recipient of the William
H. Corcoran Award for the most outstanding
paper published in Chemical Engineering
Education during 2003.
a single team. This team structure assures a multidisciplinary
functionality since the capabilities of team members to con-
tribute to a technical project vary distinctly from class to class.
The teams are formed to be balanced according to class stand-
ing, and then according to academic ability. Since the classes
engaged in this project are small, no formal method for dividing
teams was required. A more promising approach to grouping
students in larger programs is proposed by Newell, et al. 131
The premise of the project is that each team consists of new
hires in a startup company conducting business in an emerging
area of chemical engineering. The first two years, the fictional
companies were involved in biotech and nanotech enterprises.
There is, however, one problem. Despite a wealth of venture
capital and high salaries, management is fatally confused.
They are not certain exactly what product or service they are
offering. The team is charged with the task of defining that
Student Research Project
Due: 5PM Monday, November 15, 2004 |
Cross (Curcular Assignment
You've Got Work To Do
C (n11:l1 at 1l i'l n' You have your first job after com'ni: lin.i your decades of formal education. Unfortunately,
you have landed that job with a startup company which does not have a firm sense of what it does to make
money. You do know that it is focused on nanotechnology, dealing with things like carbon nanotubes, MEMs,
nanoparticles, nanosensors, or other things nano-.
As part of a multidisciplinary project team, you are going to collect information required to move your
company forward and make it a competitive force in its field- whatever that specifically may be. Those of you
attending the conference will 11- h, r information from exhibitors and from technical sessions that are tied to
your specialization. Those remaining home will obtain similar information via the web and from the technical
You team will prepare a summary report for your chief executive officer which will contain a
recommendation for a nano-related product to produce including objectives identified for obtaining
information; identification of key elements in current knowledge on the cutting-edge substance or process; and
identification of equipment, software, or other items which will contribute to your company's efforts. Specific
requirements for the report are included in the rubric on the last page. Your role (or roles) is (are) based on the
classes you are enrolled in and are summarized below. If you are enrolled in multiple classes listed below, your
responsibilities will increase.
Due to the size of the teams, each report can contain up to 12 paper summaries, so staying organized as a group
is important. Those attending the conference should use the online program to plan their strategy ahead of time
and establish their company .l-I i.- r I\ * prior to departure. Those not attending the conference should be able to
collect their data during the conference. Upon return, all team members should finalize their summaries and
work t:ii- ethir tu compile a single report. Make certain you reference all summaries of presentations or journal
articles using end notes.
Be creative, and have some fun with the project, but do keep within the scope of the project. You are actually
supposed to learn something valuable!
Deliverables: Your team will turn in four copies of a single report with the names of all team members. One
copy will go to each CME faculty member. Grading will be performed by the instructors) of the classes) for
which you receive a grade. Grading criteria may vary somewhat from class to class. The instructor of your
class retains the final authority to determine how a grade for this assignment will apply to your class.
Peer evaluation surveys of team 1,.i r r : cl 11 i,: i- will be submitted individually and used to assign individual
grades based on the team grade. Failure to contribute adequately to the team report will result in significant
reduction of individual grades.
Figure la. Page one of the project assignment from the second offering.
Chemical Engineering Education
product or service, and then to: The concise version of the assignment is that the team
identifies a fictional objective, each team member contributes
Prepare a summary report for your chief executive officer a very brief summary of two journal articles or conference
that will contain a recommendation for a nano-related
(or biorelated) product to produce or service to offer, papers related to the objective in some way, and each mem-
(or bio-related) product to produce or service to offer, .
including: objectives identified for obtaining information; ber identifies a vendor that provides a product or service that
identification of key elements in current knowledge on the would also contribute to the company's objectives. The topics
, ,,,,,. ,i.e product or process; and identification of equip- summarized and vendors identified should be tied to students'
ment, software, or other items that will contribute to your current courses in some way. The complete assignment is
company's efforts. given in Figures la and lb.
CME 200- As a person currently focused on fundamentals, you will need to identify products and
processes of interest. General summaries of research involving phase equilibria, or mass & energy balances
are a plus. Identify resources which may be of general assistance in developing a top notch nano-engineering
department for your company. You may not understand much of what you see, but a brief overview or
description should be enough.
CME 470- As a safety expert, you need to be knowledgeable of all aspects of your company's technology.
Identify nano-topics that provide a basis to conduct risk analysis to ethically protect the safety of your
company's professionals. Since you have additional expertise in separations, fluid mechanics, and reactor
design, you may also identify information useful when considering those areas of responsibility.
Team 1 Team 2
Brent Todd �
Jessica Drew *
* Indicates Project Leader. The project leader will coordinate efforts among teams, maintain the common
report file, and arrange necessary team meetings. Successful execution of this role will increase the individual
grade for this person.
Meet as a group to select your scenario well in advance of the conference. Choose your topic based on the
sessions available at the conference on Monday morning. Those obtaining journal articles can conform more
easily to the topic than those using conference papers. Express your topic in conjunction with a brief list of
cbli -.: i \ : _ that will guide the team in writing their summaries. This should take about an hour if everyone
arrives prepared. You don't have to be exceedingly specific, but you should be consistent. The conference
program is online at http://www.aiche.org/conferences/techprogram/date.asp?Day=Monday&DSN=annual04.
Take enough notes at the conference to be able to summarize the topic and tie it to your team objectives. You
are only expected to write a few sentences to a paragraph on each paper. The report may have, say, one
paragraph on separations papers, one on general chemical processes, one on reactor design, one on process
design, and a few on equipment vendors (in addition to appropriate introductions, objective statement, and
After the conference, everyone should write their summaries on their own and then send them to their Project
Leader. The Project Leader should combine them and prepare an intru,.lt.. I lun in: lu inl, the team objectives.
Gather your group It.-.r Ilmr for a xu rinne - ilinc session. Prepare your final report for submission before the
deadline. This part of the process should take no more than three hours.
Figure lb. Page two of the project assignment from the second offering.
Vol. 41, No. 2, Spring 2007 83
The assignment objectives are that the students:
* Develop a list of objectives to meet project outcomes.
* Write a coherent, concise, and high-quality report as a
* Compose referenced summaries of information relevant to
a project task.
* Function effectively as a multidisciplinary team to collect
* Identify current research related to project objectives.
* Identify vendors that produce products suitable for project
* Describe the role of biotechnology or nanotechnology in
modern engineering practice.
To accomplish this task, each team is appointed a leader
who is expected to arrange team planning meetings, facilitate
determination of the goals for the company, and coordinate
the information team members contribute toward the objec-
tives. Additionally, the leader arranges the final composition
of the report. Since this last item is a substantial task, leaders
have the option of "hiring" an editor from the team, who will
assist with this task and receive compensating
"bonus" credit for the project. No team has
used an editor to date. The team leaders are Since one
usually selected from the senior class mem- of the pr
bers who do not typically take on leadership reduce
roles but are believed by the instructor to
have the ability to lead. They are given more due to c
specific guidance, training, and instruction one ofth
prior to the start of the project. missed or
Students are assigned this task as part of the class mt
courses in which they are already enrolled. allo
Cooperation is secured from all instructors this
required to ensure participation of all three
classes (sophomore to senior). The instructors
of these courses determine how to apply the
project to their grade computations, but typically the report
counts for one or two homework assignments or as a fixed
percentage of the total grade (~5%). Additionally, the instruc-
tors of the courses from which team members are drawn
can grade the reports on their own, or use the grading of the
project faculty coordinator. To date, no faculty member has
asked to grade the reports a second time, choosing to use the
grade assigned by the coordinator.
The multidisciplinary aspect of this project is tied to the
courses in which the students were enrolled. The number of
courses involved depends on the minimum required to secure
participation from most sophomores, juniors, and seniors
(this was four courses in the first offering, but only two in
the second). For example, during the first offering, the topic
was biotechnology. Students enrolled in the following courses
participated with the course-specific assignment:
* Process Principles (sophomores): As a person cur-
rently focused on fundamentals, you will need to
identify products and processes of interest. General
summaries of research involving phase equilibria, or
mass & energy balances are a plus.
* Separations (juniors): If it's mixed up, you're the, um,
unsolution. You should identify research and equipment
associated with ,. . i" ,1 ,, different materials.
* Process Design I (seniors): Elements of process design
and simulation are your forte. You should include simu-
lation software in your .-, i,,, .*- . especially ones
that include economic analysis (especially ",. i ,1, ,").
* Reactor Design (seniors): If it reacts, it's your busi-
ness. Determining kinetic laws, sizing and designing
reactors, and. ,,t ,, m,, chemical reaction with other
processes are among the topics that you are concerned
with. Simulation at the molecular level may also float
your chemical engineering boat.
For the first offering, seniors were given this assignment in
two courses, but the assignment was limited to one course dur-
ing the second year of the project. The change resulted from
the determination that the workload was too heavy on senior
students in the first implementation. Students
were to select topics for their research that
he goals they could tie to the course in which the as-
t was to signment was made.
t was to
Since one of the goals of the project was
to reduce "lost" time due to conferences, one
rences, of the otherwise missed or rescheduled class
herwise meetings was allocated to this project. Time
scheduled spent per student on the project was intended
gs was to be 3-6 hours, not including training. Treat-
Sing the project as a laboratory exercise, this
corresponds to a lecture class time loss of 1-2
ect. hours, which is typical during the AIChE An-
nual Meeting week.
As part of the assignment, students were
provided a grading rubric to make expectations clear and to
guide them on their writing. Newell, Newell, and Dahm
provide guidelines for rubric development appropriate to this
sort of project. The rubric used in this project is provided in
Students are given creative freedom to define their objec-
tives to take advantage of available resources. This approach
differs from one concerned with developing problem-solving
skills due to the constraints associated with the conference
presentation element of the project. Since those students
attending the conference are required to summarize two
presentations, the availability of appropriate sessions on the
Monday of the conference (their last full day at conference)
is the limiting factor in their completion of the project. Con-
sequently, prior to the conference, students are directed to
the AIChE technical program online to identify presentations
Chemical Engineering Education
suitable to define objectives. Since the student conference
usually conducts 90-minute overview sessions on emerging
areas in chemical engineering, multiple students in each
group are allowed to summarize part of that session to fulfill
one of their technical summary requirements. Additionally,
students attending the conference are required to identify their
vendor from among those exhibiting at the conference. This
requires students to define their objectives carefully based
on available resources-a useful skill in dealing with poorly
Those attending the conference typically spend about two
extra hours at the conference attending technical sessions
and visiting exhibitors, still leaving significant time for
sightseeing and other activities. Those remaining home use
library resources to obtain their technical summaries and the
Internet to find vendor information. During the days follow-
ing the conference, teams are expected to meet and combine
their summaries into a coherent paper meeting assignment
objectives. Each team is required to submit its paper on the
Monday following the conference.
0C1... i.- . i t -h 0-unacceptable 1-marginal 2-acceptable 3-excellent
Write a coherent, Not all objectives Addresses all objectives, Addresses all objectives, Addresses all objectives,
concise, and high addressed, significant spelling, some notable spelling, well-written with very
quality report as a grammar, punctuation, grammar, punctuation, minor spelling,
team (20%) and style issues. and style issues, grammar, punctuation,
and style issues.
Compose referenced Most summaries fail to Most summaries include All summaries include All summaries include
summaries of include references to the references to the source references to the source references to the source
information relevant source paper or paper or presentation paper or presentation paper or presentation
to a project task (10%) presentation. End notes with minor with minor using a consistent end
style inconsistent, inconsistencies in end inconsistencies in end note style.
note style, note style.
Function effectively Some specified subject All specified subject All specified subject All specified subject
as a multidisciplinary areas not included in the areas included in the areas included in the areas included in the
team to collect report. Some team report and but not tied report and tied together report and lied together
relevant information members fail to fully together by a common by a common concept by a common concept
(15%) participate. concept (product or (product or process) with (product or process). All
process). Some team some inconsistencies, team members
members fail to fully All team members participated in all
participate, participated, aspects of the project.
Develop a list of Report indicates no Report indicates the plan Report indicates a Report indicates a
objectives to meet advance planning to prepared to obtain loosely structured plan well-structured plan
project outcomes obtain information information required for prepared to obtain prepared to obtain
(109%) required for the report, the report was minimal information cohesive information
inadequate. required for the report. required for the report.
Identify current Fewer than 2 unique Summaries of at least 2 Summaries of at least 2 Summaries of at least 2
research related to papers (conference unique papers unique papers unique papers
project objectives presentations or journal (conference (conference (conference
(25%o) articles) included for presentations or journal presentations or journal presentations or journal
each person for each articles) included for articles) included for articles) included for
participating course, each person for each each person for each each person for each
Some topics are participating course, participating course. participating course.
inconsistent with the Some topics are Topics are not Topics are consistent
team objective. inconsistent with the necessarily consistent with the team objective.
Summaries may be team objective. with the team objective. Summaries will be one
incomplete and fail to Summaries may be Summaries may not be paragraph per paper, be
establish relevance. incomplete and fail to complete and establish complete, and lie the
establish relevance. relevance, topic to the team
Identify vendors Descriptions of fewer Descriptions of at least 1 Descriptions of at least 1 Descriptions of at least 1
which produce than one unique product unique product or unique product or unique product or
products suitable for or service included for service included for each service included for each service included for each
project requirements each person for each person for each person for each person for each
(20%) participating course, participating course, participating course, participating course.
Items are consistent Items are not necessarily Items are consistent
with the team objective, consistent with the team with the team objective.
Several descriptions will objective. Some Descriptions will be
be incomplete and fail to descriptions may be complete, and tie the
tie the item to the team incomplete or fail to tie item to the team
objective, the item to the team objective.
Figure 2. Rubric distributed to students and used for project grading.
Vol. 41, No. 2, Spring 2007
The first year of this project, students completed post-proj-
ect surveys. For the second offering, students were asked to
complete both pre- and post-project surveys. Summaries of
the results for the two years combined are given here.
* Among those o i,, .in, the conferences, nine had not
attended technical sessions prior to this project, four
had attended such sessions. Afterward, all had attended
conference technical sessions.
* Prior to this assignment, seven hadpreviously located
articles in the literature, five had not. All had done so
after the project.
* Prior to this assignment, six hadpreviously identified
vendors for engineering products or services, six had
not. All had done so after the project.
* This project was thefirst time working with some of
their teammates for all but four students p.i, i,. q'1 ,,1 ,
* Of26 respondents, 22 indicated they assisted other
students with decisions they needed to make to complete
* Five students indicated they spent 0-3 hours on the proj-
ect; 13 said 4-6 hours; and nine said 6 or more hours.
The average self-reported time spent on the project was
about 6 hours in both years.
In the second year of the project, students were surveyed
both before and after the project. Table 1 summarizes the re-
sults, which indicate that students did make significant gains in
knowledge and lifelong-learning capability, with more modest
gains in their perceived ability to work in teams.
Students were also asked to name the best and worst ele-
ments of the project. The most popular responses for best
element included learning about topics not covered in the
curriculum and interacting with other classes. The worst
elements included poor student leadership, confusion about
the project (mostly in the first year), and the time required
for the project.
Instructor concerns prior to assigning the project included
the amount of grading. With a team size of about seven stu-
dents, however, the number of reports to grade was limited.
The use of the aforementioned rubric also simplified the grad-
ing process. A grade sheet for each student, with adjustments
for peer evaluation and for leadership, was provided to each
class instructor for recording and distribution to the students.
The confusion issue was also a great concern and was ad-
dressed in part by providing students in the second year with
successful examples of reports from the previous year. One
mistake made the first time this project was assigned was not
providing teamwork training to the students. This has been
rectified through a program held through the AIChE student
chapter prior to the assignment's distribution. Additionally,
library training sessions were provided in the second year,
along with focused training for team leaders and distribution
of background materials to each team on that year's topic.
Participation of other faculty in the department is also a key
concern. Changing an existing course requires effort on the
part of the instructor to integrate the project into the course.
That effort would result in little benefit to the instructor,
so there needs to be "buy-in" to the efficacy of the project
significantly improving achievement of program outcomes.
Additionally, concern regarding the project contributing to
achieving course outcomes has been raised, particularly in
the courses involving sophomores. The load on the faculty
member coordinating the assignment is about four contact
hours (teamwork training, library training if not provided
elsewhere, and project organization meetings) plus prepara-
tion time, grading time for reports, and time for meeting with
student leaders to address their concerns and questions.
The assessment of teamwork proved unsatisfying to the
instructor, consisting of the third item on the rubric (Figure
2), review of student peer evaluations, and review of student
project evaluations. Other assessment methods for teamwork
are suggested in the literature and should be considered for
the next offering. 15161
A project to vertically integrate chemical engineering
students into a multidisciplinary team was successful in
developing an introductory understanding of emerging areas
in chemical engineering. Students experienced the pain of
multidisciplinary teams as they successfully completed a
Summary of Student Responses to Pre- and Post-Project Survey Questions in the Second Year of the Project
Students were asked to respond to a set of questions and indicate their agreement according to a five-point Likert scale, where 5 indicates strong
agreement and 1 indicates strong disagreement. Sample size was eight students.
Question Pre-Project Average Post-Project Average
(Std. Dev.) (Std. Dev.)
I work well with teams. 3.625 (1.69) 4.000 (1.31)
I know the relevance of nanotechnology to chemical engineering. 1.875 (1.13) 3.875 (1.55)
I can find the technical information I need in chemical engineering from the literature. 3.250 (0.89) 4.125 (0.64)
I know what is meant by "the literature." 2.750 (1.49) 4.125(1.13)
I know what nanotechnology means. 2.875 (1.36) 3.625 (1.69)
Chemical Engineering Education
report consisting of referenced summaries of technical papers
and identification of vendors of products and services, all tied
to objectives the team previously developed and the courses
in which they were enrolled. The project made contributions
to program outcomes in communication, lifelong learning,
multidisciplinary teamwork, and contemporary issues. An ad-
ditional benefit was the increased interaction among students
in a small, nontraditional chemical engineering program.
1. ABET Criteria for accrediting engineering programs: Effective for
evaluation during the 2004-2005 accreditation cycle.
20-03.pdf>, accessed Jan.5, 2005
2. Wankat, PC., ES. Oreovicz, WN. Delgass, "Integrating Soft Criteria
into the ChE Curriculum," Proceedings of the 2000 American Society
for Engineering Education Annual Conference & Exposition (2000)
3. Felder, R.M., and R. Brent, "Designing and Teaching Courses to Satisfy
the ABET Engineering Criteria," J. Eng. Ed., 92(1), 7 (2003)
4. Miller, R.L., and B.M. Olds, "A Model Curriculum for a Capstone
Course in Multidisciplinary Engineering Design," J. Eng. Ed., 83(4),
5. Fornaro, R.J., M.R. Heil, and S.W. Peretti, "Enhancing Technical
Communication Skills in Engineering Students: An Experiment in
Multidisciplinary Design," Proceedings of the 3 IstAnnualASEE/IEEE
Frontiers in Education Conference, S2G-1, Reno, NV (2001)
6. Newell, J.A., S.H. Farrell, R.P Hesketh, and C.S. Slater, "Introducing
Emerging Technologies in the Curriculum Through a Multidisciplinary
Research Experience," Chem. Eng. Ed., 35(4), 296 (2001)
7. Glennon, B., "Development of Cross-Disciplinary Projects In a ChE
Undergraduate Curriculum," Chem. Eng. Ed., 38(4), 296 (2004)
8. Shaeiwitz, J.A., and R.Turton, "Lifelong Learning Experiences and
Simulating Multi-disciplinary Teamwork Experiences through Unusual
Capstone Design Projects," Proceedings of the 2003 American Society
for Engineering Education Annual Conference & Exposition (2003)
9. Bhavnani, S.H., and M. DayneAldridge, "TeamworkAcross Disciplin-
ary Borders: A Bridge Between College and the Work Place," J. Eng.
Ed., 89(1), 13 (2000)
10. Katz, S.M., "The Entry-Level Engineer: Problems in Transition from
Student to Professional," J. Eng. Ed., 82(3), 171 (1993)
11. Smart, J.L., W. Murphy, G.T. Lineberry, and B. Lykins, "Development
of an Extended Campus Chemical Engineering Program," Proceedings
of the 2000 ASEEAnnual Conference & Exposition. American Society
for Engineering Education (2000)
12. Silverstein, D., "Making Student Conferences an Assessable Learning
Opportunity," Proceedings of the 2005 ASEE Annual Conference &
13. Newell, J., K. Dahm, R. Harvey and H. Newell, "Developing Meta-
cognitive Engineering Teams," Chem. Eng. Ed., 38(4), 316 (2004)
14. Newell, J.A., H.L. Newell, and K.D. Dahm, "Rubric Development for
Assessment of Undergraduate Research," Chem. Eng. Ed., 38(1), 68
15. Lewis, P, D. Aldridge, and PM. Swamidass, "Assessing Teaming Skills Ac-
quisition on Undergraduate Project Teams," J. Eng. Ed., 87(2), 149 (1998)
16. Shaeiwitz, J.A., "Observations on Forming Teams and Assessing
Teamwork," Proceedings of the 2003 ASEE Annual Conference &
Exposition (2003) 1
Use CEE's reasonable rates to advertise.
Minimum rate, 1/8 page, $100;
Each additional column inch or portion thereof, $40.
Johns Hopkins University
The Department of Chemical and Biomolecular Engineering at
Johns Hopkins University invites applications for a full-time lec-
turer. This is a career-oriented, renewable appointment. Responsi-
* Teach 3 courses each semester (currently with labs).
* Manage curriculum issues, including degree requirement
updates and course development.
* Coordinate advising for undergraduate Chemical and
Biomolecular Engineering majors.
* Organize prospective freshmen activities, including open
houses and welcome letters, and serve as liaison to the
* Oversee and train graduate TAs and graders.
* Maintain retention and growth statistics.
Johns Hopkins is a private university well known for its commitment
to academic excellence and research. The Department of Chemical and
Biomolecular Engineering is among the first rank of departments in the
Whiting School of Engineering in terms of funded research activities
and educational programs. We are located in Baltimore, MD, often
referred to as "Charm City," in close proximity to Washington, DC
and Philadelphia, PA. See the departmental Web page at http://www.
jhu.edu/-cheme for additional information about the department,
including undergraduate programs and course descriptions.
Applicants must have a Ph.D. in Chemical Engineering or a closely
related field, and demonstrated excellence in teaching. Applications
must include a letter of application, curriculum vitae, and a statement
of teaching philosophy. Applicants should arrange for three reference
letters to be sent directly to the address below. All material should
arrive by May 30, 2007.
Lecturer Search Committee
Chemical and Biomolecular Engineering Department
Johns Hopkins University
3400 N. Charles St, 221 MD HALL
Baltimore, MD 21218
Johns Hopkins University is an EEO/AA employer. Women and
minorities are strongly encouraged to apply.
Vol. 41, No. 2, Spring 2007
---- U s_____________________________________
A POPULATION BALANCE BASED
in a Particle Science and Technology Course
SHERYL H. EHRMAN, PATRICIA CASTELLANOS, VIVEK DWIVEDI, AND R. BERTRUM DIEMER*
University of Maryland * College Park, MD 20742
There is a great need for chemical engineering students
to have some exposure to particle technology concepts,
since so many products are processed and handled
in particle or powder phase. For example, more than half
(by volume) of all products sold by DuPont and Dow are in
particulate form. Considering raw materials, intermediates,
and co-products, the amount of solids handled is actually
three to four times the amount sold."1 In addition, there are
increasing opportunities for chemical engineers with some
particle technology training in fields outside of the traditional
chemical process industry, such as sensing technology devel-
opment for homeland security, aerosol drug formulation, and
device development and nanomaterials processing. Over the
past decade, there has been a push to incorporate more par-
ticle technology into core chemical engineering coursework,
and, as recommended by Nelson, et al.,11" to include elective
courses in particle technology. At the University of Mary-
land, a course designed for senior-level undergraduates and
first-year graduate students was created and taught initially
Gas to Recovery
0.25 pm particles
Figure 1. Schematic of aerosol generation and particle
* DuPont Engineering Research and Technology, DuPont Company,
Wilmington, DE 19898
in 2002. The course spans "dry" particle technology from
aerosol science to powder technology unit operations.
To bridge the gap between fundamental aerosol science
and powder technology, a team design problem incorporating
population balance modeling and design of particle collection
systems was developed and assigned to students in our Particle
Science and Technology course. The main motivation for de-
veloping this assignment was to provide a realistic open-ended
industrial problem for the students to solve that would incor-
porate population balance modeling and aerosol dynamics to
describe evolution of particle size distributions. The problem
was developed collaboratively between a representative from
industry and a University of Maryland faculty member, with
additional input from the graduate student teaching assistant
for the course. The problem was assigned twice (spring 2003
and fall 2005), with changes made in the second version based
on feedback received from students in the first. Students in
the fall 2005 course provided additional feedback.
In this problem, students were given a description of a
particle synthesis and collection system as shown in Figure
1. The model particles, with properties similar to those of
Sheryl H. Ehrman is an associate professor in the Department of
Chemical and Biomolecular Engineering at the University of Maryland.
She earned B.S. and Ph.D. degrees in chemical engineering from the
University of California, Santa Barbara, and the University of California
Los Angeles, respectively.
Vivek Dwivedi is a graduate student in the Department of Chemical
and Biomolecular Engineering at the University of Maryland. He has
a B.S. and an M.E. in chemical engineering, both from the University
Patricia Castellanos is a graduate student in the Department of
Chemical and Biomolecular Engineering at the University of Maryland.
She has a B.S. in chemistry from the University of Maryland.
R. Bertrum Diemer is DuPont's principal division consultant in
reaction engineering, thermochemistry, and population balance
modeling and an adjunct professor of chemical engineering at the
University of Delaware. He received chemical engineering degrees
from Lehigh (B.S. '73) and Delaware (M.S. '80, Ph.D. '99) and is a
� Copyright ChE Division of ASEE 2007
Chemical Engineering Education
I and length?
Mb o a
silica particles, were produced in an aerosol reactor. One major simplifying assumption was made:
The size distribution of the particles leaving the reactor was assumed to be monodisperse. The particle
collection system consisted of a pipeline agglomerator, a cyclone, and a baghouse. The objective of this
design was to reduce wear on the baghouse by collecting 75% of the particle mass in the cyclone. Use
of the cyclone alone was not feasible for the conditions specified in the problem statement, because
standard cyclones typically are not very efficient at collecting sub-micron-diameter particles. Hence
the coagulation tube was added prior to the cyclone to promote formation of larger agglomerates that
could be collected more efficiently by the cyclone. In both implementations of the problem, in 2003
and 2005, students focused on the pipeline agglomerator and the cyclone in their system design.
Within the pipeline agglomerator, coagulation and breakup were assumed to be the only processes
occurring. Assuming steady state, axisymmetric incompressible plug flow, with particle size distribu-
tion dynamics varying only along the z axis of the pipeline agglomerator, the simplified form of the
aerosol general dynamic equation, expressed in the form of a population balance, is:
u (z- ) fv( ,V- )nK()n(V_- )dy -n(V) f (J,V)nK()dY
Oz 2 0 0
+f r(v)b(V;t)n(t)d@ -F()n(V) (1)
where u is the axial velocity, n is the particle population density function distributed by particle volume
in units of number concentration, z is the axial position, V is particle volume, F is the breakage rate
kernel, b is the parent particle volume in breakage, b is the breakage daughter distribution, and 3 is
the coagulation rate kernel. The left-hand side describes the evolution of the volume distribution as
a function of the distance down the axis of the pipeline agglomerator. The first and second terms on
the right-hand side represent changes in the size distribution because of agglomeration, and the last
two terms describe changes to the size distribution resulting from breakup.
Under the conditions specified in the problem statement, Brownian diffusive transport of particles
in the z direction was assumed to be negligible. Collisions were assumed to be 100% efficient, with
no influence from long-range forces. Coagulation was assumed to result in soft agglomerates, i.e.,
no sintering of the particles occurred. The coagulation rate kernel was represented as the sum of the
Brownian coagulation and the Saffman-Turer rate kernels as:
S 1/V T3 [V 13
2kT 1" ' �
3[ ,- = +NM
-0.31 V-+3 13(V- )23 3 (V )- (2)
where k is the Boltzmann constant, T the temperature, g is the fluid viscosity, v is the kinematic
viscosity, calculated by dividing the fluid viscosity by the density of the total flowing stream (fluid
plus particles), and E is the energy dissipation per unit mass. Breakage was assumed to result in two
equal-sized daughters, and this assumption greatly simplified the sectional approach. The breakage
kernel is given as:
r(V)=r 4 v13 (3)
The daughter distribution is given by:
b(V;4)= 26V 1
where delta = is the Dirac delta function, and V and ) are defined as described as above.
In discrete form, the population balance Eq. (1) can be expressed in terms of summations
uz = - n- - n ,.n + F ,+, + 2F,,n,, - F,, 1n,, - F,n, (5)
where n denotes the number of particles of volume i, and n, the number of particles of volume j.
Vol. 41, No. 2, Spring 2007
than in 2003.
some of the
in the industry
Unfortunately, approximately 2 X 106 ordinary differential equations would be required to describe
the anticipated range of particle volumes in this system. To simplify the numerical implementation
of the solution, a geometric sectionalization method was used with a geometric ratio of 2.12 3] Here,
the geometric ratio is defined as
V+1 = 2 (6)
where v is the lower bound of the ith volume interval, v+1 is the lower bound of the next greater vol-
ume interval, and q is the discretization parameter, taken in this case to be 1. This method works well
with the binary equisized daughter distribution assumed for particle breakage, and the method has
been shown to capture particle number and particle mass balances correctly. It should be noted that
for values of q, greater than 4, corrections to the original population balance discretization approach
of Litster, et al.,21 were published by Wynn. 31
The number of bins was left to the students, but it was suggested that the students cap the number
at 28 bins. Detailed comparisons between predictions of size distribution evolution made using this
sectional approach with 25 bins and predictions made using the quadrature method of moments and
the polynomial interpolative closure method of moments have been reported previously.
The population balance equations used to describe the change in the number of particles in bin i
(N) as a function of distance down the axis in the sectional approach are given by:
dN ,- NN -I N N N 1 1iN
Uz--' ' -l E N J + 1-1,4-1 .-1 i - N J-- 1.J
d -1 2N 2 2' Z J=1
with the collision kernel given by:
S(i,j)= 2 2 ( '/3+2( '/
0.31 I 3V, 2[ +32 2 3-'1
- 2F 1N 1 2- F N
2 20+2')/3 ) + 2` 1]
To convert to particle diameter from the volume distribution- since the particles are colliding at
low temperatures to form soft agglomerates-it was assumed that the agglomerates were fractal-like,
and therefore particle volume scaled with particle diameter, dp, as:
dp = do (9)
where do is the primary particle diameter, and Df is the fractal dimension, here assumed to be 1.8.
The design assignment followed lectures on particle size distributions and fractal aggregates, dif-
fusion, nucleation, coagulation, coalescence, discrete population balance modeling approaches, and
cyclone design by the course instructor (Ehrman). The industrial partner (Diemer) gave a lecture on
industrial applications of aerosols, and on sectional population balance approaches specific to the
First Implementation, Spring 2003
In the first problem assignment, in 2003, students were asked only to design the coagulation tube.
The cyclone grade efficiency, q , was given as:
1= d (10)
where the critical particle diameter, dp, or cut size, was 24 microns. The maximum pressure drop
allowed in the coagulation tube, 2 psi, was also given as a constraint. The students were restricted
Chemical Engineering Education
to integral values, in inches, for the pipe inner diameter. The
students were asked to write a program to solve the population
balance equations, using a program of their choice. Global
optimization of this problem was beyond the scope of the
course and so solution approaches involved trial and error
selection of pipeline agglomerator diameter and length, and
from the final size distribution, the cyclone mass capture
percentage was calculated from Eq. (10). This process was
repeated until a solution meeting all specifications and satisfy-
ing all constraints was identified.
Students worked on this problem in teams consisting of
three or four students. The students were given three weeks
to complete the problem. Students were given a chance to
ask the industrial partner questions in a conference call one
week before the project was due, and two teams participated
in this call. One team was successful at writing a code in
MATLAB to solve the population balance equations, and
they used a systematic approach to find pipeline agglomera-
tor dimensions that allowed for adequate agglomeration, yet
minimized pressure drop. The solution format was a brief
three-page industrial-type memo, addressed to the instructor
and industrial partner, and describing the final design and the
particle size distributions entering and exiting the cyclone.
Two teams were not able to finish their code in the time
allotted. These teams were asked to describe the solution
algorithm they would have followed if they had gotten their
code to work. Students were given the opportunity to submit
evaluations of the class at the midterm point anonymously
through our Web-based course-hosting software, and the fol-
lowing comment is representative of many of the student's
views about the project.
"I think the computer project was a bit over our heads. We
could program the ii,. .,- ,i. ,. : outlined by Dr. Diemer,
but it was more or less an exercise in MATLAB instead of
population balance modeling."
- anonymous student, spring semester, 2003
As a result of the project outcome, the project was modified
significantly before its second implementation.
Second Implementation, Fall 2005
The main difference with the second implementation is
that students were given a Matlab code to solve the popula-
tion balance equations, written by the group of students who
successfully completed the problem in 2003. To increase the
degrees of freedom in the problem, the design portion was ex-
panded to include the cyclone. Students were asked to design
the cyclone using ChemCad software, rather than being given
a grade-efficiency curve and cut size for a specific cyclone.
With the incorporation of the process simulator, the intent was
to add a secondary aspect to the project: giving students an
experience extending chemical engineering process simula-
tion software to particle technology. Students were asked to
minimize the area of the pipeline agglomerator and cyclone,
as a surrogate for minimizing capital costs. The students
were given a constraint of a maximum pressure drop in both
pipeline agglomerator and cyclone of 2 psi. The preparation
for this implementation was similar to the first one, with one
exception. Prior to Dr. Diemer's lecture, the teaching assistant
(Dwivedi) gave a hands-on tutorial on cyclone design using
ChemCad, in our computer classroom. Teams were assigned
such that at least one person on a team had prior ChemCad
experience. Prior to this project, however, these students
had not previously worked with multiphase process streams
consisting of suspended particles, so a tutorial for all students
was a necessity.
An additional individual homework assignment was given
prior to the project being assigned: Students were required to
calculate the magnitude of each component of the coagulation
kernels over a range of particle diameters. The goal was to get
a feel for the relative importance of continuum Brownian vs.
turbulent coagulation, and how the relative importance of each
changed as particle size increased. Additionally, this assign-
ment gave some assurance that all students were comfortable
with the collision kernel calculations.
In the design problem itself, the basic solution algorithm
most groups followed was:
1. Choose a . , i , and diameterfor the pipeline agglom-
2. Run the MATLAB code to determine the size distribu-
tion at the end of the agglomerator.
3. Input the size distribution, as well as the cyclone size
specifications into ChemCad and calculate the compo-
sition and size distribution of the cyclone exit stream.
4. Calculate the overall pressure drop in the system and
check to see if the constraint of 2 psi is satisfied; if not,
return to step 1 or step 3.
5. Calculate the surface area of the agglomerator and
6. Ifpressure drop was ,,11f. ,11i, below 2 psi, reduce
cyclone or agglomerator dimensions and return to step 2.
There was one exception to the basic algorithm described
above. One group focused on the cyclone and designed an
extremely small cyclone, with a pressure drop of nearly 2 psi
that had high efficiency for micron-sized particles, and thus
they only required a very short pipeline agglomerator. Five out
of six groups were able to complete the project in two weeks.
The teaching assistant held additional office hours to assist
with the process simulator and the instructor held office hours
to assist students in interpreting and modifying the MATLAB
code. The students did not request a conference call with the
industrial partner. As in 2003, the solution was submitted in
the form of a brief three-page industrial memo containing the
details of the design, the size distribution of particles entering
Vol. 41, No. 2, Spring 2007
the cyclone in histogram format, and the size distribution and
solids mass fraction of gas leaving the cyclone.
Anonymous feedback in the second implementation was
generally more positive than in 2003. The students "enjoyed
learning about some of the related new developments in the
industry concerning major companies like DuPont," and they
enjoyed working on a realistic, challenging, open-ended prob-
lem. There was considerable relief that the MATLAB code
for the sectional solution to the population balance equations
was supplied. There was, however, one comment suggesting
that it would be better to "let students make [the] population
balance model by themselves."
With the modifications made in the second implementation,
we feel we have created an interesting and tractable design
problem. We intend to continue including this assignment,
with slight modifications, as an integral part of this course
in the future. Detailed 2003 and 2005 problem statements
as well as the MATLAB code, and a guide to ChemCad
and cyclones, are available at:
S.H. Ehrman acknowledges the support of a CAREER
award from the National Science Foundation (DMR 0093649)
supporting educational advances in the field of particle tech-
nology and materials processing.
1. Nelson, R.D., R. Davies, and K. Jacob, "Teach -em Particle Technol-
ogy," Chem. Eng. Ed., , 29(1), 12 (1995)
2. Litster, J.D., D.J. Smit, and M.J. Hounslow, "Adjustable Discretized
Population Balance for Growth and Aggregation, "AIChE Journal,
41(3), 591 (1995)
3. Wynn, E.J.W., "Improved Accuracy and Convergence of the Discretized
Population Balance of Litster, et al.," AIChE Journal, 42(7), 2084
4. Diemer, R.B., and S.H. Ehrman, "Pipeline Agglomerator Design as a
Model Test Case," Powder Technology, 156, 129-145 (2005) 7
Chemical Engineering Education
CALL * FOR * PAPERS
Fall 2007 Graduate Education Issue of
Chemical Engineering Education
Each year, CEE publishes a special fall issue devoted to graduate education.
It includes articles on graduate courses and research in addition to ads
describing university graduate programs.
We invite articles on graduate education and research for our
Fall 2007 issue. If you are interested in contributing, please send us your
manuscript as a pdf file, via e-mail, by the deadline below.
Deadline for manuscript submission is May 25. 2007.
Respond to: firstname.lastname@example.org
A Student-Centered Approach To Teaching
MATERIAL AND ENERGY BALANCES
1. Course Design
LISA G. BULLARD AND RICHARD M. FIELDER
North Carolina State University * Raleigh, NC 27695
. . Lisa G. Bullard received her B.S in ChE
ike most chemical engineering curricula, the chemical
and biomolecular engineering curriculum at North from NC State and her Ph.D. in ChE from
Carolina State University begins with a course on Carnegie Mellon. She served in engineer-
material and energy balances, historically designated the ing and management positions within
Eastman Chemical Co. from 1991-2000.
"stoichiometry course." For much of its history, the course At N.C. State, she is currently a teaching
was generally feared and despised by students, with their associate professor and the director ofun-
dergraduate studies in the Department of
descriptions of it invariably including the term "weed-out." Chemical and Biomolecular Engineering.
They were put off by the fragmented nature of the subject
matter, which appeared to be a hodgepodge of loosely related
concepts from physical chemistry. Test grades were low, fail- Richard M. Felder is the Hoechst Celanese
ure rates were high, and student course ratings were routinely Professor Emeritus of Chemical Engineer-
ing at North Carolina State University. He is
the lowest of any course in the curriculum. coauthor of Elementary Principles of Chemi-
in the late 1 s the p fiction f cal Processes, an introductory chemical
Beginning in the late 1970s with the publication ofElemen- engineering text now in its third edition. He
tary Principles of Chemical Processes,['] a new approach to has contributed more than 200 publications
the stoichiometry course was adopted at N.C. State. Individual to the fields of science and engineering
topics from physical chemistry were no longer presented on neering, and writes "Random Thoughts,"
a stand-alone basis, but were introduced and applied in the a column on educational methods and
issues for Chemical Engineering Educa-
context of chemical process engineering. The traditional tion. With his wife and colleague, Dr. Rebecca Brent, he co-directs
homework problems related to chemical and petroleum pro- the National Effective Teaching Institute (NETI) and regularly offers
teaching effectiveness workshops on campuses and at conferences
cesses were supplemented by problems from the growing around the world.
range of fields that employ chemical engineers, including
environmental engineering, biochemistry and biomedicine,
Vl Copyright ChE Division of ASEE 2007
Vol. 41, No. 2, Spring 2007 9.
This two-part series describes the structure of the stoichiometry course
at North Carolina State University. The course has a variety of learning
objectives, and several nontraditional pedagogies are used in the course
delivery. This first paper outlines the course structure and policies, the
preparation given to the teaching assistants (who play an integral part in
the course delivery), and the course assignments. The next one describes
the methods used for classroom instruction and assessment.
"I _ ________________________
and microelectronics. Most lectures included brief activities
that provided practice and feedback in the methods that would
be required on homework and tests. A 1990 paper outlined
the new instructional approach and described the turnaround
in student performance and evaluations that resulted from
The stoichiometry course has continued to evolve. Since
the early 1990s, it has been taught using cooperative (team-
based) learning, with measures being taken to hold all team
members individually accountable for the entire content of
team assignments. Instructional technology has played an
increasingly important role in the course, with a variety
of software tools supplementing traditional instruction.
Assignments include traditional closed-ended problems as
well as open-ended problems that call for creative or critical
thinking or both.
A goal of the N.C. State Chemical and Biomolecular En-
gineering (CBE) Department is to equip students to be self-
directed learners by the time they graduate. Our premise is
that they are a very long way from that status when they enter
the engineering curriculum and the best way to get them there
is to use what educational theorists call .. .,i/ 'iI,,. -initiilly
providing a great deal of structure and support, and gradually
withdrawing it over the course of the curriculum. Since the
stoichiometry course is the gateway to the CBE curriculum,
we set a high level of challenge in the course to give the
students a realistic picture of the intellectual demands that
await them in the next three years, but we also provide higher
levels of structure and support than most of them have ever
In the fall 2005 semester, 110 students enrolled in two
sections of the stoichiometry course. Although each of the
authors was primarily responsible for one lecture section, we
worked together closely to generate common course materials,
assignments, and tests, and we periodically guest-lectured in
each other's sections. This two-part series of papers outlines
the structure of the course and our approach to teaching it
and offers suggestions to faculty who might wish to adapt the
approach to their own teaching. Part 1 describes the course
design, and Part 2 summarizes the course delivery.
COURSE STRUCTURE AND POLICIES
The stoichiometry course at N.C. State, designated CHE
205, is a four-credit, one-semester course, structured as three
50-minute or two 75-minute interactive lecture classes per
week taught by a faculty member plus a two-hour weekly
problem session (recitation) conducted by a graduate teaching
assistant. Prerequisites for the course include two semesters
of calculus, two semesters of general chemistry, and one se-
mester of physics. The text is the 2005 edition of Elementary
Principles of Chemical Processes,"1 which comes bundled
with a CD containing several instructional resources and
a workbook that guides students through the solution of
selected chapter-end problems. The course covers Chapters
1-9 of the text.
In the fall 2005 offering of CHE 205, handouts and work-
sheets that guided students through problem solutions were
used extensively in lectures, problem sessions, and homework
assignments. In the problem sessions, the TAs provided a
modest amount of formal instruction in Excel and E-Z Solve
(a program on the text CD that solves algebraic and differential
equations numerically); carried out active exercises that guid-
ed students through the solution of unassigned text problems
and problems from old tests; and answered questions.
The syllabus, course policies, assignment schedule, hand-
outs, study guides, sample tests, and other course materials
may be viewed at
cbe205.html>, and the course policies are also listed in Ap-
pendix 1A of this paper. Features of the course structure that
departed from traditional practice were as follows:
* After thefirstfour weeks of the course, most of the
homework was done by instructor-assigned student
teams, with measures taken to satisfy the five defining
conditions for cooperative learning (individual account-
ability, positive interdependence, face-to-face interac-
tion, development and appropriate use of teamwork
skills, and regular self-assessment of team function-
ing).3] We will say more about those measures in Part 2
of this paper.
* Late assignments (i.e., assignments turned in after the
start of class on the due date) were accepted and graded
with a 40% penalty, but a team or individual was only
allowed two late assignments in the semester.
* Homeworkproblem solutions were not posted, 0 ii. .,, ,
final answers were given so that students reworking
problems would know when their solutions were correct.
When solutions are posted, many students just copy
them verbatim without trying to understand them; copies
find their way into file cabinets in fraternity and sorority
houses; and the frequency of perfect solutions achieved
without understanding steadily rises from one semester
to the next.
* The students could refer to their texts on the midterm
and final exams, but not to their course notes or graded
homework. Our rationale for this policy is that the stu-
dents work through a number of examples in the course
notes in active learning group exercises and do most of
their homework in groups. We make it clear to them that
they each need to understand the complete solutions re-
gardless of who in the group took the lead on each part,
tell them that we will be testing that understanding on
the exams, and do so with some explicit questions about
both the in-class exercises and the homework.
* Students who missed a midterm exam without a certified
medical excuse or prior approval took a comprehensive
makeup exam late in the semester. This policy enables
the instructor to write one makeup test instead of one
for each midterm, and the number of students who miss
Chemical Engineering Education
tests because of faulty alarm clocks and getting stuck in
traffic goes down dramatically. Individual arrangements
are made with students who had legitimate and verifiable
excuses for missing tests.
* Study guides were posted on the course Web site one to
two weeks before each of the three midterm tests and the
final exam. The study guide for a test contains a compre-
hensive list of learning objectives for that test-state-
ments of all the terms, phenomena, and concepts the
students might be asked to define or explain and the
kinds of problems they might be asked to solve, includ-
ing but not limited to the problems they had solved in
homework assignments. Study guides for all tests and
the final exam are posted at
;.,,i-i,. i. -2,5 ,' ,,,,,i, i,i, / >. The study guide for the
second test is shown in Appendix lB.
* Students could appeal their homework and test grades
within a week of the time the graded assignments were
handed back. To do so, they had to justify their request
for additionalpoints in -, it,,, The one-week limit
prevents a flood of requests for regrading at the end of
the semester when students traditionally scramble for
more points wherever they can get them, and requiring
written justification cuts down significantly on frivolous
* An absolute grading system was used-no curving.
This feature of the course encourages cooperation
among the students on homework (one of our primary
course goals), but it also puts a burden on the instruc-
tors to construct tests that are appropriately challenging
without being unfair. The tests are taken individually,
and students must earn an average grade of 60 or better
on tests for the team homework grades to count toward
the final course grade. This policy precludes students
moving on in the curriculum (which requires a C- or
better in CHE 205) solely because they were on a good
* A considerable amount of external support was avail-
able to the students. The two instructors and the three
lead teaching assistants each maintained three office
hours per week, and the instructors were also open to
e-mailed questions, although they made it clear that they
were not on call 24/7 and that e-mail messages sent at
11 p.m. the night before an assignment was due would
almost certainly not be replied to until sometime the
TEACHING ASSISTANT PREPARATION
We had an enrollment of 110 students in the two sections
of CHE 205 and six teaching assistants. Three of the TAs
(all graduate students) took responsibility for the problem
sessions (facilitating them and writing, administering, and
grading computer proficiency tests), and the other three (all
seniors) graded homework assignments. The TAs and course
instructors all held weekly office hours and worked together to
grade the midterm and final exams. We designated one of the
Vol. 41, No. 2, Spring 2007
graduate students who had previously helped with the course
as TA captain, in which capacity he coordinated grading and
worked with the instructors to develop the weekly problem
The graduate studentTAs had 10 hr/wk commitments includ-
ing 3 hr/wk office hours, and the undergraduate graders had
6-8 hr/wk commitments. The fact that most homework assign-
ments were submitted by groups of three to four instead of by
individual students kept the requirement for graders from being
excessive. The course instructors spent three hours each week
in class and another three hours holding office hours.
In the first week of class, we met with the TAs and gave them
a five-page memo that spelled out their roles and responsibili-
ties. The memo included the following instructions:
* Tips on office hours: When students come to you for help
on homework problems, don't just give them answers
or let them look at the solution key. Your first response
should be, "Show me what you've done so far." If it's a
material or energy balance problem, ask them to show
you their flowchart and degree-of-freedom analysis, get
them to explain the problem and outline the solution
strategy, and try to lead them to the solution by asking
questions. Let them do all the writing rather than just
watching you do it, and don't do any algebra or number
crunching-they can do that on their own time once they
understand how to derive the system equations.
* Tips on grading: Use a grading key that specifies in detail
how much credit is given for each part of every problem,
and make sure the same mistakes get the same deductions
on every paper. A given problem or portion of a problem
on an assignment or test should only be graded by one in-
dividual. Penalize careless errors enough so that it stings,
but don't slaughter students who basically understand
what they're doing. Try to figure out what the students did
wrong and make legible comments on their papers to help
them understand their mistakes. (It takes time to do that,
but if you don't spend it when you're grading you'll have
to spend it less productively in office hours explaining the
grades.) Avoid sarcastic comments about mistakes and
compliment good work. Make sure the homework is col-
lected when it is due on Friday and the graded papers are
returned to the instructors' mailboxes by Monday morning
before the class.
TEXT, CD, AND WORKBOOK
Instruction in the course closely followed Chapters 2-9 of
Elementary Principles of Chemical Processes (EPCP). The
students were required to bring the text with them to every
lecture class, among other reasons so that when working
through problems in class they would get practice in find-
ing the information (physical properties, conversion factors,
graphical correlations, etc.) they would need to look up on
the open-book tests to come.
The latest edition of EPCP contains a CD with tools includ-
ing E-Z Solve (a user-friendly equation-solving program), a
physical property database that includes a program to evaluate
sensible heats by integrating tabulated heat capacity formulas,
the Visual Encyclopedia of Chemical Engineering Equipment
created by Susan Montgomery of the University of Michigan,
and a set of six interactive instructional tutorials that lead the
students through most of the basic problem-solving techniques
needed in the course. E-Z Solve and the physical property
database enable students to solve problems involving many
simultaneous material and energy balance calculations with
relative ease (although we require some manual solutions of
system equations and integration of heat capacity formulas
before the students are allowed to use the software tools); the
Visual Encyclopedia gives students realistic views of the unit
operations and processes they analyze in the homework prob-
lems; and the tutorials provide practice and immediate feedback
in the analytical methods at the heart of the course.
When we first began to use this edition of the text, the
students virtually ignored the software tools we had been so
sure they would find invaluable. When we later required them
to use the tools in several early assignments, however, they
overcame their inertia and discovered how helpful the tools
could be. Their subsequent use of the software increased by an
order of magnitude over what it had been without those initial
assignments, and their end-of-course evaluations reflected a
strong appreciation of the CD's usefulness.
A common student complaint in the stoichiometry course is
that the examples presented in class tend to be much simpler
than many of the chapter-end problems that show up on home-
work assignments. The 2005 edition of the course text comes
with a workbook that guides students through the solutions
of several of the more complex text problems. We asked the
students to bring the workbooks to the problem session each
week, and the TAs chose relevant problems for the students
to work through individually or in groups. We also included
at least one workbook problem in each week's homework
assignment to be completed and submitted individually,
and we encouraged students to solve unassigned workbook
problems when studying for exams. In their end-of-course
evaluations, many students reacted positively to having the
workbook as a resource.
Problem sets were assigned weekly. Most of the assigned
problems were taken from the end of the textbook chapters,
sometimes with added parts calling for reflection on the
meaning of calculated results or speculation about possible
explanations for differences between the calculated results
and results that might be measured. Every three or four as-
signments the teams were asked to assess their performance
as a team. The assignments can be seen at
felder-public/l I,. ,'i .' i,. i i.,. . l k:.html>.
The text includes several creativity exercises that call on
students to brainstorm responses to open-ended questions
related to the course content. We assigned several of those
exercises and gave an even more general assignment toward
the end of the semester:
Creativity Exercise - Extra Credit
You may earn up to 10 points of extra credit on your final
homework assignment by ,ii-.,,,, a creative expression
of your experience in CHE 205. It might be a poem, song,
puzzle, artwork- the sky's the limit! The only constraints
are that your work must be original and your submission
must be in good taste ...., l,,,,, that could be shared with
the rest of the class). You may work in groups if your idea
requires multiple people to execute or is too big for one per-
son to complete individually (but group projects will have a
higher bar for grading).
In the three years in which we've given this assignment,
we've gotten a remarkable collection of products, includ-
ing crossword puzzles, cartoons, haiku and other poems,
paintings, a murder mystery, a crocheted doll holding a tiny
model of the textbook, a music video, and a live rap song
with costumes and choreography. Some of the students kept
doing the same sort of thing in other courses: two who did a
music video in the stoichiometry course went on to do sequels
in subsequent courses on process simulation and thermody-
namics, and another student who submitted a personal course
journal in CHE 205 continued the journal through her senior
year, documenting her entire experience in chemical engineer-
ing. We spend the last day of class allowing students to share
their contributions, which are generally received with lots of
laughter and cheering. It's a great way to end the semester.
One of us has worked for several years with N.C. State
librarians to incorporate information literacy concepts
throughout the CBE curriculum, starting with the freshman
engineering course and continuing through the sophomore
stoichiometry course, the junior professional development
seminar, and the senior capstone design course. In CHE
205, librarians visit during a problem session to introduce
students to important discipline-specific resources that
chemical engineers typically use, including Perry's Chemical
Engineers Handbook, the Chemical Economics Handbook,
the Kirk-Othmer Encyclopedia of Chemical Technology, and
the Chemical Market Reporter, as well as databases including
Compendex and SciFinder Scholar. The presenters stress the
importance of proper literature citation and give students brief
practice citation exercises, and they discuss the idea that the
credibility of information depends strongly on the source,
with Perry's Handbook and a MySpace blog representing
extremes of trustworthiness.
Chemical Engineering Education
The following assignment is given to students following
the information literacy presentation. Typically they are given
two to three weeks to complete it. By linking information
competencies to assignments related to class material, we
move beyond decoupled instruction that is quickly forgotten
to "just-in-time" need-based instruction.
1. Select a chemical substance from Table B.I in your text
that begins with the same letter as your first name or the
nearest possible letter (for example Angie --Aniline).
Find and report the information listed below for this
substance in references other than the course text or
CD, and properly cite the references. Organize your
report neatly and show all units.
(a) Specific gravity, molecular weight, normal melt-
ing and boiling points, Antoine constants, heats
of fusion and vaporization at the normal m1,, ii,,,
and boiling points, and heat capacity as a func-
tion of temperature. If some of these properties
are missing for your chosen species, choose a dif-
ferent species with complete physical properties.
(b) Several examples of industrial uses of the species.
(c) Toxicity data and environmental hazards associ-
ated with the species.
(d) At least three companies that manufacture the
(e) Worldwide demand and/or sales.
(f) Unit pricing (' ,. $/gal, etc.) Your figure should
reflect bulk pricing, not pricing of small units
from laboratory supply firms such as Fisher
2. From the textbook index, select a topic that begins with
the same letter as your last name or the nearest possible
letter (for example Brent - Bubble point). Identify three
published articles (not Web sites) that deal with this
topic and list their full bibliographic citations. Then find
the articles and photocopy or print out their first pages
and abstracts (if the abstracts are not included in the
SUMMARY OF PART 1
The two-part series of papers of which this is the first part
describes the structure of the stoichiometry course at North
Carolina State University in the fall of 2005. The course had a
variety of learning objectives, including traditional objectives
related to the course content and also objectives involving
creative thinking skills, communications and teamwork, and
information literacy. Several nontraditional pedagogies were
used in the course delivery, including active, cooperative, and
inquiry-based learning, and a number of different applications
of instructional technology. This paper outlines the course
structure and policies, the preparation given to the teaching
assistants who played an integral part in the course delivery,
and the course assignments. The next paper summarizes the
methods used in the course instruction and assessment.
Vol. 41, No. 2, Spring 2007
We acknowledge with gratitude the numerous and signifi-
cant contributions to CHE 205 of the TAs (especially Adam
Melvin). Our thanks also go to Honora Nerz Eskridge for her
invaluable assistance with the information literacy segment
of the course.
1. Felder, R.M., and R.W Rousseau, Elementary Principles of Chemi-
cal Processes, 2005 Update Edition, New York, John Wiley & Sons
2. Felder, R.M., "Stoichiometry Without Tears," Chem. Eng. Ed., 24(4),
3. Smith, K.A., S.D. Sheppard, D.W Johnson, and R.T. Johnson, "Pedago-
gies of Engagement: Classroom-Based Practices," J. Eng. Ed., 94(1),
4. Roskowski, A.M., R.M. Felder, and L.G. Bullard, "Student Use (and
Non-Use) of Instructional Software," J. Science, Math, Engineering,
and Technology (SMET) Education, 2, 41 (Sept.-Dec. 2001)
5. Bullard, L.G., and H. Nerz, "The Literature Engineer: Infusing In-
formation Literacy Skills throughout an Engineering Curriculum,"
Proceedings of the 2006 ASEE Annual Conference, Chicago (June
ChE 205 Policies And Procedures
* Academic integrity. Students should refer to the university
policy on academic integrity found in the Code of Student
Conduct (found in Appendix L of the Handbookfor Advising
and Teaching). It is the instructor's understanding and expec-
tation that the student's signature on any test or assignment
means that the student contributed to the assignment in ques-
tion (if a group assignment) and that the student neither gave
nor received unauthorized aid (if an individual assignment).
Authorized aid on an individual assignment includes discussing
the interpretation of the problem statement, sharing ideas or
approaches for solving the problem, and explaining concepts
involved in the problem. Any other aid would be unauthorized
and a violation of the academic integrity policy. All cases of
academic misconduct will be submitted to the Office of Stu-
dent Conduct. If you are found guilty of academic misconduct
in the course, you will be on academic integrity probation for the
remainder of your years at NCSU and may be required to report
your violation on future professional school applications. It's not
* Homework. Students will submit homework individually for
the first few homework assignments. Early in the semester, the
instructors will designate teams of three to four individuals,
and all subsequent assignments should be submitted by those
teams unless otherwise specified. The assignment schedule will
be posted on the course Web site.
* Homework format. Use engineering paper (available in the
Student Supply Store), one side of each page; begin each
problem on a new page; and box the final answers. Each com-
pleted assignment should be in one person's handwriting (the
recorder's). Staple the pages and fold them vertically when
you hand them in, putting your name (individual assignments)
or the names and roles (coordinator, recorder, checker, and
monitor) of the ', i,. q,.m,,ii team members (team assignment), and the problem set number and date on the outside. If a student's name
appears on a solution set, it certifies that he/she has participated in solving the problems. If this turns out not to be the case, both the
iv. ,q .n, i,. .,,,- , student and the recorder will get zeros for that assignment.
* Late homework. Completed assignments should be turned in at the beginning of class on the due date (Bullard's section) or to the
homework box in the CHE lounge (Felder's section) between 5 p.m. on Thursday (day before the due date) and 9:30 a.m. on Friday
(the due date). If it's your job to turn in the homework and you're late, so is the homework. Late assignments will receive a maximum
grade of 60. Late solution sets will be accepted up to 8 a.m. on the Monday after the due date, turned in to your instructor's mailbox in
the CHE office, 2001 EB1. Once an individual or a group hands in two late assignments, however, no more will be accepted.
* Posted solutions. Problem set solutions will not be posted. It is your responsibility to make sure you find out how to solve the problems
by asking about them in class, during office hours, or in the problem session after they have been handed in.
* Individual effort assessments for team homework. Teams will periodically be asked to submit individual effort assessments with
completed assignments. These assessments will be incorporated into the assignment of homework grades. If repeated efforts to improve
teamfi, i. it. -1111, (including faculty intervention) fail, a nonparticipant may be fired by unanimous consent of the rest of the team, and a
team member doing essentially all the work may quit. (Details of the required procedures are given in the handout on team policies and
expectations.) Individuals who quit or are fired must find a team unanimously willing to accept them; otherwise they will receive zeros
for the remainder of the homework.
* Tests. There will be three tests during the semester and a comprehensive final exam. All tests will be open-book, closed-notes. Each
student's lowest test grade will count half as much as the other two. Tests will be given as a common exam for both sections on scheduled
Friday from 3-5 p.m. (see detailed course schedule for dates). Students who are unable to take the test at those times (with a documented
excuse-notjust that you don't want to) will schedule an alternate time to take the exam. To make up for the additional test time required
out of class, the class period before the exam will be an optional review session conducted by the instructor or a TA.
* Test and homework grading. The responsibility for grading tests and homework assignments resides with the graders. If you believe an
error has been made in grading on a problem set, bring it to the grader who did the grading during his or her office hours. If you believe
that you should have gotten more points than you got for any reason other than a simple addition error, write a statement making your
case and take it to the grader. If you are not satisfied with the grader's decision, bring the statement to your course instructor, who will
make the final decision.
* Missed tests. If you miss a test without either a certified medical excuse or prior instructor approval, you will take a makeup test at a
designated time during the last week of the semester. The makeup exam will be fair but comprehensive (covering all the course mate-
rial) and challenging. Tests missed with certified medical excuses or prior instructor approval will be dealt with individually. Only one
missed test can be made up.
* Problem session. All 205 students must be registeredfor one of the weeklyproblem sessions (205P). Several computer applications will
be taught during the problem sessions. Ten percent of your grade is based on problem session quizzes and in-class exercises. Attendance
is expected. You should not float between problem sessions; stay in the one in which you are registered. If it is necessary to miss a
problem session, however, you may attend another session to make up the time as long as you notify the TA of the problem session you
attend so that your attendance can be recorded.
* Attendance. Students who miss class due to an excused absence should work with the instructor or problem session TA to make up any
missed work. Documented medical excuses should be presented to the instructor. Examples of anticipated situations where a student
would qualify for an excused absence are:
a. The student is away from campus representing an official university function, e.g., participating in a professional meeting, as part of a judging
team, or athletic team. These students would typically be accompanied by a university faculty or staff member.
b. Required court attendance as certified by the Clerk of Court.
c. Religious observances as verified by Parents & Constituent Services (515-2441). For more information about a variety of religious observances,
visit the Diversity Calendar.
d. Required military duty as certified by the student's commanding officer.
For a full statement of the university attendance policy, see .
* Calculation of course grade. A weighted average grade will be calculated as follows:
D- Midterm tests = 40% (Lowest grade counts 1/2 of each of the other two)
- Final examination = 30%
- Homework = 20%
- Problem session quizzes and in-class exercises = 10%.
Weighted >97 93- 90- 87- 83- 80- 77- 73- 70- 67- 63- 60- <
average 1 96.9 92.9 89.9 86.9 82.9 79.9 76.9 72.9 69.9 66.9 62.9 60
The homework grades will only count if the average grade on class tests and the final exam is 60 or above-in other words, if you
can't pass the individual tests, then you can't pass the course.
Note: We do not curve grades in this course. It is theoretically possible for everyone in the class to get an A (or an F). Your performance
depends only on how you do, not on how everyone else in the class does. It is therefore in your best interests to help your classmates,
Chemical Engineering Education
while keeping the academic integrity policy in mind.
* Instructors' commitment. You can expect your instructors to be courteous, punctual, well organized, and prepared for lecture and other
class activities; to answer questions clearly and in a non-negative fashion; to be available during office hours or to notify you beforehand
if they are unable to keep them; to provide a suitable guest lecturer when they are traveling; and to grade uniformly and consistently
according to the posted guidelines.
* Consulting with faculty. We strongly encourage you to discuss academic or personal questions with either of the CHE 205 course
instructors during their office hours or by e-mail.
* Disabled students. North Carolina State is subject to the Department of Health, Education, and Welfare regulations implementing Section
504 of the Rehabilitation Act of 1973. Section 504 provides that: "No otherwise qualified handicapped individual in the United States.
. shall, solely by reason of his handicap be excluded from participation in, be denied the benefits of, or be subjected to discrimination
under any program or activity receiving Federal financial assistance." This regulation includes students with hearing, visual, motor, or
learning disabilities and states that colleges and universities must make "reasonable adjustments" to ensure that academic requirements
are not discriminatory. Modifications may require rescheduling classes from inaccessible to accessible buildings, providing access to
auxiliary aids such as tape recorders, special lab equipment, or other services such as readers, note takers, or interpreters. It further
requires that exams actually evaluate students' progress and achievement rather than reflect their impaired skills. This may require oral
or taped tests, readers, scribes, separate testing rooms, or extension of time limits.
Team Policies and Expectations
Your team will have a number of responsibilities as it completes problem and project assignments.
* Designate a coordinator, recorder, and checkerfor each assignment. Rotate these roles for every assignment.
* Agree on a common .... ,>, time and what each member should have done before the .,.. o,,>, (readings, taking the first cut at some or
all of the assigned work, etc.)
* Do the required individual preparation.
* Coordinator checks with other team members before the .,. o,, to remind them of when and where they will meet and what they are
supposed to do.
* Meet and work. Coordinator keeps everyone on task and makes sure everyone is involved, recorder prepares final solution to be turned
in, monitor checks to makes sure everyone understands both the solution and the strategy used to get it, and checker double-checks it
before it is handed in. Agree on next meeting time and roles for next assignment. For teams of three, the same person should cover the
monitor and checker roles.
* Checker turns in the assignment, with the names on it ofevery team member whoparticipated actively in completing it. If the checker
anticipates a problem getting to class on time on the due date of the assignment, it is his/her responsibility to make sure someone
turns it in.
* Review returned assignments. Make sure everyone understands why points were lost and how to correct errors.
* Consult with your instructor if a conflict arises that can't be worked .- ,,.., i by the team.
* Ifa team member refuses to cooperate on an assignment, his/her name should not be included on the completed work. If the noncoop-
eration continues, the team should meet with the instructor so that the problem can be resolved, if possible. If no resolution is achieved,
the cooperating team members may notify the uncooperative member in writing that he/she is in danger of being fired, sending a copy
of the memo to the instructor. If there is no subsequent improvement, they should notify the individual in writing (copy to the instructor)
that he/she is no longer with the team. The fired student should meet with his/her instructor to discuss options. Similarly, students who
are consistently doing all the work for their team may issue a warning memo that they will quit unless they start getting cooperation, and
a second memo quitting the team if the cooperation is not forthcoming. Students who get fired or quit must find a team of three willing
to accept them as a member, otherwise they get zeros for the remaining assignments.
As you will find out, group work isn't always easy-team members sometimes cannot prepare for or attend group sessions because of other
responsibilities, and conflicts often result from differing skill levels and work ethics. When teams work and communicate well, however,
the benefits more than compensate for the difficulties. One way to improve the chances that a team will work well is to agree beforehand
on what everyone on the team expects from everyone else. Reaching this agreement is the goal of the assignment on the last part of
Team Expectations Assignment
On a single sheet of paper, put your names and list the rules and expectations you agree as a team to adopt. You can deal with
any or all aspects of the responsibilities outlined above - preparation for and attendance at group meetings, making sure everyone
understands all the solutions, communicating frankly but with respect when conflicts arise, etc. Each team member should sign the
sheet, indicating acceptance of these expectations and intention to fulfill them.
These expectations are for your use and benefit-we won't grade them or even comment on them unless you ask us to. Note, how-
ever, that if you make the list fairly thorough without being unrealistic you'll be giving yourselves the best chance. For example, "We will
each solve every problem in every assignment completely before we get together" or "We will get 100 on every assignment" or "We will
never miss a meeting" are probably unrealistic, but "We will try to set up the problems individually before meeting" and "We will make
sure that anyone who misses a meeting for good cause gets caught up on the work" are realistic.
Vol. 41, No. 2, Spring 2007
Study Guide For Midterm Test 2*
To do well on the next test, you should be able to do the following:
1. Explain in your own words the terms separation process, distillation, absorption, scrubbing, extraction, crystallization, adsorption, and
leaching. (What are they and how do they work?)
2. Sketch a phase diagram (P vs. T) for a single species and label the regions (solid, liquid, vapor, gas).
Explain the difference between a vapor and a gas. Use the phase diagram to define (a) the vapor pressure at a specified temperature, (b)
the boiling point at a specified pressure, (c) the normal boiling point, (d) the melting point at a specified pressure, (e) the sublimation
point at a specified pressure, (f) the triple point, (g) the critical temperature and pressure. Explain how the melting and boiling point
temperatures vary with pressure and how P and T vary with time (increase, decrease, or remain constant) as a specified path on the
diagram is followed.
3. Estimate the vapor pressure of a pure substance at a specified temperature or the boiling point at a specified pressure using (a) the Antoine
equation, (b) the Cox chart, (c) the Clausius-Clapeyron equation and vapor pressures at two specified temperatures, (d) Table B.3 for water.
4. Use data in the text to speculate on whether distillation and/or crystallization might be a reasonable separation process for a mixture of
two given species. List the additional information you would need to confirm your speculation.
5. Distinguish between intensive and extensive variables, giving examples of each. Use the Gibbs phase rule to determine the number of
degrees of freedom for a multicomponent multiphase system at equilibrium, and state the meaning of the value you calculate in terms of
the system's intensive variables. Identify a feasible set of intensive variables to specify that will enable the remaining intensive variables
to be calculated.
6. In the context of a system containing a single condensable species and other noncondensable gases, explain in your own words the terms
saturated vapor, superheated vapor, dew point, degrees of superheat, and relative saturation. Explain the following statement from a
weather report in terms a first-year engineering student could understand: "The temperature is 75 F, barometric pressure is 29.87 inches
of mercury and falling, the relative humidity is 50%, and the dew point is 54�E "
7. Given an equilibrium gas-liquid system with a single condensable component (A) and liquid A present, a correlation for p *(T), and any two
of the variables y (mole fraction ofA(v) in the gas phase), temperature, and total pressure, calculate the third variable using Raoult's law.
8. Given a mixture of a single condensable vapor, A, and one or more noncondensable gases, a correlation for p *(T), and any two of the
variables YA (mole fraction of A), temperature, total pressure, dew point, degrees of superheat, and relative, molal, absolute, and percent-
age saturation (or humidity), use Raoult's law for a single condensable species to calculate the remaining variables.
9. For a process system that involves a gas phase containing a single condensable component and specified or requested values of feed or
product stream saturation parameters (temperature, pressure, dew point, relative saturation or humidity, degrees of superheat, etc.), draw
and label the flowchart, carry out the degree-of-freedom analysis, and perform the required calculations.
10. After completing your analysis of a vapor-liquid phase change process, identify as many possible reasons as you can for discrepancies
between what you calculated and what would be measured in a real process. Include any assumptions made in the calculation.
11. Explain the meaning of the term "ideal solution behavior" in the context of a liquid mixture of several volatile species. Write and clearly
explain the formulas for Raoult's law and Henry's law, state the conditions for which each correlation is most likely to be accurate, and
apply each one to determine any of the variables T, P, xA, or y (temperature, pressure, and mole fractions of A in the liquid and gas
phases) from given values of the other three.
12. Explain in your own words the terms bubble point, boiling point, and dew point of a mixture of condensable species, and the difference
between vaporization and boiling. Use Raoult's law to determine (a) the bubble point temperature (or pressure) of a liquid mixture
of known composition at a specified pressure (or temperature), and the composition of the first bubble that forms; (b) the dew point
temperature (or pressure) of a vapor mixture of known composition at a specified pressure (or temperature), and the composition of the
first liquid drop that forms; (c) whether a mixture of known amount (moles) and composition (component mole fractions) at a given
temperature and pressure is a liquid, a gas, or a gas-liquid mixture, and if the latter, the amounts and compositions of each phase; (d) the
boiling point temperature of a liquid mixture of known composition at a specified total pressure.
13. Use a Txy or Pxy diagram to determine bubble and dew point temperatures and pressures, compositions and relative amounts of each phase in
a two-phase mixture, and the effects of varying temperature and pressure on bubble points, dew points, and phase amounts and compositions.
Outline how the diagrams are constructed for mixtures of components that obey Raoult's law. Construct a diagram using a spreadsheet.
14. For a process system that involves liquid and gas streams in equilibrium and vapor-liquid equilibrium relations for distributed compo-
nents, draw and label the flowchart, carry out the degree-of-freedom analysis, and perform the required calculations.
15. Explain in your own words the terms solubility of a solid in a liquid, saturated solution, and hydrated salt. Given solubility data, deter-
mine the saturation temperature of a feed solution of given composition and the quantity of solid crystals that precipitate if the solution
is cooled to a specified temperature below the saturation point. Perform material and energy balance calculations on a crystallizer, and
identify sources of error in your calculation.
16. Given a liquid solution of a nonvolatile solute, estimate the solvent vapor pressure lowering, the boiling point elevation, and the freez-
ing point depression, and list the assumptions required for your estimate to be accurate. Give several practical applications of these
phenomena. Identify sources of error in your calculation.
17. Given the description of a familiar phenomenon involving more than one phase, explain it in terms of concepts discussed in this chapter.
Given an explanation of such a phenomenon, evaluate its scientific soundness. 1
* This test covered through Chapter 6 of the text.
100 Chemical Engineering Education
Ij 1 laboratory
LIQUID-LIQUID MIXING LABORATORY
For Chemical Engineering Undergraduates
SANAZ BARAR POUR, GREGORY BENOIT NORCA, Louis FRADETTE, ROBERT LEGROS, PHILIPPE A. TANGUY
Ecole Polytechnique * Stn. Centre-ville, Montreal
n the chemical industries, fluid mixing in stirred tanks is Sanaz Barar Pour received her B.S. from the University of Tehran in
involved in a wide variety of operations such as homog- chemical engineering in 2004. She is currently a master's student at Ecole
enization of miscible liquids (blending), gas dispersion, Polytechnique of Montreal. Her research activity is related to solid-liquid
dispersion in highly viscous media.
mixing of immiscible liquids emulsificationn or dispersion),
and suspension of solid particles. Emulsification is a process Gregory Benoit Norca received his B.S. from Ecole Polytechnique of
Montrealin mechanical engineering. Currently he is following an engineer-
consisting of dispersing droplets of an oil phase into a water ing program in ENSTA, France. His field of interest is mechanical structure
phase (O/W) or a water phase in an oil phase (W/O). A sur- design and analysis in automobile, aeronautics, and aerospace.
factant is normally added to stabilize the dispersion. Louis Fradette completed his B.S. and M.S. degrees at Laval University
s o n is te b s f m m p (Quebec City). He holds a Ph.D. from Ecole Polytechnique of Montreal and
This operation is the basis for many manufacturing pro- Institute National Polytechnique de Lorraine, France. He has more than
cesses in the food, cosmetics, and pharmaceutical industries, 10 years of process engineering experience in various fields such as pe-
to name a few. The suspension of solid particles in agitated troleum refining, petrochemicals, and software development. Since 2004,
he has been working as a research associate at Ecole Polytechnique of
vessels is required in many industrial reactors dealing with Montreal with research activities related to the development of efficient
catalytic reactions, crystallization, leaching, polymerization, continuous and batch processes for the production of emulsions.
dissolution, ion exchange, and adsorption. The typical objec- Robert Legros is a professor of chemical engineering at Ecole Poly-
tives of solid-liquid suspension are to produce a homogenous technique of Montreal. Currently he is the chair of the department. He
and particle size) and to promote the received his B.S. from Ecole Polytechnique in 1983 and his Ph.D. from
slurry (concentration and particle size) and to promote the the University of Surrey in 1987. His academic research involves solids
rate of mass transfer between the solid and liquid phases. In thermal treatments in fluid beds, modeling of combustion reactor, heat
most applications, determining the minimum impeller speed and mass transfer, and hydrodynamics of spouted beds. Some of his
current research interests are related to pharmaceutical engineering,
for off-bottom suspension or to disperse a liquid phase in namely in powder technology and downstream processes.
mechanically agitated liquids in stirred vessels has consider-
Philippe A. Tanguy is a professor of chemical engineering at Ecole
able importance.[1 Polytechnique of Montreal. He received his B.S. in 1976, his doctorate
In order to provide chemical engineering students with prac de specialite in 1979 from Universite de Paris, and his Ph.D. in 1982
from Laval University. His research interests are in non-Newtonian fluid
tical experience on suspension and emulsification processes, mechanics, CFD and process engineering involving complex fluids, in
laboratory solid-liquid and liquid-liquid mixing experiments particular coating processes, and agitation and mixing operations. He
is currently Director of the Research Unit in Industrial Flow Processes
have been developed. The objective is to highlight the main (URPEI).
� Copyright ChE Division of ASEE 2007
Spring 2007 10
aspects governing process efficiency, namely the influence
of the operating parameters and the effect of the impeller
type. The mixing laboratory is a part of a senior-year unit
operations course and a graduate mixing course, both of-
fered by the Department of Chemical Engineering at Ecole
Polytechnique of Montreal. The experiments are carried
out as a group exercise. Each group consists of a maximum
of three students, and must perform the required laboratory
work in less than four hours. After finishing the experiments,
each group should hand over a full report or prepare an
oral presentation the following week. Both report and oral
presentation consist of the following sections: description
of the experiment objectives, theoretical basis, engineering
method used, experimental set-up and operating conditions,
experimental data, analysis of the results, and comments or
recommendations for improvements.
The mixing system used in the experiments is a modified
Turbotest (VMI Rayneri) laboratory mixer. It consists of a
transparent polycarbonate vessel of 165 mm inner diameter
and 230 mm height, with an open top fixed to a rigid table for
safe operation. Two impellers are tested for liquid-liquid dis-
persions: a radial-flow impeller (6-blade Rushton turbine) and
an axial-flow impeller (marine propeller). In the solid-liquid
suspension work, the 6-blade Rushton turbine is compared
to a pitched blade turbine (four 45� blades). The impellers
are mounted on a rigid shaft driven by a DC motor, in which
speed is carefully regulated in a range from 10 to 1,800 rpm
by means of a DC controller. The motor is mounted on a rigid
structure that can be moved to adjust the vertical position of
the impeller. A standard mixing configuration is used as a
starting point, with the impeller placed on the vessel centerline
at 1/3 of the liquid height in liquid-liquid dispersion and 1/6
of the liquid height in solid-liquid suspension. The agitation
torque is measured by a noncontact type torque meter (range
between 0.1 and 1.42 N.m) fitted between the motor and the
The term immiscible liquid-liquid system designates two
or more thermodynamically incompatible liquids present as
separate phases. Emulsion is defined as a liquid-liquid dis-
persion stabilized by means of one or more surfactants that
allow long-term stability (from days to months depending on
the formulation). The term liquid-liquid dispersion is kept for
nonstabilized (unstable) systems where the dispersed state is
maintained by continuous agitation. Agitators that provide
high shear and good pumping capacity are common choices
for liquid-liquid dispersion and emulsification.
When the densities between the dispersed phase and the
continuous phase are different, and the agitator does not gener-
ate a sufficient circulation throughout the vessel, settling and
coalescence occur. Consequently, it is important to determine
the minimum speed for dispersion of the droplets.
Skelland and Seksaria 31 proposed a dimensionless correla-
tion to predict the minimum speed for liquid-liquid dispersion
in two liquids with different densities:
SD05 = C20 J [ [ D2 g (1)
g D N P YD
where D is the diameter of the impeller, g the gravitational
acceleration, T the tank diameter, o the interfacial tension,
g. and gd the viscosity of the continuous phase and the dis-
persed phase, respectively, and po and Pd the density of the
continuous phase and dispersed phase, respectively. In Eq.
(1) Ap = |pd - p I C20 is a model constant scaling the ease
of forming a suspension. Table 12-5 in the Handbook of In-
dustrial .\li.vim','1 gives more information as to the value of
these constants for different types of impellers.
How regimes are characterized by the value of the Reyn-
olds number, Re, which is the ratio of inertial forces over
the viscous forces. The Reynolds number of an agitator in a
liquid-liquid system is:
Re = D (2)
where p and jT are the density and viscosity of the mixed vol-
ume phases. For diluted dispersions ( < 0.01), these values
are equal to density and viscosity of the continuous phase. The
laminar regime corresponds to Re < 10, the transition regime
to 10 < Re < 104 and the turbulent regime to Re > 104.
The emulsification time is the time necessary to reach a
stable droplet size and droplet size distribution. This time
depends on the frequency of droplet passage in the agitator's
region or, alternatively, to the circulation time t:
t N (3)
where V is the tank volume, Nq the circulation number, N the
rotational speed of the agitator, and D the diameter of the agi-
Chemical Engineering Education
Figure 1. Droplet size as a function of time.41
tator. Circulation number or pumping number, N , varies with
impeller type and hydrodynamic regime. The typical value of
circulation number for Rushton turbine and marine propeller
is 0.67 and 0.53, respectively. Figure 1 shows a typical curve
of a droplet-size evolution in an emulsification tank.
The droplet mean diameter can be defined by the follow-
n- d lm /m-n
dmn d ndm (4)
where m= 1,2,3; n= 0,1,2; and m > n.
A common choice for the mean diameter is the Sauter mean
diameter (d32), since it is directly related to the interfacial area
per unit volume, a , which determines the transfer rate of en-
ergy, mass, and/or the chemical reaction in the dispersion:
a, = (5)
where P is the volume fraction of dispersed phase.
C EXPERIMENT 1
The objective of this experiment is to determine the influ-
ence of the type of agitator and its position on the speed
required for complete dispersion in multiphase systems with
separate phases. The corresponding power consumption will
be obtained. In liquid-liquid dispersion, knowing the mini-
mum agitator speed for complete dispersion, Nmn , enables us
to design the mixer efficiently.
In order to reach the objectives the detailed procedure is
1) Use the configuration shown in Figure 2 with the Rush-
2) Add 3 L (85%v) water and 0.5 L (15%v) of sunflower
oil up to H=165 mm level.
3) Set the rotational speed equal to zero rpm and verify
that the torque screen displays zero.
Figure 2. Experimental setup.
4) Gradually increase the rotational speed of the agita-
tor up to the point where no continuous layer of the
dispersed phase remains in the vessel. Write down
the torque and the rotational speed and then stop the
agitator. Calculate the power consumption by using the
P= 2.T.N.M, (6)
where P is the power consumption, N is the rotational
speed of the agitator, and M. is the corrected torque.
The shaft guiding system induces a residual torque due
to friction; hence, the torque value must be corrected by
subtracting the residual torque from the measured torque for
M, = Mm -M, (7)
where Mm and Mr are measured and residual torques, re-
5) Wait 3-4 minutes to allow the phases to separate. Place
the Rushton turbine at position C/H = 0.5 and repeat
from step 4.
6) Repeat the experiment with the marine propeller.
In each case determine the flow regime. Use the following
values for viscosity and density: H = 0.008 Pa.s and p =
985 kg/m3. What is the influence of the vertical position of
the agitator on the minimum speed for dispersion and power
consumption? What is the best position for the agitator in
this setup? Which agitator would you use for liquid-liquid
The complete dispersion speed, Nd, is determined in each
case. The Rushton turbine is more efficient than the helical
marine impeller in terms of power consumption. For the
Rushton turbine, dispersion was not obtained in laminar
regime in both cases. For the helical marine impeller, disper-
sion was obtained in turbulent regime where C/H=1/3 and in
transitional regime where C/H=1/2. Hence, by placing the
agitator at C/H=1/2, the complete dispersion speed and power
consumption are decreased by more than 50% in both cases.
The best position of the agitator would be the interface of the
two phases. For this application of liquid-liquid dispersion,
the Rushton turbine is preferred.
The objective of this experiment is to determine the effect of
mixing duration and rotational speed of agitator on the particle
size and particle size distribution. Particle size distribution
is one of the most important characteristics of liquid-liquid
dispersion. A laser granulometer will be used to measure the
particle size of the droplets.
1) Place the Rushton turbine on the shaft as in Experiment 1.
2) Add water up to H=165 mm level.
3) Add 7 mL of surfactant in order to stabilize the emul-
4) Start the agitator in order to dissolve the surfactant
completely in the water. Set the agitator at 225 rpm,
add 35 mL of sunflower oil, and start the chronometer
5) Take samples at time = 1, 4, 7, 10, 13, 16, and 20
minutes or beyond and measure the particle size of the
droplets by means of granulometer until the diameter,
d3, becomes constant. The samples can be taken by a
syringe from the same location in the vessel. Typical
experimental results are shown in Table 1.
6) Repeat the same procedure for 300 and 375 rpm.
7) Save the files in the Excel format (.xls) according to the
procedure written in the user manual of the granulome-
ter. These files will be used in order to investigate the
droplet size distribution.
Plot the droplet size distribution in terms of volume frequency
(vol% vs. droplet size distribution) at 1, 4, 7, and 13 minutes
for the first speed (225 RPM). What do you observe?
An example of typical plots is presented on Figure 3.
Plot the Sauter diameter, d32, as a function of time for each
speed. How does the droplet size change with time? How does
it change with the rotational speed? How does the equilib-
rium time change with the
TABLE 1 rotational speed?
Time (min) d32 ([rm) An example of typical
1 30.09 curves is shown in Fig-
4 24.50 ure 4. This figure shows
S22.1 that the droplet size is
decreasing with time.
10 14.98 Increasing the rotational
13 13.05 speed, however, will re-
16 11.04 duce droplet size and
20 11.04 equilibrium time.
0 50 100 150 200 250 300
Drop size (um)
350 400 450 500
Figure 3. Particle size distribution obtained
at 1, 4, 7, and 13 minutes.
A suspension is a dispersion of particulate solids in a con-
tinuous liquid phase that is sufficiently fluid to be circulated
by a mixing device.5s] The systems resulting from incorporat-
ing powders and fine particles throughout the liquid medium
shall be considered as dispersion or colloidal suspensions.
In agitated vessels, the degree of solid suspension can be
classified as follows: [4
* On-bottom motion
* Complete off-bottom suspension
* Uniform suspension
The required level of suspension depends on the desired
process result and the unit operation involved. For example,
a high degree of suspension is required for crystallization or
slurrying, whereas a lower degree of suspension is usually
sufficient for the dissolution of a highly soluble solid.
The state of suspension known as complete off-bottom
suspension is characterized by complete motion, with no
particle remaining at rest on the vessel bottom for more than
one or two seconds. This criterion is known as the Zwieter-
ing'] criterion. Operation at the minimum impeller speed for
a just-complete suspension condition is adequate for many
reaction or mass transfer processes, and much of the study on
the solid-liquid mixing is concerned with the measurement
and correlations of N values.
According to Zwietering,['1 the following correlation can be
used to estimate the just suspended impeller speed, N:
N= Suo g (P - P X 3d 2D-085 (8)
S=Re omFr045 (9)
Reiip = is the impeller Reynolds number, the Froude
numbers Fr = N , D is the impeller diameter, d is the
mass-mean particle diameter, X the mass ratio of suspended
solids to liquid, v is the kinematic viscosity of the liquid, g
is the gravitational acceleration, p5 and p, are the density of
particle and the density of liquid, respectively.
The objective of this experiment is to determine the just-
suspension impeller speed, N , and the value of S for a radial
and an axial impeller, namely a Rushton turbine and a pitched-
blade turbine. In most of the chemical process industries
it is essential to provide complete off-bottom suspension.
Chemical Engineering Education
Below this speed the total solid-liquid interfacial area is not
completely and efficiently used for mass transfer. Above this
speed the mass transfer rate increases slowly and the power
dissipated increases considerably. Therefore, it is important
to determine the just-suspension impeller speed, N .
1) Place the Rushton turbine on the shaft as before.
2) Add water into the tank to H level with H=T.
3) Set the impeller at 1/6 of the liquid height C/T=1/6. It
is necessary to keep this bottom clearance constant in
all the experiments.
4) According to the given concentration of 1,000 Rm
glass beads, weigh the exact amount of glass beads by
means of a balance and then add them to the tank.
5) Set the rotational speed equal to zero rpm and verify
that the torque screen displays zero.
6) Increase the rotational speed of the impeller until you
reach the Zwietering visual criterion for just suspen-
sion. Note the just suspended speed.
7) With this concentration of solids, proceed with the
change of impellers and repeat step 6.
8) Increase the amount of glass beads to the next de-
sired concentration and repeat steps 5-7 for the new
concentration. Note N for both impellers at the new
9) Repeat step 5-8 for two different diameters of glass
10) Prepare a PEG/water fluid mixture with 0.02 Pa.s of
viscosity. Use Figure 5 to estimate the amount of PEG
(Polyethylene Glycol, grade 35000) to be mixed with
water until you reach H = T. After dissolving the PEG,
wait two minutes until a clear fluid is obtained. (PEG
increases the viscosity of the solution without having
considerable effect on the density).
11) Repeat steps 5-8 with this
fluid for both impellers and
glass beads with 3000 pm
With the experimental measure-
ments and Figure 5, fill in Table 2 and
answer the following questions:
a) Determine the value of S for
both impellers from the first ex-
periment when the fluid was water.
What is the value of S for each
b) For the second experiment
with higher viscosity fluid, replace
the value of S that you have ob-
tained in the Zwietering correlation
to compute N . Compare just sus-
O 5 10 15 20 2E
Figure 4. Sauter diameter and equilibrium time as a
function of rotational speed for Rushton turbine.
10 15 20 25
Massconcentration (wt%of PEG 35000)
Figure 5. Viscosity vs. mass concentration
of PEG (grade 35000).
phase dP(pm) s(Hz) N (rps)
1000 0.5 22.9 10.81 4.75
1000 1.5 25.9 12.32 4.66
1000 5 32.3 15.25 4.97
- 103 dp(pm) X N (Hz) N, (rps) S
3000 0.5 26.6 12.56 4.43
3000 1.5 33.1 15.63 4.47
3000 5 37.3 16.61 4.61
War d ) X s () Experimental Calculated
Water dp (pm) X N (Hz) N (p) N (rps)
+ _ Ns (rps) Nis (rps)
PEG 3000 0.5 34.4 16.24 16.57
= 0.02 3000 1.5 39.2 18.51 19.11
3000 5 43.9 20.73 21.35
-- N=225 rpm
----- N=300 rpm
-- --N=375 pm
"'L "'""*- *.
~~-- - - -------- .- ..............
pension speed from the correlation with what you got from
the experimental value.
c) Draw N / (P - p )0 45 vs. dp (gm) for two different im-
pellers and both fluids. In each curve, how can you estimate
the just-suspension impeller speed for the glass beads with
diameter 2000 nm?
Which of these impellers would you use for suspension?
Why? See Figure 6
The pitched-blade turbine is preferred in solid-liquid sus-
pension application. Generally, axial flow impellers are more
efficient than radial flow impellers to suspend the solids.
As mentioned, two-phase mixing is involved in numerous
industrial processes, hence, a general knowledge of these
operations is essential for the chemical engineer. These experi-
ments are designed to allow students to become acquainted
with basic understanding of important parameters that affect
liquid-liquid and solid-liquid mixing. Analyzing their results
in a critical manner, and answering specific questions, will
help the students to investigate the behavior of different im-
pellers in different mixing situations. All this should provide
students with criteria to choose the best mixing system for a
a interfacial area per unit volume [m 1]
C20 coefficient of variation dimensionlesss]
d32 Sauter mean drop diameter, general use [m]
d nominal diameter of drops in size class i [m]
dmn droplet mean diameter [m]
d particle size or diameter [m]
D impeller diameter [m]
Fr Froude number [-]
g gravitational acceleration [m/s2]
m number of size classes representing drop size distribution
M corrected torque [N.m]
Mm measured torque [N.m]
M residual torque [N.m]
n number of drops in size class i
N impeller speed [rps]
N minimum impeller speed to just suspended solid particles
in vessel [rps]
N minimum impeller speed to suspend liquid drops in vessel
N flow number
P power [W]
Re Reynolds number
Remp impeller Reynolds number
S Zwietering constant
t contact time between two colliding drops [s]
0 1000 2000 3000 4000
Figure 6. Ns / (ps-p'45 vs. d as a function of
concentration of solid particles.
T tank diameter [m]
V volume of tank [m3]
X mass ratio of suspended solids to liquid [kg solid/100 kg
R� viscosity of continuous phase [Pa.s]
Ld viscosity of dispersed phase [Pa.s]
[i bulk viscosity of liquid-liquid mixture [Pa.s]
v kinematic viscosity of the liquid [m2/s]
po density of continuous phase [kg/m3]
p, density of liquid [kg/m3]
Ps density of solid or particle [kg/m3]
P bulk density of liquid-liquid mixture [kg/m3]
Ap = Pd c density difference between phases [kg/m3]
o interfacial tension [N/m]
P volume fraction of dispersed phase
1. Armenante, PM., Y.T. Huang, and A.T. Li, "Determination of the
Minimum Agitation Speed to Attain the Just Dispersed State in Solid-
Liquid and Liquid-Liquid Reactors Provided With Multiple Impellers,"
Chem. Eng. Sci., 47(9-11), 2865 (1992)
2. Ascanio, G., R. Legros, and PA. Tanguy, "A Fluid Mixing Laboratory
for Chemical Engineering Undergraduates," Chem. Eng. Ed., 37(4),
3. Skelland, A.H., and R. Seksaria, "Minimum Impeller Speed for Liquid-
Liquid Dispersion in Baffled Vessels, "Ind. Eng. Chem. Process Des.
Dev., 17, 56 (1978)
4. Paul, E.L., V.A. Atiemo-Obeng, andA.M. Kresta, Handbook oflndus-
trial Mixing, John Wiley & Sons, Inc. (2004)
5. Uhl, V.W, and J.B. Gray, Mixing Theory and Practice, Vol. 1 and 2,
Academic Press, London (1966)
6. Harnby, N., M.E Edwards, and A.W Nienow, Mixing in the Process
Industries, Antony Rowe, Great Britain (1985)
7. Zwietering, T.N., "Suspending of Solid Particles in Liquid by Agita-
tors," Chem. Eng. Sci., 8(3-4), 244 (1958)
8. Baldi, G., R. Conti, and E. Alaria, "Complete Suspension for Particles in
Mechanically Agitated Vessels," ( ...I.. i . Sci., 33(1), 21 (1978) 1
Chemical Engineering Education
--- - ^ K.___________________________-
CONCEPTESTS FOR A
JOHN L. FALCONER
University of Colorado * Boulder, CO 80304
McDermottE11 found that few students developed a
functional understanding of the material in an in-
troductory physics course. Instead, students entered
the class with misconceptions, and these misconceptions were
not displaced by lectures. McDermott stated that students must
be allowed to apply their own ideas, so that when they fail,
students are more likely to learn the correct concepts. She
found that "an effective instructional approach is to challenge
students with qualitative questions that cannot be answered
through memorization." This is because students in science
and engineering courses often solve quantitative problems
by memorizing an algorithm; as a result they do not obtain a
functional understanding, which means they cannot use their
knowledge in new situations.J11
A recent paper[21 discussed the use of conceptests in a chemi-
cal engineering thermodynamics class. These conceptests
show that many students don't understand basic concepts
that they used in previous courses, such as vapor pressure
(Example 1) and the ideal gas law (Example 2). The use of
conceptests and clickers will be described briefly, and then ex-
amples of conceptests for thermodynamics will be presented.
These conceptests replace much of the lecture. Explaining
the motivation for clickers at the beginning of the semester is
important. Also, few quantitative problems are done in class,
but the students do at least as well on quantitative problems
on exams. The students answer multiple-choice, qualitative
questions using small, hand-held clickers. Students then
discuss their answers with fellow students. We used infrared
clickers for the first three years, and in 2006 we used RF
clickers,[41 which cost the same but are much faster and do
not require any classroom installation.
The motivation for using clickers'51 with conceptual ques-
tions[6 7] is to increase students' conceptual understanding
and to change their misconceptions. This approach provides
John L. Falconer is a professor of chemical
and biological engineering at the University
of Colorado in Boulder. He is a President's
Teaching Scholar at the university, which
is the university's highest teaching rec-
ognition. His research interests are in
heterogeneous catalysis, photocatalysis,
adsorption and diffusion in zeolites, and
zeolite membrane synthesis, characteriza-
tion, modeling, and separations. He has
published more than 180 papers in refereed
journals, and has five patents. He received
his Ph.D. in chemical engineering from Stanford University in 1974, and
has been at the University of Colorado since 1975.
@ Copyright ChE Division of ASEE 2007
Vol. 41, No. 2, Spring 2007
instant feedback on student understanding from everyone in
class, so that the instructor can use class time to concentrate
on clearing up confusing concepts. It creates a more-engaged
learning environment and it allows students to determine how
well they understand key concepts. It also allows them to
apply their own ideas and to learn from and teach their fel-
low students. Students thus hear explanations that may differ
from those presented by the instructor; a student who recently
learned the concept may often better understand what confuses
a fellow student. Conceptests and clickers also make teaching
more enjoyable for the instructor. Students like this mode of
instruction: Class discussions are more lively, attendance is
higher, and students are more motivated to be prepared.
CLASSROOM USE OF CLICKERS
An approach for using clickers and peer instruction[5 6] is
1. The instructor poses a multiple-choice conceptual question.
2. Students, working alone, enter an answer using their
individual clickers, and the answers are collected by the
instructor's computer. The clickers have unique ID num-
bers, which can be registered at the start of the semester.
3. After a few minutes, the instructor pauses the acquisition
of answers, and uses the histogram of answers on his/her
computer to determine the students' level of understand-
ing. The students are not shown the histogram. A good
question typically has less than 50% correct answers.
4. The instructor then asks students to discuss their answers
with neighbors (peer instruction6]1, and asks them to
either change their answers or convince their neighbors
who disagree with them to do so. The instructor can walk
around the classroom to answer questions and determine
how students are doing.
5. After a few minutes, the instructor looks at the histo-
gram of answers again. If most students have the correct
answer, he/she either explains why it is correct or asks
students to explain why.
6. If the students have not converged on the correct answer,
the instructor can try to clarify the question or spend
more time on the concept.
We typically used two to four clicker questions in a 50-
minute class. The student answers are graded to provide
motivation, with three points for the correct answer and two
points for a wrong answer. The five days with the lowest
averages for each student were not counted in grading. The
clicker grade counted 10% of the total course grade, and the
class average has been close to 90%. This grading provided
sufficient motivation for students to attend class and to be
prepared. To encourage cooperation instead of competition,
the course was not graded on a curve; a fixed grading scale
was presented at the beginning of the semester.
Reading was assigned for each class to prepare students
for in-class conceptests. A graded reading quiz, consisting
of two to four questions, had to be completed online before
class using WebCT software. Each homework assignment
and each exam had conceptual questions similar to those in
class, but they were not multiple choice, and the students had
to provide explanations for their answers. Approximately 40%
of each exam grade was from conceptual questions. The exam
questions were different from those in class, but they tested
the same concepts.
DEVELOPING CLICKER QUESTIONS
The type of clicker questions used can make a significant
difference in students' learning and motivation. Some sug-
gestions for effective clicker questions are:
* Use conceptual questions, not numerical calculations.
* Identify the concepts that are most confusing.
* Develop good wrong answers that bring out student
misconceptions. Previous exams are a good source of
* Aim for 50% or less correct answers; if 90% are correct
initially, the question is probably not a good use of class
As shown in the examples for chemical engineering ther-
modynamics, additional conceptests can be developed with
small variations on a conceptest by:
* Increasing instead of decreasing a variable
* (C i, ,, ,, a different variable
* Using aflow system instead of a static system
* Using a liquid-to-solid instead of gas-to-liquidphase change
* Holding a different variable constant
* Using a different method to change the same variable
Asking students to choose between several graphical rep-
resentations is an effective conceptest, such as asking which
process on a T-S diagram is correct (Example 8), or which
curve is incorrect on a P-H diagram (Example 9), or how
does the mole fraction change with pressure (Example 12).
Using this approach, approximately 275 conceptests have
been developed for a chemical engineering thermodynamics
course, and PowerPoint versions of them can be obtained
Anonymous student feedback has been positive. Of the 53
students in the fall 2005 thermodynamics course, 38 specifi-
cally mentioned in an anonymous end-of-semester evaluation
that they liked clickers and conceptests. One student did not
like them and said that the lectures rely too heavily on click-
ers. One comment indicated that students have not always had
good experience with clickers: "Clickers, surprisingly, were
Chemical Engineering Education
The correct answers are underlined.
Question 1. Liquid water is in equilibrium with air at 50 "C in a piston/cylinder system at 1 atm pressure. The total pressure
is raised to 2 atm by pushing on the piston. Temperature is constant. At equilibrium, the partial pressure of the water:
C. Remained the same
When this question was given as an online quiz at the beginning of the semester in 2005 (ungraded, but
credit given for completing the quiz), only 14 students out of 51 selected the correct answer (C), and 22
selectedA. Since chance would result in 17 correct answers, this question indicates students have signifi-
cant misconceptions about vapor pressure. This question was also put on the final exam, but it was not
multiple choice. Instead, students had to describe what happened and why. On the final exam, 35 students
answered correctly with the correct explanation. This was more demanding than the online question at
the start of the semester since they had to provide an explanation.
Variations on this question: Instead of a pressure increase to 2 atm, the students can be asked what happens if: va por
a) Total pressure decreased to 0.75 atm
b) Liquid water added to system at constant T, P
c) Water vapor added at constant T, P
d) Air added at constant T, P
e) Air removed at constant T, P
f) Water vapor and air removed at constant T, P
The same concept can be used in a flow system, where the total pressure is lowered. Instead of asking what happens to
the partial pressure of water, the question can ask if the amount of water in the vapor phase (or liquid phase) increases or
Question 2. A piston/cylinder system at 45 �C contains 0.9 mol hexane vapor and 0.1 mol hexane liquid at equilibrium.
The pressure was doubled at constant temperature. What is the final state of the system?
A. The system contains all vapor
B. The system contains all liquid
C. Some of the liquid had vaporized
D. Some of the liquid has condensed
When this question was given as an online quiz at the beginning of the semester in 2006, only 5 students out of69 correctly
selected B. More students (9) selected C; i.e., they answered that liquid vaporized when the pressure increased. Note that all
these students had done multiple problems dealing with vapor pressure in a prerequisite course.
Question 3. A constant-volume tank contains CO2 at 2 atm. Nitrogen is injected into the tank. What
happens to the partial pressure of CO2 if it all remains in the tank? Assume ideal gases.
C. Stays the same
When this question was given as an online quiz at the beginning of the semester in 2005, only 6
students out of 51 correctly selected C. Most students (32) selected B, demonstrating how prevalent
misconceptions are about a basic concept that was used extensively in several previous courses. ,
Vol. 41, No. 2, Spring 2007 10
Question 4. An explosion takes place in a closed tank that is completely insulated. What happens to the energy of the
system (tank plus contents)?
A. Stays the same
C. Increases Insulation
D. Insufficient information
When this question was given as an online quiz at the beginning of the semester in 2005, ., ..m
only 9 students out of 51 correctly selected A, whereas 37 selected B.
Variations on this question
a) Use endothermic reaction instead of exothermic.
b) Make it isothermal instead of adiabatic.
Heat of Reaction
Question 5. Which has the higher adiabatic temperature for an oxidation reaction? A feed that contains
A. a stoichiometric amount of pure 02
B. 50% excess oxygen (pure)
C. a stoichiometric amount of air
Variations on this question:
a) Use an endothermic reaction and change the amount of inert in the feed.
b) Ask which has higher temperature if the flow rate to the reactor is cut in half.
Question 6. Solid water becomes a liquid when the pressure increases isothermally. The entropy of the water:
C. Does not change
Variation on this question
Use ethanol so that liquid becomes solid as pressure increases.
Question 7. Adiabatic expansions and compressions are shown in the two graphs. One curve represents a reversible
process and one an irreversible process in each figure. Which curves are for the irreversible processes?
A.1&3 B. 1&4 C.2&3 D. 2&4
10 Chemical Engineering Education
Question 8. Which diagram corresponds to a Carnot engine in which both adiabatic steps in the cycle are irrevers-
ible? Correct answer: (c)
4 -4 3
4 S 3
4 - 3
Question 9. Which diagrams are not possible for a vapor compression refrigeration cycle using a throttling valve?
A. 1 B. 1.2 C. 2 D. 2,3 E. 3
actually used effectively; they are almost always otherwise
a complete waste of time."A few additional sample student
"The most conceptually challenging ChE course I have had
and the class I have been the most consistently motivated
"Really liked lectures, not rigid, butfocused on what we
had trouble understanding."
"Most effective methods of teaching were concept tests."
"Concept tests make me think .:- . ., ii, and completely
about every subject."
"I know a lot more now than I ever dreamed I would know.
One of the biggest learning techniques was concept tests."
1. McDermott, L.C., "Oersted Medal Lecture 2001: Physics Educa-
tion Research: The Key to Student Learning," Am. J. Phys. 69, 1127
2. Falconer, J.L., "Use of ConcepTests and Instant Feedback in Thermo-
dynamics," Chem. Eng. Ed., 38, 64 (2004)
4. Free software that acquires student responses
and grades the results is provided by iclickers. The receiver connects
to a USB port and contains a small LCD screen so that the instructor
can see the student responses.
5. Duncan, D., Clickers in the Classroom, Addison Wesley (2005)
6. Mazur, E., Peer Instruction, Prentice Hall (1997)
7. Landis, C.R., A.B. Ellis, G.C. Lisensky, J.K. Lorenz, K. Meeker,
and C.C. Wamser, ( ,-...... ,, ConcepTests: A Pathway to Interactive
Classrooms, Prentice Hall (2001)
8. Contactjohn.email@example.com 1
Vol. 41, No. 2, Spring 2007
Binary Vapor-Liquid Equilibrium
Question 10. Your technician reported that when he decreased the temperature for a binary gas-phase mixture of C6
isomers, n-hexane condensed before 2,2 dimethylbutane (DMB). Can this happen?
A. Yes, if n-C6 has a lower vapor pressure than DMB
B. Yes, if n-C has a higher vapor pressure than DMB
C. No. because both species have to condense
D. It depends on the system pressure as to whether one or two species condense
When this question was given as an online quiz at the beginning of the semester in 2006, only 1 student out of 69 cor-
rectly selected C. Note that all students have solved multiple Raoult's Law problems in a prerequisite course.
Variations on this question
a) Increase pressure instead of lowering temperature.
b) Start with liquid and raise temperature and state that only one component evaporates.
c) Start with liquid and lower pressure and state that only one component evaporates.
Question 11. One mole of pure hexane is in vapor liquid equilibrium at 1 atm and 70 �C in a piston and cylinder. After
0.2 mol of heptane liquid is injected, the system returns to equilibrium at the same T and P. Psat(hexane) > Psat(heptane)
What are the final contents of the system?
A. All liquid
B. All vapor
C. Liquid and vapor with yhee > hxane
D. Liquid and vapor with yhexne < Xhexane
Variations on this question
a) Inject heptane vapor instead of liquid.
b) Start with pure heptane and inject hexane liquid or vapor.
Question 12. The pressure increases for 50/50 vapor mixture of A and B initially at low pressure. The liquid is ideal.
Which plot corresponds to how xA and YA change with pressure? PAsat > P at
Correct answer: (b)
Variation on this question
Start with liquid and raise the temperature and show similar plots vs. temperature.
1.0 A XA 1.0 B 1.0 C xA
0.5 0.5 A XA 0.5
0 0 0
12 Chemical Engineering Education
Entropy of Mixtures
Question 13. Two ideal gases are mixed isothermally. For which of the fol-
lowing does the entropy of the gases increase?
A. constant V
B. constant P
C. both constant V and constant P
Question 14. Organic molecules adsorb in the pores of zeolites. Consider adsorption from hexane liquid and vapor. The
adsorbed hexane concentration in the zeolite pores is:
A. higher in A
B. higher in B
C. the same in both
D. insufficient information
Variations on this question
(a) Use a binary mixture of hexane and acetone, with acetone
enriched in the gas phase. Ask which zeolite has a higher
concentration of adsorbed acetone.
(b) Use a binary mixture of hexane and acetone, with acetone enriched in the gas phase. For system A, make the liquid 50/50
and for system B make the vapor 50/50. Ask which zeolite has a higher concentration of adsorbed acetone.
Question 15. Two identical flasks at 45 �C are connected by a tube. Flask A contains water, and B contains 50% more
water plus it contains ethanol. What happens as the system approaches equilibrium?
A. Water moves from A to B
B. Ethanol moves from B to A
C. Water moves to B and EtOH to A
D. Both water and EtOH move to A
E. No change in levels
Variations on this question
a) Both flasks contain only water, but at different temperatures.
b) Both flasks contain only water at the same temperature, and then salt is added to one.
Question 16. A can of soda at 0 C contains liquid water with a low concentration of dissolved CO2. The gas-phase CO2
pressure is 1.5 atm. Compare fugacities in the liquid phase.
A. Water fugacity is higher
B. CO2 fugacity is higher
C. Water and CO, fugacities are the same
Vol. 41, No. 2, Spring 2007 11.
Question 17. For this reaction, Pd and PdO do not mix in the solid phase:
Pd(s) + 0.5 O,(g) - PdO(s)
This reaction goes to equilibrium at 400 �C in a closed container for two starting conditions:
(a) 100 g Pd, 1 mol 02
(b) 100 g Pd, 1 mol 02, 1 g PdO
The initial 0, pressure is high enough to reach equilibrium. Which statement is correct?
A. The 02 pressure is higher for (a)
B. The 02 pressure is higher for (b)
C. The 02 pressure is the same for both conditions
Question 18. Calcium carbonate decomposes according to:
CaCO3(s) - CaO(s) + CO2(g)
10 mol 0.2 mol 10 mol initial conditions
At equilibrium, you push down on the piston until the volume is half the original volume.
Temperature is constant. What happens?
A. CO2 pressure almost doubles
B. CaO and CO, react; CO, pressure does not change
C. At equilibrium so nothing changes
D. All the CO2 reacts
Variation on this question
Start with 10 mol of CaO instead of 0.2 mol.
Chemical Engineering Education
M]n1= class and home problems
GO TO THE MOVIES
(Stimulating Problems for the Contemporary
JIMMY L. SMART
University of Kentucky * Paducah, KY 42002
Coarse, unbleached pages filled with dull text, sparsely
interspersed with boring graphs. That accurately
describes many of my undergraduate engineering
textbooks in the 1960s. My, how times have changed! Today,
in an environment with Xboxes, cellular telephones, digital
TVs, iPods, and BlackBerries, it is increasingly difficult to
attract and hold the interest of undergraduate students.
I try to engage students in my classroom actively, with
generally mixed results. We work through classroom theory
and equation derivations. I really gain their attention, however,
when I relate ideas presented in their text with my real life
experiences in industry. Another way to spark their interest
is to prepare challenging and interesting class and homework
problems. Today, many textbook problems are dull and life-
less. For more fun, I introduce test problems centered around
popular movies the students have probably watched. Students
Vol. 41, No. 2, Spring 2007
find these test problems memorable. Over the years, I have run
into former students who, during the course of our conversa-
tion, recall and chuckle over some of my "movie problems"
they remembered on a final exam.
� Copyright ChE Division of ASEE 2007
The object of this column is to enhance our readers' collections of stimulating problems in
chemical engineering education. Ideal problems, which may be "open-ended," are those that mo-
tivate the student either by the novel illustration of a particular principle, or by the elucidation of
a difficult concept in a more traditional setting. Practical relevance is encouraged. The text portion
of a manuscript (excluding figures) should not normally exceed 10 double-spaced pages (about
2,500 words). Please send manuscripts to Professor James O. Wilkes (e-mail: wilkes@umich.
edu), Chemical Engineering Department, University of Michigan, Ann Arbor, MI 48109-2136.
Preliminary ideas may be discussed with Prof. Wilkes before submitting a manuscript.
Jimmy Smart is an associate professor
of chemical and materials engineering at
the University of Kentucky. He received
his B.S. from Texas A&M and his M.S
and Ph.D. from the University of Texas
at Austin, all in chemical engineering. He
has more than 20 years industrial experi-
ence with companies such as IBM and
Ashland Chemical. His research areas
include applications of membranes to
purify water supplies and treatment of
The following five problems are representative of some
of the "movie problems" that I have used on tests in various
courses, including reactor design, heat transfer, mass transfer,
engineering economics, and fluid mechanics. These problems
tend to be open-ended. They can be challenging and can of-
ten be worked a variety of ways-giving different answers,
depending upon what basis assumptions were made. It is not
necessarily expected that students get these problems 100%
correct; some problems are more reasonably worked than oth-
ers. What is interesting is the students' approach to the prob-
lems-their thinking process. After all, the thinking process
is most of what they will retain from their college experience,
not whether the Hagen-Poiseuille equation is most appropriate
for laminar fluid flow rather than turbulent flow.
Solutions to these five problems can be viewed by sending
me an e-mail firstname.lastname@example.org.
Problem 1. Reactor Design"'
Spider-Man 2 (2004)
Spider-Man is a comic book/movie hero who generates spi-
der silk (fibrous biopolymer filaments) for a variety of reasons,
including use of swing lines for rapid movement between city
skyscrapers. Formation of these biopolymer filaments is the
result of an enzymatic reaction involving secretion of amino
acids within a tubular gland located in his arms. The filament
is extruded through spinnerets located at the tip of each of
these flexible tubes. We will model these tubular glands as
mini-PFRs. Assume Spider-Man can consciously control the
length of his tubular glands to achieve the necessary conver-
sion for generation of strong polymeric fibers.
Spider-Man is poised to complete a long swing, down
through a major thoroughfare of Manhattan. He is preparing
to extrude a suitable filament to support his swing. From
previous experience he knows he will need at least a 68.6%
conversion of the amino acid reactants to support his weight
along with mechanical stresses associated with the swing.
Model the conversion of amino acids, A, as a gas phase
reaction occurring in a PFR. The irreversible gas-phase
reaction(1) A k B,
reaction(2) 2A k2 C,
where rA = kiAPA
where r 2=k
where r2A = k2APA
where reaction (1) is a competing reaction leading to a matrix
stiffener product, B, and reaction (2) is the desired reaction
leading to the biopolymer filament, C. Pure A enters the
base of each gland at a rate of 5 gmol/s at 300 K and 1.0 atm
pressure. The exiting molar flow rate through each tube is
comprised of the following products: matrix stiffener, 0.408
gmol/s, and biopolymer, 1.51 gmol/s.
Calculate the required length of Spider-Man's tubular
gland to achieve a minimum conversion of 68.6 % of A.
In lieu of a math solver, use average reaction rate constants
and average partial gas pressures to obtain a rough estimate
of glandular length.
AH1 = - 5,000 J / mol B
750 J/ mol C
Cp, = CpB = Cpc = 1.04J /g K
Average molecular weight of the reaction mixture = 14.0
k, = 0.0015 gmol/cm3 s atm at 300 K, with E = 9,900 cal/mol
k, = 0.07 gmol/cm3 s atm2 at 300 K, with E, = 1,500 cal/mol
Problem 2. Heat Transfer"2
Pirates of the Caribbean,
The Curse of the Black Pearl (2003)
Captain Jack Sparrow
(Johnny Depp) is on board
the English ship, the Inter-
ceptor, in pursuit of his
old ship, the Black Pearl.
In an effort to stop the
ship, Captain Jack's crew
heats cast iron cannonballs
to 2000 �F and fires them
at the Black Pearl. One
such 6-inch diameter ball
was fired (remaining in the
air for 65 seconds) and a
lucky shot landed the ball
into an insulated barrel (42 gal) of kerosene on the deck of
the ship. If the ambient air and kerosene temperature were
Chemical Engineering Education
30 �F and kerosene spontaneously combusts at 325 �F (entire
barrel contents must attain this temperature),
(a) Does the barrel burst into flames?
(b) What is the final temperature of the kerosene?
Properties of cast iron are r = 23 Btu/hr ft �F, C
Btu/lb �F, and SG = 7.4
Properties of kerosene are C = 0.5 Btu/lb �F and SG
Assume (h),,ball = 16 Btu/hr ft2 �F
Problem 3. Mass Transferm'
You are the chief military engineer with William Wallace
(played by Mel Gibson), who is in the middle of the siege of
an English castle. The castle is surrounded by a moat that is
10 ft wide X 10 ft deep X 0.5 mile outside diameter. Wallace
has decided to blockade the castle and wait until the water
level in the moat has been reduced by evaporation to a level
of 5 ft deep. At that point, men and equipment can breach the
moat and attack the castle.
120 I I I I
y = 84.7 - 0.059x + 9.20 x 105 x2-2.821 x 108x3
- 0 -
I- 85 -
6 5 . . . . . . . . . .L. . . . . . . . . L. . L .
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Military Time (0600 = 6 am, 2000 = 8 pm, etc.)
The bottom of the moat is clay-lined and there is no wind
during the daylight hours (6 a.m. - 6 p.m.). At night, a steady
south wind of 10 mph blows across the surface of the moat
from 6 p.m. - 6 a.m. The country is in the middle of a seven-
year drought. The daily air temperature varies according to
Figure 1. Assume the temperature of the water remains 20 �F
less than the air temperature and radiation losses to the night
sky bring the water back to the same point (65 �F) at 6 a.m.
Assume the relative humidity of the surrounding air remains
constant at 10%. Provide to Wallace your estimate as to
how many days before he can storm the castle and win
the victory for Scotland.
Use the following heat convection correlations and mass
transfer analogy to prepare your estimate:
Natural convection for horizontal plate (cold surface facing
up): NuL = 0.27 RaL14
Forced convection for horizontal plate:
laminar flow, Nu = 0.332 Re "Pro
turbulent flow, Nu = 0.0360 Reo s8 Pro
Antoine equation for water: In p"' = 11.64 - [rS4 I I' +
375.5)], where p (=) atm, T (=) �F
The following properties are for air at T,:
S= 0.0153 Btu/hr ft�F
gPp2 /2 = 1.96 x 106(F ft3)
Pr = 0.706
v=0.17x10 3ft2 s
DAB =2.8 x 10 4f2 /s
p = 0.076 lb / ft3
C = 0.24 Btu lb OF
Vol. 41, No. 2, Spring 2007
Figure 1. Variation of weather temperature
during a typical 24-hour day.
Problem 4. Fluid Mechanics'31
Master and Commander: The Far Side of the World
In the movie Master and Commander: The Far Side of
the World, it is 1805 and Captain Aubrey (played by Russell
Crowe) is trying to outmaneuver a large French warship, the
Acheron. At one point in the movie, the French are in hot
\ turbulent 2 x 10
Turbulent \ 1 x 10o3
% 5xl 4
2 X 10. 4
1 x 10to
0.002 xlO-b I
I Laminar Turbulent
j >"' smooth plate
105 106 107 108 109
Figure 2. Friction drag coefficient for a flat plate.
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
Froude number (Fr) = V/(e g)0.5
Figure 3. Relationship of drag coefficient
along boat hull with Froude number.
0 1 1 1 1 1 1 1 1 I I I I I I I 1 1 I I I I
400 800 1200 1600 2000 2400
Military time (0800 = 8 am, 2000 = 8 pm, etc.)
Figure 4. Variation of wind velocity throughout a typical
Chemical Engineering Education
pursuit of his smaller English warship, the Surprise. So far,
the ships have been in a dead heat and Captain Aubrey cannot
seem to lose his attackers. He figures that if he can just get
his little ship to go a little faster (greater than 13 knots or 15
mph), he can outdistance the French. Using Figure 4, what
is the earliest time of the day that he can expect to begin
to pull out of range of the French guns?
Model the ship as a V-shaped structure with a draft (depth
in the water) of 17 ft. The length of the ship is 85 ft, width is
30 ft, and total available sail area is 5 X 104 ft2. Use Figure
2 to estimate friction drag.
The total drag on the ship, tt = + Q + m, is equal
to friction drag + pressure drag + wave-making drag. Esti-
mate pressure drag to be 1.5 4and wave-making drag is a
function of the Froude number, Fr (see Figure 3). To simplify
calculations (instead of integrating along the length of the
ship), take an average between "near bow values" and "near
stern values" for 4and m .
Force on the sails can be calculated from Newton's 2nd Law
as Fs = PairA (Vw n)2. Assume an ambient temperature of 65
�F and the molecular weight of air is 29 lb/lb mole. Neglect
acceleration and inertial effects of the ship.
Problem 5. Engineering Economics"41
Forrest Gump (1994)
Forrest Gump is the story of a man who, over the course
of three decades and despite having an IQ of only 75, leads
a most extraordinary life. After being discharged from the
Army, Forrest is joined by his former Army commander, Lt.
Dan, to start a new business venture. The business is named
"The Bubba Gump Shrimp Corporation," after Forrest's fallen
Army comrade, Bubba.
Forrest has attracted some potential venture capitalists to
invest in his new shrimp company. He has scheduled a meet-
ing for next week to present his financial case to a group of
The Bubba Gump Shrimp Corporation (BGSC) will need
$2.8MM start-up costs for equipment, warehouse space, utili-
ties, advertising, personnel, insurance, and distribution. It is
projected that O&M costs during the first year will be $180K
and increase by 6% for every year thereafter. The company
will have to come under labor union organization in the fourth
year and this will increase annual O&M costs by an additional
2%. Selling price for the standard 3 lb bag of frozen shrimp is
expected to be flat at $12 per bag. Sales projections start out
at 252 bags/day (at a stream factor of 0.8) the first year and
increase by 20% each year thereafter. During the lean years
(years one through three), arrangements have been made
with another shrimp company to use Forrest's equipment to
package some of their product. This will provide an additional
income of $40K each year. Assume the MARR is 9%, which
includes an inflation component of 4%. BGSC will have an
effective tax rate of 35% and will use MACRS-GDS (seven-
year property) depreciation allowances. Using a planning
horizon of 10 years and a salvage value of $0.9MM, complete
an economic analysis based upon then-current dollars.
(a) Calculate the present worth of Forrest's invest-
(b) What is his discounted cash flow/return on in-
(c) What is his discounted payback?
(d) Assume that Chicken-of-the Sea, Inc., wants
to get into the shrimp business and approaches
Forrest at the end of his fourth year of operation
and offers to buy him out. Prepare a rationale
for a reasonable selling price to be presented to
A area, ft2
CD drag coefficient, dimensionless
CDf drag coefficient due to friction, dimensionless
Cp constant pressure heat capacity, Btu/lb �F
Cp, constant pressure heat capacity per mass for component i,
D, diffusion coefficient for component A diffusion through
component B, ft2/s
-' drag, lb,
E activation energy for rate constant i, cal/mol
F force, lb
Vol. 41, No. 2, Spring 2007
Fr Froude number, dimensionless
g acceleration due to gravity, ft/s2
GDS general description system (depreciation method)
Gr Grashof number, dimensionless
h convective heat transfer coefficient, Btu/hr ft2 �F
k reaction rate constant, mol/cm3 s atm
- thermal conductivity, Btu/hr ft �F
L length, ft
Characteristic length, ft
MARR minimum attractive rate of return, %
MACRS modified accelerated cost recovery system (depreciation
NuL Nusselt number along a characteristic length, hL/ r
Nu Nusselt number along a flat plate length, hL/
O&M operating and maintenance
P, partial pressure of component i, atm
p"" pure component saturated vapor pressure, atm
PFR plug flow reactor
Pr Prandtl number, dimensionless
RaL Rayleigh number along a characteristic length, Gr X Pr,
Reynolds number along a characteristic length, dimensionless
reaction rate for ith reaction, mol/cm3 s
coefficient of thermal expansion, 1/T
heat of reaction, J/mol
kinematic viscosity, ft2 s
viscosity, lb/ft s
1. Rawlings, J.B., and J.G. Ekerdt, Chemical Reactor Analysis and Design
Fundamentals, Nob Hill Publishing, Madison, WI (2004)
2. Welty, J.R., C.E. Wicks, R.E. Wilson, and G.L. Rorrer, Fundamentals
ofMomentum, Heat, and Mass Transfer, 4th Ed., John Wiley & Sons,
New York (2001)
3. Munson, B.R., D.E Young, and T.H. Okiishi, Fundamentals of Fluid
Mechanics, 5th Ed., John Wiley & Sons, New York (2006)
4. White, J.A., K.E. Case, D.B. Pratt, and M.H.Agee, PrinciplesofEngineering
Economic Analysis, 4th Ed., John Wiley & Sons: New York (1998) 1
Chemical Engineering Education
HOW TO PREPARE NEW COURSES
WHILE KEEPING YOUR SANITY
RICHARD M. FIELDER
North Carolina State University
Education Designs, Inc.
Think of a two-word phrase for a huge time sink that
can effectively keep faculty members from doing the
things they want to do.
You can probably come up with several phrases that fit.
"Proposal deadline" is an obvious one, as are "curriculum
revision," "safety inspection," "accreditation visit," and "No
Parking." (The last one is on the sign posted by the one open
space you find on campus minutes before you're supposed
to teach a class, with the small print that says "Reserved for
the Deputy Associate Vice Provost for Dry Erase Marker
But the phrase we have in mind is "new prep"-prepar-
ing for and teaching a course you've never taught before.
This column describes the usual approach, which makes this
challenging task almost completely unmanageable, and then
proposes a better alternative.
THREE STEPS TO DISASTER, OR,
HOW NOT TO APPROACH A NEW COURSE
1. Go it alone. Colleagues may have taught the
course in the past and done it very well, but it
would be embarrassing to ask them if you can use
their materials (syllabi, learning objectives, lecture
notes, demonstrations, assignments, tests, etc.), so
instead create everything yourself from scratch.
2. Try to cover i , i % til,,it. known about the subject
in your lectures and always be prepared to answer
any question any student might ever ask. Assemble
all the books and research articles you can find and
make your lecture notes a self-contained encyclo-
pedia on the subject.
3. Don't bother making up learning objectives or
a detailed syllabus-just work ;ih, . , out as you
go. It's all you can do to stay ahead of the class in
your lectures, so just throw together a syllabus that
contains only the course name and textbook, your
name and office hours, and the catalog description
of the course; invent course policies and procedures
on a day-by-day basis; and decide what your learn-
ing objectives are when you make up the exams.
Here's what's likely to happen if you adopt this plan. You'll
spend an outlandish amount of time on the course- 10 hours
or more of preparation for every lecture hour. You'll start ne-
glecting your research and your personal life just to keep up
with the course preparation, and if you're unfortunate enough
to have two new preps at once, you may no longer have a
personal life to neglect. Your lecture notes will be so long and
dense that to cover them you'll have to lecture at a pace no
normal human being could possibly follow; you'll have no
time for interactivity in class; and you'll end up skimming
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 r
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
Rebecca Brent is an education consultant
specializing in faculty development for ef-
fective 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 200;
Vol. 41, No. 2, Spring 2007
some important material or skipping it altogether. Your poli-
cies regarding late homework, absences, missed tests, grading,
and cheating will be fuzzy and inconsistent. Without learning
objectives to guide the preparation, the course will be incoher-
ent, with lectures covering one body of material, assignments
another, and tests yet another. The students' frustration and
complaints will mount, and the final course evaluations will
look like nothing you'd want to post on your blog.
There's a better way.
A RATIONAL APPROACH TO
NEW COURSE PREPARATION
1. Start preparing as soon as you know you'll be teaching a
Dedicate a paper file folder and a folder on your computer
to the course and begin to assemble ideas and instructional
materials. While you're teaching the course, continue to file
ideas and resources as you come up with them.
2. Don't reinvent the wheel.
Identify a colleague who is a good teacher and has taught
the course you're preparing to teach, and ask if he/she would
be willing to share course materials with you. (Most faculty
members would be fine with that request.) In addition, try
finding the course on the MIT OpenCourseWare Web site
() and download materials from there.
Open courseware may contain visuals, simulations, class
activities, and assignments that can add considerably to the
quality of a course and would take you months or years to
construct from scratch. The first time you teach the course,
borrow liberally from the shared materials and note after each
class what you want to change in future offerings. Also con-
sider asking TAs to come up with good instructional materials
and/or inviting students to do it for extra credit.
3. Write detailed learning (l1/.. ii .. . give them to the students
as study guides, and let the ( I... i,.. ' guide the construction
of lesson plans, assignments, and tests.
Learning objectives are statements of observable tasks that
students should be able to accomplish if they have learned what
the instructor wanted them to learn. Felder and Brent recom-
mend giving objectives to students as study guides for tests,12
and show an illustrative study guide for a midterm exam.3
Before you start to prepare a section of a course that will
be covered on a test, draft a study guide and use it to design
lessons (lectures and in-class activities4) and assignments that
provide instruction and practice in the tasks specified in the
objectives. As you get new ideas for things you want to teach,
add them to the study guide. One to two weeks before the test,
finalize the guide and give it to the students, and then draw
on it to design the test. The course will then be coherent, with
mutually compatible lessons, assignments, and assessments.
Instead of having to guess what you think is important, the
students will clearly understand your expectations, and those
with the ability to complete the tasks specified in the objec-
tives will be much more likely to do so on the test. In other
words, more of your students will have learned what you
wanted them to learn. The objectives will also help you avoid
trying to cram everything known about the subject into your
lecture notes. If you can't think of anything students might do
with content besides memorize and repeat it, consider either
dropping that content or cutting down on it in lectures, giving
yourself more time to spend on higher-level material.
4. Get feedback during the course.
It's always a good idea to monitor how things are going
in a class so you can make mid-course corrections, particu-
larly when the course is new. Every so often collect "minute
papers," in which the students anonymously hand in brief
statements of what they consider to be the main points and
muddiest points of the class they just sat through. In addi-
tion, have them complete a survey four or five weeks into
the semester in which they list the things you're doing that
are helping their learning and the things that are hindering it.
Look for patterns in the responses to these assessments and
make adjustments you consider appropriate, or make a note
to do so next time you teach the course.
5. Do i.. i % l,,i. you can to minimize new preps early in your
career, and especially try to avoid having to deal with several
of them at a time.
Some department heads inconsiderately burden their newest
faculty members with one new prep after another. If you find
yourself in this position, politely ask your head to consider
letting you teach the same course several times before you
move on to a new one so that you have adequate time to work
on your research. Most department heads want their new fac-
ulty to start turning out proposals and papers in their first few
years and will be sympathetic to such requests. It might not
work, but as Rich's grandmother said when told that chicken
soup doesn't cure cancer, it couldn't hurt. 7
1R.M. Felder and R. Brent, - 1jl1 ., . speaking," Chem. Engr.
Ed., 31(3), 178-179 (1997),
2R.M. Felder and R. Brent, "How to teach (almost) anyone (almost)
anything," Chem. Engr. Ed., 40(3), 173-174 (2006),
ncsu.edu/felder-public/Columns/T, I . , 1. i...i . l.11 >.
3R.M. Felder, Study guide for a midterm exam in the stoichiometry
4R.M. Felder and R. Brent, "Learning by doing," Chem. Engr. Ed.,
37(4), 282-283 (2003),
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/
Chemical Engineering Education
--- - ^K__________________________-0
Through Stiction in Control Valves
in a Process Control Class
RANGANATHAN SRINIVASAN, RAGHUNATHAN RENGASWAMY, AND SANDRA HARRIS
Clarkson University * Potsdam, NY 13699
industrial ,nii %\' ' in the last decade have indicated
that only about one-third of industrial controllers provide Ranganathan Srinivasan received his
BE. in electronics and instrumentation
acceptable performance. Performance demographics of Bengineering from Annamala University,ation
26,000 PID controllers, collected across a wide variety of his M.Tech. in systems and control from
processing industries in a two-year time span, indicate that the IIT Bombay, and his Ph.D. in chemical
engineering from Clarkson University. He
performance of 16% of the loops can be classified as excellent, currently works for Honeywell Automa-
16% as acceptable, 22% as fair, 10% as poor, and the remain- tion and Control Labs in Phoenix, and is
actively pursuing research in the areas
ing 36% are in open loop.[3.4] Since PID controllers constitute of Loop Performance Monitoring and
97% of all industrial controllers, poorly performing loops pose Nonlinear Control Strategies.
significant problems with huge financial implications. Hence, Raghunathan Rengaswamy is an as-
controller performance assessment (CPA) is an important sociate professor in the chemical and
area that is worthwhile to introduce in undergraduate process biomolecular engineering department
at Clarkson University. He received his
control curriculum. There are a number of articles that have B.Tech. from IIT Madras and his Ph.D.
discussed approaches for introducing control advances made from Purdue University, both in chemi-
cal engineering. His research interests
in multivariable, nonlinear, and distributed parameter systems include Fuel Cells and Process Systems
in the undergraduate curriculum.[5-9] This article proposes the Engineering.
introduction of CPA in an undergraduate classroom through
the use of a nonlinear phenomenon in control valves that
leads to oscillations. -
Performance degradation in control loops manifests it- Sandr the armet of associate parofiosso
self as: poor set point tracking, poor disturbance rejection, lecular Engineering at Clarkson University.
excessive final control element variation, or oscillations in She received her B.S. and M.S. from MIT,
and her Ph.D. from UCSB, all in chemi-
measurement signals. Sustained oscillations in control loops cal engineering. Her research interests
can be due to multiple reasons: are in the areas of Process Control and
1. Valves with high static friction (also known as stic-
tion). Presence of dead band and hysteresis in valves Copyight hEDiision of ASEE 2007
Vol. 41, No.2, Spring 2007 12.
can also cause limit cycles in ., it. ,1,1, processes.
2. Poorly tuned controllers in nonlinear processes with
3. Insufficient digital resolution (qi, , r -,,, effects in
data acquisition cards).
4. Controller saturation.
5. Oscillations that are external to the loop.
6. A combination thereof.
Reasons for oscillations are summarized in Figure 1.
CONTROLLER PERFORMANCE ASSESSMENT
In industrial controllers, routine operating data archived for
each PID loop includes-but is not limited to-Controller
Output (OP), Process Output (PV), Set Point (SP), loop type,
and controller settings. This archived data can be used to iden-
tify potential areas of improvement, trends, and problems in an
incipient fashion for preventative maintenance. As mentioned
before, industrial surveys over the last decade have indicated
that the performance of nearly two-thirds of all controllers
can be improved. In light of this, as envisioned by Kozubo101
and many others, several CPA tools that can automatically
detect, diagnose, and, if possible, improve the performance of
problematic control loops are being developed (see Reference
11 for a survey and analysis of commercial products being
developed). A specific set of requirements for such CPA tools
from the authors' perspective is listed below:
1. Automated computation procedure that can evaluate
approximately 1,000 loops or more a day.
2. The CPA tool must use noninvasive techniques.
3. Minimal use of process knowledge, as it might be
infeasible to build or maintain a knowledge base for
several thousand loops.
4. Acceptable false alarm and detection rates.
5. The i,. ., , i... used must be theoretically sound,
modestly complex, and efficient.
6. Problem loops should be detected and reported using
routine P' i,, information.
7. It should be possible to diagnose the possible causes)
forperformance , i, -,,i.i..,n
8. Suggest and implement corrective action (where ap-
plicable) to uisi,.n the cause ofpoor performance.
Objectives 1,2,3, 5, and 6 have been adequately addressed
using information technology and advanced computational
platforms. Statistics for objective 4 have not been reported
in the open literature to date. Objective 7, i.e., diagnosing
the cause for poor performance, has received considerable
Most present-day CPA tools assess control loop perfor-
mance using some variant of the Minimum Variance Control
(MVC)[251 concept. In this approach, the minimum error
(between the set point and the measurement) realizable by
any controller is bounded by the performance of the MVC.
Harris�261 showed that a lower bound, on a closed-loop
output variance realized using MVC, can be obtained by
analyzing routine operating data, provided the process dead
time is known a priori. A normalized index for assessment
of feedback controller performance determined against a
benchmark of MVC was then introduced by Desborough
and Harris.[27 The normalized controller performance index
is given below:
v(b) = 1 2 (1)
L- Dead band
- Hardware |
L Quantizer, saturation
- Controller tuning
Upstream Oscillating loop
(can be due to any internal
cause in an upstream loop)
L 30% of all loops oscillate
*10-20% of all loops oscillate due
to Static Friction or stiction
L Performance improvement
where 02m is the output variance that can be
achieved using MVC calculated by solving a
system of linear equations generated based on
a dead time b; 02 is the variance of measured
output; and i2 is the mean-squared deviation
from the set point. The index �(b) is bounded by
[0, 1], while � z 0 indicates the best achievable
performance, i.e., minimum variance control, and
S1 indicates poor performance, showing that
returning of the controller may be necessary. Over
the last decade, there have been a number of other
academic investigations on the development of
robust performance indices.["1
While the normalized index can be used to
identify poorly performing loops, it provides
very little diagnostics toward ascertaining the
root cause of oscillations. In addition to this, a
process engineer (or control engineer) responsible
for more than 400 loops may not have the time
to diagnose the problem and improve each poorly
Chemical Engineering Education
Figure 1. Common causes for control-loop oscillations.
Teaching stiction in undergraduate classes has a number of advantages.
Other than its obvious importance, teaching stiction helps introduce
the concept of CPA to undergraduate students.
performing loop. Hence, it may not be possible to choose
an appropriate corrective action just based on performance
indices alone. The last decade has seen an increasing interest
in the area of detecting and diagnosing the cause of oscilla-
The last objective, i.e., suggesting corrective action, is gain-
ing importance. Corrective actions can include: identifying
new controller parameters,[281 valve maintenance or stiction
compensation,[29 30] eliminating upstream disturbances, or a
Focus of This Article
It has been reported that 20% to 30% of all control loops
oscillate due to valve problems caused by static friction
(also called stiction).1 3] Stiction is a real industrial process
control problem. Teaching stiction in undergraduate classes
has a number of advantages. Other than its obvious impor-
tance, teaching stiction helps introduce the concept of CPA
to undergraduate students. Stiction phenomenon can also be
used to introduce nonlinear behavior, such as limit cycles,
over and above the usual linear analysis taught in a control
curriculum. Finally, thinking about controller tuning with
stiction will force the students to think harder, broadening
their understanding of control concepts. For example, in
traditional control thinking high gain is the usual suspect for
causing oscillations. Whenever there are oscillations, control
engineers are taught to reduce the gain. For stiction-related
oscillations, reducing the gain can actually make the oscilla-
tions worse. Therefore, this experiment can be used to
make the students think about control more carefully (
to prevent these mistakes.
CONTROL VALVE AND STICTION P
A control valve consists of two main parts: a valve;
and an actuator that forces the stem to move. Addi-
tionally, it may contain a positioner that controls the
valve stem, allowing it to correspond with the control
signal. Stiction in control valves is thought to occur Sp +
due to seal degradation, lubricant depletion, inclusion
of foreign matter, activation at metal sliding surfaces
at high temperatures, and packing around the stem.
The resistance offered from the stem packing is often
cited as the main cause of stiction. Stiction happens
where static friction is substantially higher compared
with dynamic friction. There is an initial phase where
the valve fails to respond to the control signal until
the static friction is overcome. Once the static friction is
overcome, as friction reduces to the dynamic friction level,
the valve slips suddenly. It may then stick at the new position
or follow the control signal. This stick-slick behavior causes
oscillations (limit cycles) in both the process variable and
In the following sections, a simulation and an experimental
case study are described to illustrate the effect of stiction
phenomenon in control loops.
SIMULATION CASE STUDY
Figure 2a shows a basic regulatory control loop, and Figure
2b shows the loop structure in the presence of stiction. The
identification of a plant's linear dynamics (either in open or
closed loop) includes valve dynamics (see Gp in figure 2a).
Valve dynamics are observed only after the start of stem
movement; stiction phenomenon, if present, precedes valve
dynamics. This is represented in Figure 2b.
Several models for stiction have been proposed in the
literature. 3132] Here, we consider a simple stiction model
parameterized by one parameter, "d," given by Eq. (2).
X x(t - 1) if Ju(t) - x(t - 1) < d,
X u(t) otherwise
Eq. (2) is characterized by a single parameter, "d," termed
as stiction band. Here x(t) and x(t - 1) are present and past
Figure 2. (a) Regular process control loop. (b) Process
control loop in presence of stiction.
Vol. 41, No.2, Spring 2007
stem movements and u(t) is the present controller output. The
stem moves from one position to the other once it overcomes
the dead band "d."
In the process industries, stiction measurement is done
when the loop is in manual mode. A slow, increasing ramp-
type control signal is given as valve input. The valve input
is increased until a noticeable change in the process variable is
observed. Stiction is reported as a percent of the valve travel or
span of the control signal. The stiction model given by Eq. (2)
coincides with the procedure used for measuring stiction
and is reported as the span of the control signal. Readers are
referred to Srinivasan, et al., 23 24] for a detailed discussion on
the applicability of this simple model for modeling stiction.
A continuous system, 1 , with a discrete PI
controller (K = 0.4, T = 1) was considered. The sam-
pling time (T) was fixed to 0.1. A simulation setup
in MATLAB was designed using the simple stiction
model [Eq. (2)]. This is shown in Figure 3. Setpont
Simulation was carried out for 100 seconds, with a
sampling time of 0.1 seconds. To induce limit-cycle
behavior, however, a small deviation from the steady-
state is necessary and was given. Figure 4 shows
the control loop data when subjected to a stiction
measure of d = 0.1. While we used an experimental
setup at Clarkson University for teaching stiction, the
simulation case study presented here could be used
instead of the experimental setup, if such a setup is
EXPERIMENTAL CASE STUDY:
Figure 5 depicts the liquid-level system at Clarkson
University. It is a water-flow system with a linear nee-
dle plug valve assembly. The actuator is configured 02
to "Air to Close" with "Fail to Open" settings. The 0 15
installed control valve does not have a positioner. The
level measurement (PV) is acquired in the computer
using a Data Acquisition card (PMD-1208LS). The os05
level control was accomplished with a PI controller 0
implemented in Matlab (Simulink) environment with
a sampling time of 0.5 seconds. Simple step tests in -0 05
control signal indicated a first-order linear process
with a gain (K = -4.5) and approximate time constant
(T = 80 seconds). The parameters of PI controller
were K = 0.88, T = 0.0138 sec , obtained using the 0E 5
IMC rule for filter-parameter = 4. The control valve 0
exhibited negligible static friction (< 0.1%). There
are several ways the stiction phenomenon can be 0 0
demonstrated on this setup. The static friction in the 0
control valve can be increased by tightening the pack-
ing around the stem. This will introduce stiction in the
control valve. We have done this, and made extensive
use of such data in our research work. While this is possible,
if the experiment is going to be used for an undergraduate lab,
tightening the stem might not be necessary. Continuous tight-
ening and loosening of the stem might damage the packing,
leading to valve replacement. It is sufficient to use a stiction
model in the Simulink environment. A Simulink implementa-
tion of the whole system is shown in Figure 6.
The stiction experiment was assigned as a project to a group
of students who took the process control course taught by
Professor Sandra Harris. A detailed instruction sheet on how
to operate the liquid-level system and Matlab interface (in-
Figure 4. Loop data when stiction band d = 0.1.
Chemical Engineering Education
Figure 3. Simulation case study implemented in MATLAB.
eluding how to incorporate stiction) was provided to students.
They were then asked to work the experiment and answer the
(1) Bring the process to a steady state (say Level Set point
= 30%) using closed-loop computer control with a
stiction band of d = 0%.
(2) Set the stiction band to d = 4%.
(3) Give a step change of 10 %, i.e., the Level Set point is
changed to 40.
(4) Observe the measurements for 15 minutes. Explain
why the data oscillates.
(5) ( ,,>.,.' the controller ..ii;. (Note this time instant).
(a) Increase the Integral gain (equivalent to
reducing the value T) and observe the data for
Figure 5. Liquid-Level System.
10 minutes. Comment on oscillation period and
(b) Reset the integral gain to its original value.
Increase the controller gain, K, to a new value.
Observe the data for 10 minutes. Comment on
oscillation period and amplitude.
(6) Set the stiction band to d = 0. Comment on oscillation
period and amplitude. Do the oscillations stop?
(7) Observe the outputfor 10 minutes.
(8) Set the stiction band to d = 4%. Observe the data for 10
minutes. Does the loop start . Ia.l,,,, again? What is
the oscillation period and amplitude? Comment.
In this section, we present a student report on the experi-
ment. It can be seen from the report that the group had to think
about the effect of various controller tuning
concepts and also understand oscillations
generated through nonlinearities.
The liquid-level control experimental
apparatus in Clarkson University's Un-
dergraduate Laboratory was used for this
experiment. A GUI developed using Simu-
* link was used to adjust both the controller
Output qo parameters and the simulated stiction. All
relevant outputs and controller parameters
50 5, IN pI
Initial Level Initial Val PV data
P Position + Position
10 + I
Set piont Measuerernent -
PID Controller1 Add
To Valve I I
Liquid Level Measuerement
Figure 6. Liquid-Level System controlled from Simulink.
Vol. 41, No.2, Spring 2007 127
were recorded and plotted. First, the liquid-level set point
was set to 40% and the process was run until steady state was
achieved. Next, the level set point was decreased by a step
of 10% and the stiction band was set to 4%. The oscillatory
effect of the stiction band on the step response was observed.
After about 15 minutes, the integral time constant of the
controller was increased, and the oscillations in liquid-level
were observed for about 7 minutes. The integral action was
then reset to its original value and the controller gain was in-
creased. After another observation period of about 7 minutes,
the controller gain was reset to its original value. The stiction
band was then set to 0, thereby eliminating the simulated stic-
tion. The process was observed for about 5 minutes before
the stiction band was set back to 4%. The effect of stiction
reintroduction was observed for about 5 minutes.
The experimental results were classified into six distinct
regions. These are shown in Figures 7 and 8. Figure 7 shows
the set point and the corresponding percent level vs. time.
Figure 8 shows the controller output and the valve position
percentages vs. time.
Region 1: Region 1 shows the start of the process. In this
region, the percent level was relatively steady around the set
point of 40%, as shown in Figure 7. The controller acted ide-
ally and adequately compensates for process disturbances. As
shown in Figure 8, the controller output and
valve position signals were nearly identical,
Measurement () as the valve responded almost perfectly to
0 int N%) the output of the controller.
Region 2: The transition between Region
1 and Region 2 occurred as simulated valve
stiction was simultaneously introduced with
a set point change to 30%. The effects of
stiction are easily seen in Figure 8, where
the valve position signal began to change
in steps rather than closely following the
controller output signal. The set point change
also demonstrated the activity of the propor-
tional element of the controller, such that as
the step was made the output was proportion-
ally adjusted. Shortly after the step change,
the oscillatory effects of stiction, in conjunc-
3 tion with the effect of the integral action of
3 the controller, were clearly observed. The
further the level deviated from the set point,
the more the controller output was adjusted
by the integral action, until finally the 4%
. . "" stiction band of the valve was overcome and
the valve responded. Because of the stiction
band, the valve response overcompensated,
causing the level to increase or decrease.
Again, the controller attempted to compen-
sate for the error, resulting in the oscillatory
behavior seen in both Figures 7 and 8. The
large disturbance at around 750 seconds
was a demonstration of a method used to
combat stiction. This anomaly resulted in a
"resetting" action for the controller, leading
to a period of stable operation even though
stiction was still a factor.
Region 3: Region 3 was characterized
by a decreased period, and thus increased
6 frequency, of oscillation, as seen in Figures
3o 4a 7 and 8. This was a response to an increase
in the integral action of the controller, caus-
Chemical Engineering Education
Figure 7. Set point and percent level.
Figure 8. Controller output percent and valve positi
ing the controller to respond more aggressively. With a more
aggressive integral mode, the controller attempted to make
corrections more quickly, resulting in the increased frequency
Region 4: In Region 4, the integral time constant was reset
to its previous value and the controller gain was increased.
This caused two distinct differences in the controller output,
the valve position, and thus liquid level. By adjusting the con-
troller's integral action back to the original setting, the period
of oscillation returned to a value similar to that experienced in
Region 2. By increasing the controller gain, the amplitude of
the oscillations increased. This was demonstrated toward the
end of this region as the liquid level began to quickly change
through a wide range of values.
Region 5: Region 5 was characterized by the deactivation
of the stiction band simulation and the controller gain being
reset to its original value. Once these actions were taken, the
valve once again responded ideally to the controller output,
and the level returned to a more stable value. This behavior
was similar to the behavior in Region 1.
Region 6: In Region 6, the stiction band simulation was
again activated with a value of 4%, without changing any
of the controller parameters. The oscillatory behavior of the
process was observed with a frequency and amplitude similar
to that experienced in Region 2.
CONCLUSIONS AND FUTURE WORK
When most control technologies are initially implemented
in an industrial environment, they are well tuned and operate
optimally. For a variety of reasons, however, it is usual for
controller performance to degrade over time. Finding out
which controllers are performing poorly, identifying the cause
of poor performance, and suggesting corrective actions are
the overall problems of CPA. Since it is the responsibility of
control engineers to solve such problems on an almost daily
basis, it is worthwhile to introduce CPA at an undergraduate
level. In this article, a possible approach introducing under-
graduate students to the concept of stiction in control valves
is discussed. A student report on the lab experiment was also
presented. There are several avenues for future work. Since
the process control class does not have a lab component at
Clarkson, it was difficult to teach stiction as a part of the cur-
riculum. Only individual student groups could do this experi-
ment as a project. To circumvent that problem, we are planning
to make this experiment Web accessible. 33] This would make
it possible for stiction to be taught in the classroom. Further,
this would also make it possible for other universities to use
the experiment in their classrooms.
Stiction phenomenon also introduces limit cycles, and it is
possible to predict the limit cycle characteristics through, for
example, describing function analysis. It was felt, however,
that these concepts would be inappropriate for the undergradu-
ate control class at Clarkson. As part of future work, we will
try to introduce this experiment with limit cycle analysis as
part of a graduate course in process control. Further, stiction
is just one aspect of CPA. As discussed in the introduction,
there is a need to teach CPA at an undergraduate level. With
this goal, we are working on a three-tank setup that can be
used as a lab experiment, to discuss the whole gamut of
The authors are grateful to Mark Cooke, Sr., for interfac-
ing the Liquid-Level System with the computer. The authors
also thank the Department of Chemical and Biomolecular
Engineering for providing financial support for this work.
We would also like to acknowledge the student group of Jon
Mosenteen, Brian Ricks, Nathan Victor, and Kelly Weitz for
participating in the stiction experiment. The authors thank
the National Science Foundation for partial support through
the grant CTS-0553992.
1. Bialkowski, WL., "Dreams Versus Reality: A View From Both Sides
of the Gap," Pulp and Paper Can., 94(11), 19 (1993)
2. Ender, D.B., "Process Control Performance: Not as Good as You
Think," Control Eng, 9, 180 (1993)
3. Desborough, L.D., and R.M. Miller, Increasing Customer Value of
Industrial Control Performance \ I............ -i -..,...... Experience.
CPC-VI, Arizona (2001)
4. Desborough, L.D., Control Loop Economics, Honeywell Proprietary
5. Gatzke, E.P, E.S. Meadows, C. Wang, and E J. Doyle, "Model-Based
Control of a Four-Tank System," Computers and ( ...i.. i 24, 1503
6. Joseph, B., C. Ying, and D. Srinivasgupta, "A Laboratory to Supple-
ment Courses in Process Control," Chem. Eng. Ed., 36(1), 20 (2002)
7. Cooper, D.J., D. Dougherty, and R. Rice, "Building Multivariable
Process Control Intuition Using Control Station," Chem. Eng. Ed.,
37(2), 100 (2003)
8. Rusli, E., S. Ang, and R.D. Braatz, "A Quadruple-Tank Process Control
Experiment," Chem. Eng.Ed., 38(3), 174 (2004)
9. Young, B.R., J.H. van der Lee, and W.Y. Svrcek, "A Nonlinear, Multi-
input, Multi-output Process Control Laboratory Experiment," Chem.
Eng. Ed., 40(1), 54 (2006)
10. Kozub, D.J., Controller Performance Monitoring and Diagnosis:
Experiences and ( ,.,n,. . .. CPC V, Lake Tahoe, CA (1996)
11. Ranganathan, S., "Control Loop Performance Monitoring: Model-
ing, Diagnosing, and Compensating Stiction Phenomenon in Process
Control Valves," Ph.D Thesis, Clarkson University (2005)
12. Hagglund, T., "A Control-Loop Performance Monitor," Control Eng.
Prac., 5(11), 1543 (1995)
13. Thornhill, N.E, and T. Hagglund, "Detection and Diagnosis of Oscil-
lation in Control Loops," Control Engg.Prac., 5, 1343 (1997)
14. Horch, A., "A Simple Method for the Detection of Stiction in Control
Valves," Control Engg. Prac., 7, 1221 (1999)
15. Forsman K., "On Detection and Classification of Valve Stiction, "TAPPI
Process Control, Electrical and Information Conference, Williamsburg
16. Rengaswamy R., T. Hagglund, and V. Subramanian, "A Qualitative
Shape Analysis Formalism for Monitoring Control Loop Performance,"
Eng. App. ofArtificial Int., 14, 23 (2001)
17. Thornhill, N.E, B. Huang, and H. Zhang, "Detection of Multiple Oscil-
Vol. 41, No.2, Spring 2007
lations in Control Loops," J. of Process Control, 13(1), 91 (2003)
18. Stenman, A., E Gustafsson, and K. Forsman, "A Segmentation-Based
Approach for Detection of Stiction in Control Valves," Int. J. Adap.
Control and Signal Proc., 17, 625 (2003)
19. Xia, C., and J. Howell, "Loop Status Monitoring and Fault Localiza-
tion," J. of Process Control, 13(7), 679 (2003)
20. Choudhury, M.A., S.L. Shah, and N.E Thornhill, "Diagnosis of Poor
Control Loop Performance Using Higher Order Statistics," Automatica,
40(10), 1719 (2004)
21. Choudhury, M.A., S.L. Shah, and N.E Thornhill, Detection and Quan-
tification of Control Valve Stiction, DYCOPS, Boston, (2004)
22. Yamashita, Y., "An Automatic Method for Detection of Valve Stiction
in Process Control Loops," Control Eng. Prac., in production
23. Srinivasan, R., R. Rengaswamy and R.M. Miller, "Control Loop Perfor-
manceAssessment 1: A Qualitative Pattern MatchingApproach for Stiction
Diagnosis," Ind. and Eng. ( ...... , . Research, 44, 6708 (2005)
24. Srinivasan, R., R. Rengaswamy, S. Narasimhan, and R.M. Miller,
"Control Loop Performance Assessment 2: Hammerstein Model Ap-
proach for Stiction Diagnosis," Ind. andEng. ( i....... Research, 44,
25. Astrom, K.J., Introduction to Stochastic Control Theory, Academic
Press, New York (1970)
26. Harris, T.J., "Assessment of Control Loop Performance," Canadian
Journal of Chem. Eng., 67, 856 (1989)
27. Desborough, L.D., and T.J. Harris, "PerformanceAssessment Measures
for Univariate Feedback Control," Canadian Journal of Chem. Eng.,
70, 1186 (1992)
28. Bezergianni, S., and C. Georgakis, "Evaluation of Controller Perfor-
mance-Use of Models Derived by Subspace Identification," Int. J.
Adap. Control and Signal Processing, 17, 527 (2003)
29. Hagglund, T., "A Friction Compensator for Pneumatic Control Valves,"
J. of Process Control, 12, 897 (2002)
30. Srinivasan, R., and R. Rengaswamy, "Stiction Compensation in Pro-
cess Control Loops: A Frame-work for Integrating Stiction Measure
and Compensation," Industrial and Eng. Chem. Research, 44, 9164
31. Olsson, H., "Control Systems with Friction," Ph.D. Dissertation, Lund
Institute of Technology, Sweden (1996)
32. Armstrong-Helouvry, B., P Dupont, and C.C. De Wit, "A Survey of
Models, Analysis Tools, and Compensation Methods for the Control
of Machines with Friction," Automatica, 30(7), 1083 (1994)
33. Selmer, A., M. Goodson, M. Kraft, S. Sen, andV.E McNeill, "Perform-
ing Process Control ExperimentsAcross the Atlantic," ( I.. .. i i.,
39(3), 232 (2005) 1
Chemical Engineering Education
TEACHING PROCESS ENGINEERING
USING AN ICE CREAM MAKER
GONCL KALETUNC, KEVIN DUEMMEL, AND CHRISTOPHER GECIK
Ohio State University * Columbus, OH 43210
P processing of food materials provides excellent opportu-
nities for both high school and college students to learn Goni Kaletun9 is an associate professor in
the Department of Food, Agricultural, and Bi-
science and engineering concepts. Use of ice cream logical Engineering at Ohio State University.
processing as a teaching tool provides hands-on experience She received her B.S. and MS. degrees from
the Middle East Technical University, Ankara,
with the added incentive of consuming the final product. This Turkey, andher Ph.D. degree in food engineer-
experience encourages students to think about the science and ing from the University of Massachusetts. She
promotes undergraduate research by employ-
engineering behind the processing of any food product that ing students in her laboratory and engaging
they come across during their daily lives. them to write research proposals. Herresearch
interest is in thermal and theological analysis
A number of articles and books about ice cream processing of food and biological materials.
and properties of ice cream can be found in the literature. Kevin R. Duemmel is currently with the Ohio
These studies describe the unit operations in ice cream pro- Department of Transportation. He was a me
chanical-design engineer with the Ohio State
cessing, the effect of processing conditions on the microscopic University Department of Food, Agricultural,
and macroscopic structures of ice cream, and the relationships and Biological Engineering. He has 15 years
experience in engineering design and construc-
between the structural, physical, and sensory properties of tion, including electronics and instrumentation.
ice cream. Several engineering and science concepts are as- A licensed professional engineer, he has a
B.S. in electrical engineering, an MS. in civil
sociated with manufacturing ice cream and the final product, engineering, andis a Ph.D. candidate in Food,
including applications of material and energy balances, heat Agricultural, and Biological Engineering at the
and mass transfer, mixing, freezing, freezing point depres- Ohio State University.
sion, emulsion, foam formation, and viscosity. Some of
these concepts are introduced and discussed through lectures Christopher Gecik is design engineer for
the Department of Food, Agricultural, and
and problem solving in sophomore-level courses in various Biological Engineering at Ohio State Univer-
chemical, food, or biological engineering departments. In sity. He has a B.S. and M.A. from Ohio State
University along with coursework from other
our department, a 10-week sophomore course dedicated to institutions. He has been designing electrical,
discussion of material and energy balances in relation to food electronic, and mechanical systems at Ohio
and biological s s is o . Te c e ao i s State for more than 10 years, and has worked
and biological systems is offered. The course also includes as a "special effects designer" on Broadway
and at Disney World.
� Copyright ChE Division of ASEE 2007
Vol. 41, No.2, Spring 2007 13
an introduction to fluid mechanics in the last two weeks. A
laboratory experiment based on ice cream processing in a
small scale batch mode, using a one-liter capacity electric
ice cream maker, is implemented as a real-life example to
illustrate and discuss applications of material and energy
balance, as well as mixing and viscosity.
The objective of the experiment is to facilitate student
learning of engineering concepts by a team-based problem-
solving approach. This article describes in detail the specific
assignments given to students, the data collected, and analysis
of the data to further their understanding of the basic science
and engineering principles behind ice cream processing.
MATERIALS AND METHODS
Ice cream ingredients, including heavy cream (36% fat),
whole milk (3.2% fat), sugar, vanilla extract, and salt, were
purchased in a local grocery store.
An electric-powered ice cream maker (Cuisinart ICE20,
Denver, CO) was used to make the ice cream. Figure 1 is a
photograph of the ice cream maker assembled showing vari-
ous parts, including the gate- (or anchor-) type mixer and the
lid. The freezer bowl is removed in order to display the internal
parts. The ice cream maker was modified to allow measure-
ment of the temperature, rotational speed (rpm), and mixing
force (kg) delivered to the mixture during processing.
Modifications for Rotational Speed Measurement
The ice cream maker rotates the freezer bowl at 38 rpm,
slowing slightly as the ice cream mixture thickens. Figure
2a shows the bottom view of the ice cream maker. A bicycle
computer (Zone 5, Echowell Electronic Co., Ltd., Taiwan) was
used as a tachometer to measure the mixer's rotational speed.
The bicycle computer works by counting pulses created by
a permanent magnet that repeatedly rotates past an electrical
reed switch. The rotational speed is determined by multiplying
the number of pulses per unit time with the bicycle wheel's
circumference. An appropriate "wheel circumference" was
calculated by counting the number of teeth on the gears. The
"bicycle wheel's circumference," corresponding to 38 rpm,
was calculated to be 2445 mm based on the gear chosen to
carry the magnet in the ice cream maker.
An accessible gear inside the ice cream maker was chosen
to carry a small permanent magnet that rotated past the reed
switch. The enclosed area in Figure 2a is expanded for clear
viewing of the mounting location of the permanent magnet and
the reed switch inside the ice cream maker (Figure 2b). The
reed switch supplied with the bicycle computer failed quickly
in actual use. Therefore, a more robust reed switch (COTO
RI-48A, COTO Technology, Providence, RI) was installed
in the ice cream maker and was held in place with hot-melt
glue. A 0.1 pF ceramic capacitor was also placed in parallel
across the reed switch to remove high-frequency noise from
the tachometer circuit (buried under glue in Figure 2b).
Modifications for Torque Measurement
The ice cream maker used is unique in its design because
the freezer bowl containing the ice cream mixture rotates. This
design is well-suited for modification of the ice cream maker
for measurements of mixing force and temperature. A station-
ary mixer allows insertion of temperature probes into the ice
cream mixture. The mixer in the ice cream maker is kept
- Strain gauge
. cr nchrr mixer
Figure 1. Ice cream
(except freezer bowl).
Bottom view of the
ice cream maker.
Chemical Engineering Education
stationary with tabs molded inside the top of the lid, which rotating
is locked into the base unit by means of pegs on the inside the free
surface of the base of the lid. Such arrangement allows the gate ately in,
mixer to remain stationary against the force of the ice cream inside s
mixture rotating together with the freezer bowl, and provides to be -1
easy assembly or disassembly of the equipment. In order to after co
measure the mixing force on the mixer, it must be released
from the lid and allowed to rotate against a moment-arm. The
top of the lid containing the tabs was removed to allow the Stude
mixer to rotate with the bowl and transfer its rotational force the labc
to an added assembly. The assembly consists of the mixer mixing
fitted with a clear acrylic disk to which a shaft is attached at
center. The upper end of the shaft is concentrically supported
with an aluminum frame attached to the modified lid (Figure
1). A pulley with a radius of 25 mm is placed on the shaft and
a flexible constraint (a tape-measure blade) is attached to the
pulley. The flexible constraint is parallel to the radius of the
pulley, always tangential to the circumference of the pulley,
in spite of any rotation of the pulley. The length of the torque
arm therefore remains consistently 25 mm.
A small postal scale (Pelouze SP5, Sanford, Oak Brook, IL)
was disassembled to remove the load cell and the digital read-
out electronics. The scale has a measuring range of 0 to 2,200
grams. The digital readout electronics were mounted inside a
case. The load cell was mounted in line with the tape-measure
blade. The mixing force developed in the ice cream mixture, RESU
due its increasing viscosity, is transferred from the mixture
to the mixer, shaft, pulley, flexible constraint, and load cell Formu
nearly without friction. Torque is calculated by multiplying the Ma
the mixing force with the length of torque arm (25 mm). Stude
Temperature Measurement amount
Two digital thermometers (Taylor TruTemp, #3516, Oak tons.
Brook, IL) were used to measure the temperature of the ice
cream mixture during processing. Small holes were drilled exr
into the acrylic disk for the insertion of thermometers and 0.86%
one larger hole was added for pouring the ice cream mixture
into the assembled equipment after the freezer bowl started The sl
Composition of Ingredients and Ice Cream Mixture
Products Fat Protein Water Carbo- Salt
(%) (%) (%) hydrate (%)
Heavy 36.0 - 64.0 -
Whole 3.2 3.2 88.8 4.8
Sugar - - - 100.0
Salt - 100.0
Vanilla - - -
Ice cream 10.5 1.9 67.1 19.5 0.2
Vol. 41, No.2, Spring 2007
. One thermometer was placed close to the center of
zer bowl. A second thermometer was placed immedi-
side the inner wall of the freezer bowl (Figure 1). The
surface temperature of the freezer bowl was measured
5 �C before adding the ice cream mixture and -9 �C
mpletion of ice cream processing.
lure for Ice Cream Experiment
:nts followed the procedure outlined below during
)ratory to collect temperature, rotational speed, and
Mix the ingredients in a bowl.
Weigh a fixed volume of the ice cream mixture.
Place thermometers in the lid and complete assembly
of ice cream maker.
Start the ice cream maker.
Pour the ice cream mixture into the freezer bowl
equilibratedd at -20. �C in the freezer).
Collect temperature (center and side), mixing speed,
and mixing force data.
After the run is completed, weigh a fixed volume of the
LTS AND DISCUSSIONS
nation of Ice Cream Using
triall Balance Concept
nts were given a pre-lab assignment to calculate the
s of individual ingredients needed to prepare 0.7 kg of
ice cream for a given set of compositional specifica-
he specified ingredient composition and fat content
nent for ice cream are outlined in Table 1. Vanilla
md salt are considered pure components that comprise
md 0.2% of the ice cream mixture, respectively.
students were asked to apply the material balance con-
cept to set up simultaneous
linear equations that can be
solved to obtain the amount
Vanilla Amount of each ingredient (Figure
(%) of each 3, next page). The students
ingredient solved the five simultane-
S(g) ous linear equations, based
166 on the matrix method using
MATLAB software (7.01,
409 The Mathworks Inc.), to
determine the amounts of
- 118 heavy cream, milk, sugar,
- 1 vanilla, and salt needed
100.0 6 to prepare the ice cream
0.86 700 mixture (Table 1).
Application of the Material Balance Concept
to Determine Ice Cream Quality
Ice cream quality is defined by overrun. Overrun is calcu-
lated as the percent increase in volume of ice cream based on
the volume of the ice cream mixture used (Eq. 1).
SVolume of ice cream - Volume of mix 100
Overrun = olm * 100 (1)
Volume of mix
Volume can be written as mass divided by the density to
obtain the following equation.
Overrun (m, / po)- (mmix / Pmix) *100 (2)
Overrun = 100 (2)
where subscript ic denotes the ice cream and mix denotes the
ice cream mixture.
The following equation is written based on conservation
mmx = m, + mair (3)
Although the percent volume increase can be as high as 120%,
the mass contributed by the air is negligible because the den-
sity of air (1.239 kg/m3 under standard conditions) is much
smaller than the density of the ice cream mixture. Therefore
Eq. (3) can be approximated as:
mmix mlc (4)
The overrun is calculated by using the densities of the ice
cream mixture and the ice cream.
Overrun= (1/ P)- (1/Pmx) *100 Pmix -P * 100 (5)
1 / Pmx) Po
The prepared ice cream mixture had a density of 1079 kg/m3,
typical for ice cream mixes. After 30 minutes of processing
time in the ice cream maker, the ice cream had a density of
692 kg/m3, which gives an overrun value of approximately
56%. The volume of water expands upon freezing. Using the
concept in Eq. (5), the percent volume expansion for water
per unit mass upon freezing was calculated to be 9.3 using
the densities of water and ice at 0 'C. 1 The expected volume
expansion of the ice cream mixture containing 67.1% water
by weight was calculated to be 6.3% at 0 �C. The volume
expansion of the ice cream mixture due to freezing was also
determined experimentally to be 5% by freezing a known
volume of the ice cream mixture in a freezer at -20 �C. The
difference between the experimental value and the calculated
value of volume expansion can be attributed to several fac-
tors. The density of ice increases slightly as the temperature
decreases, making the expected volume expansion 6% at -20 C.
Furthermore, the presence of solutes such as sucrose, lactose,
and salt in the ice cream mix decreases the available water
for freezing, leading to decreased volume expansion observed
experimentally. The percent volume increase of the ice cream
mixture during processing due to air incorporation alone was
estimated to be 51%. Volume expansion due to air incorpora-
tion is the major fraction of the overrun and depends on the
rotational speed of the mixer and the developing viscosity of
the freezing ice cream mixture, which is influenced by the
formulation and cooling rate.
Calculation of Heat Removed
Students were asked to plot temperature data collected as
a function of time during ice cream processing (Figure 4).
Several physical phenomena can be illustrated using Figure
4. In the figure, four regions can be identified. Region I, the
cooling of the ice cream mixture, is followed by the freezing
of the ice cream mixture (II), cooling of the frozen mixture
(III), and the constant temperature region (IV).
The freezing of the ice cream mixture started at -1.7 �C,
demonstrating the freezing point depression due to the pres-
ence of dissolved sugars (sucrose and lactose) and salt in the
aqueous portion of the ice cream mixture. The major contri-
bution to the total amount of solutes in the ice cream mixture
was from sucrose (82%), followed by lactose (14%), and salt
(4%). The freezing point depression calculated based on the
molality of sucrose was 1.4 �C, revealing the contributions of
lactose and salt in
the freezing point
depression to be
to freezing tem-
perature, the heat
removed by the
freezer bowl is not
expected to reduce
the temperature of
ice cream mixture,
but to remove the
heat of freezing
(0.36) HC + (0.032) WM = (0.104) (700)
(0.048) WM + (1.0) S = (0.197) (700)
(0.64) HC + (0.888) WM = (0.671) (700)
(1.0) Sa = (0.002) (700)
(1) V = (0.0086) (700)
Figure 3. Material balance equations.
I II III IV
o 12 1
0 0 Z22*
i- 0 .5 5
E 13 4 0.6 U)
0 5 10 15 20 25 30
Figure 4. Temperature (*), Rotational speed, rps, (A), and
Torque, Nm, (*) of ice cream mixture during processing.
Torque values are multiplied by 30 to fit into scale.
Chemical Engineering Education
with the phase change of water from liquid to solid (ice). Dur-
ing freezing of a pure substance such as water, the temperature
is expected to remain constant until freezing is completed.
Ice cream, however, is an interesting system to teach students
because the freezing point depression increases as the ice
cream system becomes concentrated due to the freezing of
water. Therefore, instead of a constant-temperature freezing
process, we observe an approximately 1.1 �C decrease in
the freezing point of the ice cream during the phase change
(Region II, Figure 4).
When the phase change is completed, the temperature of the
ice cream starts to decrease (region III) to approximately -6 �C
and remains constant (Region IV). The constant temperature
region occurs because the heat removed from the ice cream
becomes equal to the heat generated by friction, due to mix-
ing, and the driving force for heat transfer decreases as the
freezer bowl warms up to -9 �C from an initial temperature of
-15 �C. The enthalpy calculations for each region and the heat
removed during the process are summarized in Table 2.
Time : 27.5 min
Strain gauge reading : 0.18 kg
Mixing Force : (0.18 kg) (9.80 m/s2) =17.6 N
Torque: (17.6 N) (0.025 m) = 0.44 N . m
Rotational speed (N): 37 rpm = 0.62 rps = 3.87 radians/s
Power (P): (0.44 N. m) (3.87 radians/s) = 1.7 W
Density of ice cream (p): 692 kg/m3
Diameter of mixer (D): 0.126 m
Power number (N )
p N3 D5 (692) (0.62)3 (0.126)5
Mixing Power and Viscosity Analysis
The students were asked to collect mixing speed and rota-
tional force data during ice cream processing. The data was
used to calculate the mixing power delivered to the ice cream
mixture during processing. The power input was calculated
using the torque and the rotational speed in radians per sec-
ond. Figure 4 shows the change in mixing torque and speed
as a function of time during ice cream processing together
with temperature data to illustrate the relationship between
the four heat removal regions and the fluid behavior during
In Region I, during cooling of the ice cream mixture, a
slight increase in the mixing torque dissipated to the ice
cream mixture was observed and the mixing speed remained
constant. In Region II, the phase change progresses causing an
increase in the mixing power while the mixing speed remains
constant. Upon completion of the phase change, we observe
a substantial continuous increase in mixing torque with a
concomitant decrease in mixing speed, due to the increasing
viscosity of the system (Region III). In Re-
gion IV, despite the constant temperature,
the mixing torque and speed continue to
increase and decrease, respectively. The
higher rate of mixing torque increase in
Region IV than in Region III might be due
to the air incorporation in the ice cream
mixture that occurs in Region III.
The data are further analyzed to es-
timate the viscosity of the ice cream.
The power consumption of a mixer is
described as the empirical relation-
ship between the dimensionless power
number, N , and the Reynolds (Re)
number. The relationship depends on the
impeller geometry and the fluid regime
(Newtonian vs. non-Newtonian). In this
study, the characteristic power curve for
an anchor impeller was used to determine
the Re number from the calculated
N .[7 The Re number was used
to estimate the viscosity of the
ice cream produced. The steps
of the calculation are shown
in Figure 5. The characteristic
power curve used was developed
for Newtonian fluids. Ice cream
is not expected to behave like a
Newtonian fluid. Therefore, the
estimated viscosity is referred to as
apparent viscosity, and is equal to
9.7 Pa's at the maximum shear rate
(Ymx) of 90 s 1, based on the tip
speed of the impeller calculated
Re number corresponding to Np = 325 is determined from reference  to be 0.7
Re =- 2 -0.7
Viscosity (u) = 9.7 Pa. s
Figure 5. Calculation steps to estimate the viscosity of the ice cream produced.
Temperature, Enthalpy, and Rate of Heat Removal for
Each Region During Ice Cream Processing
Region I Region II Region Region
Initial temperature, �C 11.0 -1.7 -2.7 -6
Final temperature, �C -1.7 -2.7 -6 -6
Time interval, min 0-4 4-11 11-21 21-27.5
Specific heat capacity, kJ/(kg�C) 3.3 1.88 1.88
Enthalpy of freezing for water, -334
Enthalpy, kJ 29.3 157 4.3 0.0
Rate of heat removal, kJ/min 7.3 22.4 0.43 0.0
Vol. 41, No.2, Spring 2007
by using the following equation81 :
where d is the anchor diameter, D is the inside diameter of the
freezer bowl, and N is the rotational speed (rps).
Implementation of the Ice Cream Experiment
in Various Settings
The ice cream experiment is used as a teaching tool with
small groups in a sophomore college class. The experiment
is also used in a large group setting as part lecture, part
hands-on program for high school teachers and students. The
program, tied "Science and Engineering of Ice Cream," is
designed for high school science teachers to take away
an education tool for deployment in their classrooms. In
the "Food Engineering" session of the "Get Real About
Science" program, high school students are exposed to
several engineering and science concepts in an enjoyable
and engaging program.
In small group settings, the tool was used as a laboratory
experiment with groups of three to four students. Students
were asked to calculate the amount of each ingredient by
solving material balance equations prior to performing the
experiment. The combination of ingredients was changed
(skim milk, 1% milk, or 2% milk vs. whole milk, light cream
vs. heavy cream) so that each group solved a different set
of material balance equations and produced ice cream with
The student experience was expanded to use the ice cream
experiment as a team-based term project in small groups. After
a class discussion on the potential operating variables that
affect the processing and product characteristics, each group
chose one variable to investigate. The operating variables
to investigate included use of artificial sweetener (Splenda)
to partially or completely replace sugar, or changes in the
rotational speed, the freezer bowl temperature, and the fat
content of the ice cream. Students completed their experi-
ments, analyzed the data, prepared reports, and presented their
results at the end of the class.
For a large group setting, a temperature vs. time graph was
distributed to each student. After an initial introduction, the
ice cream experiment was started. One student was given
the responsibility to keep track of the time so that at minute
intervals students could read temperature data. Each student
marked the temperature data points on their graph paper. By
the completion of the ice cream experiment, each student
had the temperature profile of ice cream processing on their
graph paper and was ready to discuss the various phenomena
that occurred during the process. During the minute waiting
periods, the history of ice cream processing and ice cream
related facts were discussed with students. This approach
allows one to run a hands-on experiment with a single setup
in a large group setting while keeping everybody's attention
on the program.
The ice cream laboratory experiment is designed not only to
illustrate to students the basic engineering principles behind
ice cream processing, but to provide a tool to apply analytical
and critical thinking to other engineering applications. Any
ice cream maker using a stationary mixer and rotating bowl
can be modified similarly to measure the mixing force and
temperature. Students learn how to analyze a given system's
performance and the interrelationships among raw material
formulation, operating variables, process parameters, and
product properties. Such analyses are important for optimiza-
tion of processes. Students also understand that knowledge
of such interrelationships can be used, for example, to adjust
rationally operating variables to compensate for any change
in raw material properties to achieve the same process pa-
rameters and final product properties. The advantages of
using a modified ice cream maker are: a simple batch system
that can be easily made at low cost, an experiment that can
be completed in one hour, and an experiment that is ideal
to introduce and illustrate multiple engineering and science
concepts. The setup also provides a novel method to estimate
the apparent viscosity of ice cream in the ice cream maker
under the processing conditions.
This work was supported with funds provided by the Price
Chair Teaching Improvements Grants Program in the College
of Food, Agricultural and Environmental Sciences at Ohio
1. Clarke, C., , Physics Education, 38(3), 248 (2003)
2. Clarke, C., The Science of Ice Cream, The Royal Society of Chemistry,
Cambridge, UK (2004)
3. Goff, H.D., E. Verespej, and A.K. Smith, "A Study of Fat and Air
Structures in Ice Cream," International Dairy Journal, 9, 817 (1999)
4. Hartel, R.W, "Ice Crystallization During the Manufacture of Ice
Cream," Trends in Food Science and Technol., 7, 315 (1996)
5. Marshall, R.T., H.D. Goff, and R.W. Hartel, Ice Cream, KluwerAca-
demic Plenum Publishers, New York (2003)
6. Tinoco, I., K. Sauer, and J.C. Wang, Physical ( ...... " i I . Hall,
Englewood Cliffs, NJ (1995)
7. Foucault, S., G. Acanio, and P Tanguy, "Power Characteristics in
Coaxial Mixing: Newtonian and Non-Newtonian Fluids," Ind. Eng.
Chem. Res., 44, 5036 (2005)
8. Steffe, J.E, Rheological Methods in Food Process Engineering, Free-
man Press, East Lansing, MI (1996) 1
Chemical Engineering Education
IM!1- class and home problems
TEACHING TRANSPORT PHENOMENA
AROUND A CUP OF COFFEE
JEAN STEPHANE CONDORET
Ecole Nationale Supdrieure d'Inginieurs en Arts lii""i".
We are all aware that teaching scientific matter is
much more accepted by students when it can be
related to situations they can experience in their
everyday life. A good example is the cooling of a cup of cof-
fee, whose scientific analysis is much more instructive than
we could have thought at first sight. Indeed, we will see that
all heat transfer mechanisms (conduction, convection, and
radiation), as well as those of mass transfer, (because of the
evaporation of the coffee) are involved. This problem was
often addressed as "leisure in science" or "first approach
of science," and a quick search on the Web shows that this
problem has been proposed at all levels of education, from
beginning to university. The approach presented here is aimed
at being rigorous, but because we do not intend to use very
powerful numerical modeling, simplifications will be made.
An important quality for an engineer is to make the "right"
simplification, i.e., which results only in slight inaccuracies,
while respecting the correct hierarchy for the parameters. In
the case chosen here no chemical reaction is present, but the
coupling of heat and mass transfer in a nonstationary process
is a common situation in chemical engineering. It can be
encountered, for instance, in small industrial units when a
Set Technologiques * Toulouse 31078
tank, after a batch transformation, is let to cool freely before
discharge. Another very important characteristic of the study
is that experiments to assess the modeling are easy to perform
with very simple tools, such as a thermometer, a stopwatch,
and a balance (to estimate the loss by evaporation). Such ex-
periments could even be done in a kitchen, in full accordance
with the "everyday life" aspect of the situation. The method
to approach the problem, and the reflection about transport
phenomena that it induces, make it a good basis for discus-
sion between students and teachers. To avoid a lengthy paper,
all equations given here are not discussed deeply and, for a
student, may deserve an additional look into textbooks or,
better, a discussion with teachers.
� Copyright ChE Division ofASEE 2007
Vol. 41, No. 2, Spring 2007
The object of this column is to enhance our readers' collections of interesting and novel prob-
lems in chemical engineering. Problems of the type that can be used to motivate the student by
presenting a particular principle in class, or in a new light, or that can be assigned as a novel home
problem, are requested, as well as those that are more traditional in nature and that elucidate dif-
ficult concepts. Manuscripts should not exceed 14 double-spaced pages and should be accompanied
by the originals of any figures or photographs. Please submit them to Professor James O. Wilkes
(e-mail: email@example.com), Chemical Engineering Department, University of Michigan, Ann
Arbor, MI 48109-2136.
Jean St6phane Condoret is a professor
of chemical engineering at the Institut Na-
tional Polytechnique of Toulouse (France).
He graduated in 1977 from the Institut de
Genie Chimique in Toulouse. His Ph.D. thesis
dealt with heat and mass transfer in packed
beds. Since 1987, he has been involved in
supercritical technology for chemistry and
PRESENTATION OF THE PROBLEM AND
The problem: we put a cup of hot coffee on a table. Its
initial temperature, 00, is around 80 �C, and ambient air is
at temperature 0a, with, for instance, a relative humidity of
50% (that means half-saturated). What is the temperature of
the coffee after 10 min. for instance? Or, more widely, when
will I be able to drink it safely and what parameters influence
this duration? A scheme of the situation is given in Figure
1 and the different fluxes will be discussed in paragraph 2.
To solve the problem we have to make a list of simplifying
1. Temperature 0 is homogeneous .-.. h, 1, the liquid in
the cup. There is no temperature gradient in the liquid, and
the inner wall temperature of the cup is equal to that of the
liquid, because internal free convection is sufficiently high.
These are very important hypotheses and we will devote a
specific paragraph to assess them.
2. Even if our system is time dependent, we will use
steady-state equations to model the heat and mass transfer
fluxes. This "pseudo steady-state approximation" is very
often proposed and is fully justified here, because establish-
ment of transfers is more rapid than evolution of tempera-
ture of the liquid. It is always difficult to demonstrate this
statement, and intuition is often the only indicator. Such
ambiguity is rarely addressed, but it has been discussed by
Cussler in his book about mass transfer.'M
3. There is no heat loss .-i . , 1i the bottom of the cup,
because the table blocks the heat flux. Nevertheless, we may
foresee that; ,,ii-o,, the cup on a massive metallic surface
will speed up the cooling. In this case, the bottom heat flux
would not obey the steady-state law, (see, in textbooks,
the chapter devoted to conductive transfer in semi-infinite
medium). This will not be considered here.
4. At the vertical cylindrical wall of the cup and at the
surface of the liquid, heat loss occurs by free convection and
No heat loss at the bottom
Figure 1. Schematics of the fluxes.
radiation. Moreover, at the liquid surface, evaporation of the
liquid simultaneously takes place. This evaporation induces
an extra heat loss, corresponding to the heat needed for
vaporization of water, that is provided from a decrease in the
internal energy of the liquid and the cup. Forced convection
by blowing air is not considered here, ohii,. , i, it could be
very easily implemented .1- -.. i ,,i,,. i. , q . i. ,,iq.,,tii . -. of
the coefficients of convection. It is important to mention here
that, at temperatures below 200 �C, chemical engineering
calculations usually neglect radiation fluxes because they
are * -..q ., 11,, with forced convection fluxes, which are much
larger. When dealing only with free convection, this omission
would lead to significant errors, even at low temperature.
5. The coffee cup is simulated by a cylinder, external
height H , internal diameter D, with constant wall thick-
ness e. and with a thermal conductivity i . The area of the
external vertical wall surface is A- and that of the horizontal
liquid surface is A. Also, this !i ... it, I upon the geometry
of the cup is not very restrictive and can be adapted for other
cases. Liquid is supposed to fill the cup almost entirely.
6. Coffee is similar to water, and properties are evalu-
ated at 60 �C.
Description of the Equations for Modeling
The heat loss through the wall and at the liquid surface
results in a temperature decrease that may be described by
the instantaneous heat balance equation, where accumulation
of internal energy in the water and the cup (considering ho-
mogeneous temperature) equals the sum of all instantaneous
heat losses. Because water evaporates, we also need an in-
stantaneous mass balance:
(mCpo + MCp ,) dt
- heat losses
- = -evaporative flux (2)
where m and Cpe refer, respectively, to the mass and specific
heat of the cup, and M and Cpwt, to mass and specific heat of
the water. We can now describe the different heat losses and
express them using steady-state equations of heat and mass
transfer, as stated in hypothesis 2.
Heat Loss at the Vertical Wall of the Cup, Qw
This is a transfer, in series, by conduction through the wall,
then, in parallel, free convection and radiation to ambient
air. As stated in hypothesis 1, internal convection at the in-
ner wall is not considered. This global transfer is accounted
for by a global coefficient U , referred to the external area,
area, given by : Uw
hn + hR De+ D
Qw = UweA(H - Oa)
Chemical Engineering Education
We need to evaluate the coefficient h for free convection
at the vertical wall. An equation for such a heat transfer coef-
ficient can be found in:[21
h = 1.35 KH (5)
where Eq. (5) is adapted to be used directly for free convec-
tion in air. SI units are used throughout.
Radiation transfer is accounted for by a radiation coefficient
hR. To estimate hR, we can approximate our case by a situation
in which a small gray surface at 6 radiates toward a large gray
enclosure, the room at a . In this case, and if 6 and 6 are not
very different, it can be shown (see any heat transfer textbook,
for instance Reference 3), that hR is proportional to the third
power of the mean absolute temperature:
hR 4 ( + 273)+ (, + 273)J (6)
where e represents the emissivity of the surface and o is the
Stefan-Bolzman constant. This linearization of radiation
fluxes is very convenient and is a great help to account for
the radiation without adding complex equations.
An important qualityfor an engineer is to make
the "right" simplification, i.e., which results
only in slight inaccuracies ....
Note that the convection coefficient, as well as the radiation
coefficient, depends on the outer wall temperature 6w. Indeed,
it is not convenient in the computation to evaluate the outer
wall temperature, so, for estimation of these coefficients, we
will equate the outer wall temperature to that of the liquid.
It results in some inaccuracy for hnv and hR. Eventually, this
inaccuracy is likely to be weak because, for usual materials
and thicknesses, the thermal resistance of the wall is low in
respect to the outer thermal resistance, and the outer wall
temperature is actually not very different from the inner wall
temperature. It does not mean that the thermal resistance
of the wall is neglected here, because it does appear in the
equation of U [Eq. (3)]. The extreme case of an insulating
wall (as for an expanded polystyrene cup, see paragraph 4)
where the inaccuracy is maximum is well described because
the "inaccurate" term has a weak numerical influence in the
computation of U [Eq. (3)].
Heat loss, by Heat Transfer Only, at the Surface of
the Liquid, Qs
It also occurs by free convection and radiation, in parallel,
and is accounted for by a global coefficient h with:
h = hns + hRs (7)
Qs = hsAse( - a) (8)
hn represents the coefficient for free convection at a horizontal
Vol. 41, No. 2, Spring 2007
surface. From Reference 2, for air, it is given by:
h = 1.31 [ -
Heat loss resulting from the evaporation, Qevap
We must first estimate the evaporative molar density of
flux, N . At the interface, air is saturated at the surface
liquid temperature 6, and water partial pressure is equal to
its vapor pressure Pv(6) at this temperature. Far from the
surface, for half-saturated ambient air, the water partial pres-
sure is 0.5Pv(0a). In the case of a single component we can
find explanations in mass transfer textbooks (see for instance
Nw, = kCCT I (P, () - 0.5P, (0a)) (10)
k is the mass transfer coefficient referred to a molar concen-
tration difference at low or equimolar transfer fluxes. F is the
logarithmic mean of the partial pressure of air, P- = P - Pw
at the surface and far from the surface, and it accounts for the
influence of the bulk flow of air. So:
(PT -0.5P, (0a)) - (PT - P (0a))
PT - 0.5P, (0a)
PT - Pv a)
CT is the total molar concentration. Pv(6) can be computed
from a vapor pressure law for water, such as Clapeyron's or
Antoine's law. Here we have used, from Reference 4:
105 96681 66821
P,0) = 10 228+e whereP isinPa, in C (12)
An important feature is now to estimate the mass transfer co-
efficient k . This can be done using the analogy between heat
and mass transfer, as first proposed by Chilton and Colbum.
For the air-water system, because the Lewis number, Le, is
close to 1, it g-n r
kc- hn� (13)
The molar density of the flux is then
Nwa = 1 (P () - 0.5P (a)) (14)
After some rearrangements, using the perfect gas law, mass
Wwat hns~ wa- As, (Pv (0) -0.5Pv (0a)) (15)
Now, knowing the evaporative mass flux, the heat loss by
evaporation, Qevap, is given by:
Qevap= WwatAHv (16)
where AHv is the heat of vaporization of water per kg of
Finally the system of differential equations to solve is:
dO hs31t 1
(mCp +MCp_)- =-hK (0)AK ( O- a)-U (U)AW ( O-a)- AH "a -AF (P (0)- 0.5P1 (0a))
dt "TC,, Cp_ F
dM hns A As(v (0)- 0.5Pv (0a))
dt C,,Cpar F
with initial conditions at t = 0, 6 = 00, and M = M0. This system can
be solved numerically by the variable step Runge-Kutta method, for
instance. For all our computations, we have used a very convenient
commercial software, Mathcad 13, where automatic resolution of
such system of equations is implemented. Listing of the program
can be found at
RESULTS OF THE MODELING AND
COMPARISON WITH EXPERIMENTS
The experimental apparatus, including a numerical thermometer,
a balance, and a stopwatch, is seen in Figure 2. Three different
porcelain cups (No.'s 1, 2, and 3) were used filled with water. The
cups were put on the balance plate, hot water from an electric kettle
was poured in, and the temperature and mass variation of the liquid
were recorded. A piece of insulating material was set under the cups
to prevent direct contact with the balance plate. It proved to be
useful with respect to hypothesis 3. Physical and geometrical data
are given in Table 1. Figures 3 a, b, and c presents the comparison
between experimental temperature and the modeling as described
above. The modeling appears very good, although it slightly un-
derestimates the cooling rates in all cases. A simple explanation
could be that area of the handle was not taken into account in the
computations (indeed, cup No. 2, which gave the best results, had
a small handle). Figure 4 also presents good agreement between
experimental and modeled mass variation. As an example, Table 2
gives computed values of different terms of the equation for experi-
ments of cup No. 2, and relative contribution of each flux can be
appreciated. It can be seen, for instance, that the evaporative flux,
except at the end, is quite significant (see below, paragraph 4). Also
note that radiation and free convection coefficients are in the same
range (around 7 Wm 2 C 1)
Figure 2. Experimental apparatus.
Figures 3. Variation of the
temperature of the liquid
for the 3 different cups:
a) cup No. 1, Mo = 78.6 g,
0 = 22.3 �C, = 82.5 C
(: experiment; - : model)
b) cup No. 2, Mo = 102.9 g,
0o =21.8 �C,00 = 79 C
---: simplified model;
-----: model without
c) cup No. 3, M = 87.2 g,
0o =21.1 �C, = 80 �C
(0: experiment; -: model).
t >,_ -
9U* ---- ---------------
0 20 400 600
45 *------ -------
An4 --- -- -- -- --
0 200 400 oo0
0 200 400 600
Chemical Engineering Education
USE OF A SIMPLIFIED ANALYTICAL SOLUTION
One may feel frustrated to need a numerical solution for the system of differential equations. First note that the mass of water
varies only slightly (less than 3%), so we can suppress the mass balance equation and consider the mass of water as a constant,
equal to M0. Furthermore, Table 2 shows that variation, in respect to the temperature, of heat and mass transfer coefficients
is not very large. We have found that use of parameter values computed at the mean temperature leads to very similar results.
Eventually, we can only consider one simplified differential equation, using averaged values:
dO hM h(a (R 1
(MoCpwat -mCp) -= -h, A� (0- a)-UwavA (0-0a)- " As, (P (0)- 0.5Pv (0a))AHv (19)
dt airCpar Fav
no.-- - --- --- --- ---.
0 200 400 600 800 1000
Nevertheless, even with the proposed averaging, Eq. (19) has still no
obvious analytical solution, due to the exponential term in the expression
of P (6). But, if the function P (6) is approximated by a parabolic equa-
tion, P (0) = b02+ cO + d, we can propose an analytical solution. We found
by numerical fitting, that
b =18.367 c =-1237.2 d = 27753 in the range 40 �C to 80 �C
In this case, Eq. (19) is a differential equation with separated variables,
whose solution is:
t(e) = (M Cp
mC 2 2bB+ A cB 2 2bBe + A + cB (
mCp - arctan 2b + A + -- arctan (20)
S 4b= \- 4b \ + 4bBdm - A2 2AcB - cB
A = -hsavAse Uwavwe
Figure 4. Loss of mass (g) for the 3
a) cup No. 1, M� = 78.6 g,
0 = 22.3 �C, 0 = 82.5 �C
(*: experiment; -: model)
b) cup No. 2, M = 102.9 g,
a = 21.8, �C, 0 = 79 �C
( A: experiment; -: model)
c) cup No. 3, M = 87.2 g,
0 =21.1 �C, = 80 �C
( : experiment; - : model)
Numerical Values Given by the Model for Experiment of Cup No. 2 (M. = 0.1029 kg, Oa = 21.8 �C)
Values are in SI units as given in the nomenclature.
time temp. mass(g) Qevap Qw Qs hnw hns hRw hRs Uwe
0 79.0 102.9 12.0 8.6 2.1 7.5 7.4 7.1 7.6 14.1
90 75.2 102.5 9.5 7.8 1.9 7.3 7.3 7.0 7.5 13.9
180 71.9 102.1 7.8 7.2 1.7 7.2 7.2 6.9 7.3 13.7
270 69.0 101.8 6.6 6.7 1.6 7.1 7.1 6.8 7.2 13.5
360 66.3 101.5 5.6 6.3 1.5 7.0 7.0 6.7 7.2 13.3
450 64.0 101.3 4.9 5.9 1.4 6.9 6.9 6.6 7.1 13.2
540 61.8 101.1 4.3 5.5 1.3 6.8 6.8 6.5 7.0 13.0
630 59.8 101.0 3.8 5.2 1.2 6.7 6.7 6.5 6.9 12.9
720 58.0 100.8 3.4 4.9 1.2 6.7 6.6 6.4 6.9 12.7
810 56.3 100.7 3.1 4.6 1.1 6.6 6.6 6.4 6.8 12.6
900 54.7 100.6 2.8 4.4 1.0 6.5 6.5 6.3 6.8 12.5
Vol. 41, No. 2, Spring 2007 141
Numerical Values of the Parameters of the Three Different Cups
Values are in SI units as given in the nomenclature.
D He ew m X Cp
cup 0.0520 0.0495 0.0040 0.1092 1 970 0.924
cup 0.0512 0.0610 0.0020 0.0642 1 970 0.924
cup 0.0520 0.0635 0.0040 0.1278 1 970 0.924
hns 9t wa 1 A Hv
dm= d- 0.5Pv (0a)
Figure 3b compares the numerical and the analytical so-
lution of the equation, and shows that the simplification is
The case may be further simplified if we consider that
evaporation does not occur (insulating cover on the cup). In
this case Eq. (18) becomes:
(MoCpwt + mCp) - U = -UwvAw (0 - Oa) (25)
which is very easily integrated to
=e (MCp,,,+Cp) (26)
Results of this analytical solution are presented in Figure
3b, showing that the final temperature is significantly higher in
this case. This situation exists in real life. It corresponds to the
"fast food coffee," which is served in expanded polystyrene
cups with a cover that insulates and blocks evaporation. This
absence of evaporation combined with an increased thermal
resistance of the wall (expanded polystyrene has a very low
conductivity) results in very slow cooling. This explains why
we often burn our lips at the end of a fast food meal when we
drink our coffee without precaution, as we cannot imagine it
is still so hot after the duration of the meal!
COMMENTS ON THE HYPOTHESIS OF
HOMOGENEOUS LIQUID TEMPERATURE
We can use the analogy with the well known case of heating
or cooling of a solid. The homogeneity of the solid tempera-
ture is usually assessed by considering the Biot number,
Bi = (27)
where L is a characteristic length of the system. The Biot
number evaluates the ratio between inner conductive trans-
fer and outer convective transfer. When the Biot number is
much smaller than 1, homogeneity of the solid temperature is
insured. In our case, the Biot number can be written as:
Bi= UD (28)
So withUw z 14Wm2C1,D = 4 x 102 m, andX = 0.67W
m' C 1, we obtain Bi = 0.8. This value is not "much" smaller
than 1, but we have considered here that only thermal con-
duction occurs in the liquid, while free convection is actually
present, and greatly increases the inner transfer. For instance,
we can estimate the enhancement of the "apparent" conductiv-
ity by the value of the Nusselt number, Nu. To evaluate this
value, we can use a simplified sketch and consider inner free
convection in a horizontal cell. Equations for such situation
(23) can be found in Reference 9:
(24) Nu= h= 0.069Ra033 Pro
with 3x105 < Ra = H3gA < 7x109
and Pr Cp (29)
In our case, if we want to accept a temperature difference
AO = 1 C, between bottom and surface of the liquid, Eq.
(29) predicts a conductivity enhancement of around seven-
fold that now allows a better fulfillment of the Biot criterion.
Remember that this very simplified approach aims only at es-
timating if we are in the acceptable range. If we now consider
the case of the industrial tank with a characteristic length of
1 m, Eq. (29)-which gives a conductivity enhancement of
120-fold- allows maintaining the Biot number at a low value,
and the hypothesis of homogeneity is still valid.
The agreement between modeling and experiments (Figures
3 and 4) was surprisingly good. Indeed, every experienced
researcher knows that a totally predictive model is often
disappointing and parameter adjustment is common practice
(conversely, students are very confident in these predictive
models!). Nevertheless, be aware of the numerous simplifica-
tions we used that here proved to be reasonable. As a practical
conclusion, note that when preparing a cup of coffee another
scenario is possible: hot coffee from the pot is poured into the
cup. In this case there is first cooling of the coffee by exchange
of enthalpy with the cup. The cup and the liquid quickly reach
an equilibrium temperature, 6,, given by the equation:
0 MCpwCt, + mCpo0
MoCpwt + mCp,
Indeed, the temperature decrease is significant and this
speeds up considerably the desired cooling. Evaluation of the
kinetics of this process is not easy, but is useless because its
rapidity (a few tens of seconds) can be easily demonstrated.
So, an even more efficient cooling process would be to pour
the coffee again into a new cup (as massive as possible), and
repeat if necessary. Because everyday life situations are an
unlimited source of scientific questions, what will happen if
we add sugar to the liquid? Will this influence the cooling
rate? This is another story, worth being discussed-around
a cup of coffee!
A area (m2)
A term defined by Eq. (22)
B term defined by Eq. (23)
Bi Biot number, Eq. (27)
Cp specific heat (J kg 1 C 1)
Chemical Engineering Education
total concentration (Mol m 3)
logarithmic mean of partial pressures of air (Pa)
free convection heat transfer coefficient (Wm2 C ')
radiation heat transfer coefficient (Wm2 C 1)
low or equimolar flux mass transfer coeff. (kg s 'm 2)
characteristic length (m)
Lewis number= ratio of thermal and massic diffusivities
mass of water (kg)
mass of the cup (kg)
molecule weight (kg Mol1)
molar density of flux (Mol s m 2)
Nusselt number, Eq. (29)
heat flux (W m2)
Prandt number, Eq. (29)
vapor pressure (Pa)
evaporative heat flux (W m2)
Rayleigh number, Eq. (29)
global heat exchange coefficient (W m 2C 1)
mass flux (kg s 1)
massic latent heat of water (J kg ')
thermal diffusivity (m2 s 1)
thermal expansion coefficient (K 1)
emissivity of the surface
kinematic diffusivity (m2 s 1)
density (kg m 3)
Stefan Boltzman constant = 5.67 x 108 (W m2 K4)
0 temperature (�C)
1. Cussler, E.L., Diffusion, Mass Transfer in Fluid Systems, p. 24, Cam-
bridge Univ. Press
2. Coulson, J.M., and J.E Richardson, Chemical Engineering, Vol 1, 205,
3. McCabe, L., and J.C. Smith, Unit Operations of Chemical Engineering,
422, McGraw Hill
4. Hirata, H., S. Ohe, and K. Nagahama, Vapor Liquid Equilibria, Else-
5. Treyball, R., Mass Transfer Operations, p. 49, McGraw Hill
6. Chilton, T.H., andA.P Colburn, Industrial and Engineering( ( .....
7. Coulson, J.M, and J.E Richardson, Chemical Engineering, Vol. 1, p.
303, Pergamon Press
8. Parulekar, S.J., "Numerical Problem Solving Using Mathcad," Chem.
Eng. Ed., 40(1) 2006
9. Incropera, EP, and D.P DeWitt, Fundamentals of Heat and Mass
Transfer, Wiley, New York (1985) 1
Vol. 41, No. 2, Spring 2007
---- U s_____________________________________
THE DEVELOPMENT AND DEPLOYMENT
OF A VIRTUAL UNIT OPS LABORATORY
SREERAM VAIDYANATH, JASON WILLIAMS, MARCUS HILLIARD, AND THEODORE WIESNER
Texas Tech University * Lubbock, TX 79409-3121
The success of engineering education relies heavily on
training a student to apply the theoretical knowledge
gained to practical situations. In traditional pedagogy,
the theoretical element is typically provided through class-
room lectures and tutorials. For the practical part, engineering
laboratories play the major role. Luis Ando, et al.,E11 call this
"learning by doing."
According to Hansen,21] only 25% of what students hear
stays with them, about 45% of what students hear and see
is retained, and about 70% of what they do is retained when
they use "learning by doing." Douglas Cooper[31 mentions
that "such practice is motivating, promotes critical think-
ing, facilitates understanding in the use and limitations of
the theory, and helps prepare students for challenges of the
Even though theoretical and practical knowledge are equally
important, they come at different costs. In conventional train-
ing, the theoretical knowledge is imparted through classroom
lectures and tutorials. This knowledge is relatively afford-
able. The practical component, which requires a laboratory
setup, comes more expensively. The costs incurred include
procurement of equipment, setup, maintenance, operation,
and training. Moreover, the laboratory equipment is typically
available only for limited time periods.
BENEFITS OF COMPUTER-BASED EXPERIMENTS
Kadiyala and Cryness]5 published an exhaustive overview
of computer-based instruction in engineering over the past 15
years. Reviewing 760 reports, they found convincing evidence
that information technologies can enhance learning when the
pedagogy is sound, and when there is a good match of technol-
ogy, techniques, and objectives. The use of computer-based
laboratories can also contribute significantly to reducing the
costs of practical training.
Students can access the computer-based laboratories more
easily than accessing physical labs. In fact, every computer
that can run lab software becomes a lab. With the prolifera-
tion of laptop computers, it is literally "lab anywhere."Also,
a student can "redo" the experiments at home and try out new
ideas as soon as he or she conceives them. In a traditional
physical environment, this would be difficult, as the student
must wait for the laboratory classes. Additional time must be
spent to reconfigure equipment to explore alternate scenarios.
Moreover, in many cases, the enthusiasm to try out new things
diminishes with time. Therefore, the computer-based labora-
tory caters to the realization of "flashes of thli iughC" that could
Sreeram Vaidyanath is a software engineer with Microsoft in Bellevue,
WA. He completed a Master's of Science degree in Computer Science
at Texas Tech University in December 2005, specializing in image
registration. Prior to coming to Texas Tech, he earned a Bachelor of
Technology degree from the University of Calicut, Kerala, India. His
home is Palakkad, India.
Jason L. Williams is a doctoral student in chemical engineering at Texas
Tech University. Prior to entering the TTU doctoral program, he earned a
B.S. in chemical engineering from TTU in May 2002, during which time
he also contributed to the Virtual Unit Operations Laboratory.
Marcus Hilliard is a doctoral student in chemical engineering at the
University of Texas in Austin. Prior to entering the UT doctoral program,
he earned a B.S. degree in chemical engineering from TTU in May
2002, during which time he contributed to the Virtual Unit Operations
Theodore F. Wiesner is an associate professor of chemical engineering
at Texas Tech University. His research interests are computer-based
instruction, biomedical engineering, and bioprocess engineering. Prior
to entering academia, he worked in the chemical process industries in
the areas of polymer manufacture and wastewater treatment. He earned
his B.S. degree from Kansas State University, his M.S. from the University
of Houston, and his doctorate from the Georgia Institute of Technology,
all in chemical engineering. He is a member of American Institute of
Chemical Engineers, the American Society of Engineering Education,
and the Biomedical Engineering Society.
� Copyright ChE Division ofASEE 2007
Chemical Engineering Education
Little wonder, then, that computers are increasingly be-
ing deployed in industry.P811] Flowsheeting software, such
as ASPEN or CHEMCAD, is used to design new processes
and simulate existing facilities. Importantly, and perhaps
more frequently, computer technology is used to collect and
analyze data from operating processes to optimize them. Im-
mediately after graduation, an engineer is most
likely to work in an environment where he or
she monitors and perhaps controls the physi- Locate appr
cal equipment from a computer screen. Yet matemr
Sorby, et al., in Reference 7, state that, "The
engineering curriculum has evolved over the
years to include some computer applications,
but computer techniques for the most part Non-dime
are not an integral and pervasive part of the discreet
curriculum as they are in the industrial sector.
To develop students who will succeed in their
engineering careers, it is important that they
be introduced to computer techniques early
in their educational programs and that these Pilot algorit
skills are continuously used and reinforced
throughout the curriculum." Hence, training
students to collect data via computer interfaces
prepares them to serve in such anticipated
industrial environments. Convert to
Further, computer-based experiments pro-
vide a safe venue for students to experiment
in understanding what might happen if they
do not conform to the standard operating
procedure and normal operating conditions. Design us
Some experiments, such as the runaway of an
exothermic reaction, are difficult to treat cost-
effectively and safely in a physical laboratory.
In a computer-based lab, however, a student
can safely explore such scenarios and be better Adapt physic
prepared for real-world situations. Computer- for v
based experiments can likewise actually train
students to run the physical equipment in the
same or subsequent courses. This is the same
rationale behind training pilots in ground-
based flight simulators prior to having them Finisht
fly in real aircraft. In fact, several chemical m
processing firms have implemented training
on dynamic process simulators for their plant
operators prior to the actual running of a fa- Figure 1. Fl
cility. Such virtual training can be of value to ing the des
new engineers as well. each softwa
In this paper we report on our experience in
developing and deploying a computer-based
laboratory'. We conclude that, while the up-front investment
in terms of labor in virtual laboratory development is sub-
stantial, the benefits of augmenting a physical laboratory with
computer simulations can justify the development.
DEVELOPMENT OF THE VIRTUAL
UNIT OPERATIONS LABORATORY
We sought to introduce computer-based experiments into
our curriculum by simulating our senior unit operations labo-
ratory. The unit operations laboratory at Texas Tech University
is employed to reinforce to senior chemical
engineering students the basic chemical en-
te dynamic gineering principles associated with various
moes pieces of process equipment. The major pieces
of equipment used in our laboratory include
a double-pipe heat exchanger, a packed col-
unn ammonia absorber, and a cooling tower.
alize and The unit operations laboratory is also used to
models familiarize students with the safety concerns
regarding each piece of equipment and about
operational issues. The equipment used is
comparable to pilot-scale units of industrial
laboratories. We expect the students to acquire
MathCAD the following specific engineering knowledge
and skills in the unit operations lab course:
1) implement laboratory and process plant
safety; 2) analyze experimental data; 3) under-
stand and apply the theories of shell and tube
Sscripting heat transfer, gas absorption, and humidifica-
tion; and 4) scale equipment to the industrial
level using pilot plant data. In addition to these
specific chemical engineering skills, we require
the students to display significant learning and
erface in development in all the areas of ABET Engineer-
W ing Criterion 3 (a-k).
Guiding Principles in Virtual
;ign process for
re module of the
In our experience, effective virtual labs will
have the following four key principles
(a) Authentic interface. An important feature
of a computer-based experiment is that
it must be as faithful to the physical
experiment as possible. Consequently, it
is necessary to run the experiment at real
time. This will let the student have a "feel"
for the time needed for the experiment in
the physical case. Moreover, the experi-
ment should incorporate as many practical
factors (that make the system deviate from
ideal behavior) as possible. Another way to
improve the "reality" factor is to make the
indicators and controls in the computer-
based experiment as similar as possible to
the physical ones.
(b) Pre-laboratory preparations. In a conventional labo-
ratory course, the students undergo pre-laboratory
A CD containing the VUOLmay be obtained by the reader atno cost
by e-mailing Theodore Wiesner at Ted. Wesner@ttu.edu.
Vol. 41, No. 2, Spring 2007
preparation before the actual laboratory session. This
provides them the background information of what
is happening, the working of the apparatus, and the
results to be expected. The pre-laboratory prepara-
tion usually involves reading relevant sections of the
textbook and/or working out problems that will give
the basic idea about the experiment. For the student to
get the maximum learning experience, pre-laboratory
preparations are important, and hence, this aspect of
the physical laboratory must be carried forward to the
computer-based laboratory also.
(c) Documentation and formatting. As for any software,
computer-based laboratories must be well-documented
and well-formatted. The documentation must be rich
enough to guide the user from start of the experiment
to the calculations and must be as user-friendly as
possible. Thef ... 11,i,, should be easy enough for the
student to follow without excessive training. This helps
a student to learn andpractice the sessions without the
need for another person for guidance.
(d) Equivalent learning experience. It is very easy to
do mathematical computations using computers. In
a computer-based experiment, however, calculations
done manually in a physical experiment should be per-
formed manually in the computer-based experiment as
well. The student learns much of the theory behind the
experiment by way of , ,. i,i,, , various parameters
needed. Otherwise, he or she would fail to appreci-
ate the importance of the various parameters. Once
the student is required to calculate other parameters
using the available data, the relative importance of
the various parameters becomes apparent. In short,
the computer-based experiment must attempt to give
the same learning experience that the physical lab
GENERAL SOFTWARE PARADIGM
This section explains the general software design methodol-
ogy used to create each module of the virtual unit operations
laboratory. The design is a multi-step process and is illustrated
in Figure 1 (previous page). The finished software module is
a LabVIEW virtual instrument (.vi) file.
Locate Appropriate Dynamic Mathematical Models
The starting point for the development of each module is
to find out appropriate dynamic mathematical models for the
process under consideration. The models must be dynamic in
order to faithfully represent the unit operation under study,
particularly its unsteady state behavior.
This involves a thorough review of published models for
a unit operation.
By way of example, we illustrate the mathematical treat-
ment of a double pipe heat exchanger. The energy balances
for this experiment are given in Eqs. (1) and (2).
&T &T 4U
- + v- = (T - T) (tube-side) (1)
at Oz pCpD
&7TL sgn 4D1U
- +sgnv 4U (T-T) (shell-side)(2)
t "Z psCps (D -D2)
The exchanger is subject to the following initial and bound-
T(z,0)= T (z)
T (z,0)= Ts (z)
T(0,t) = Tt (t)
T (0,t)= Ts,ie (t) cocurrent
T (L, t)= T, le (t) countercurrent (3)
T is the tube-side temperature, t is time, v is the tube-
side velocity averaged across the cross-section, and z is the
distance along the exchanger. U is the overall heat transfer
coefficient, D1 is the diameter of the inner tube, T is the
shell-side temperature, and p and C are the density and the
heat capacity of the tube-side fluid. The subscript s indicates
the analogous properties of the shell-side fluid; the subscript
0 indicates initial conditions; and sgn = +1 or -1, indicating
cocurrent or countercurrent flow respectively. L is the length
of the exchanger.
Nondimensionalization and Discretization
We now introduce the following dimensionless variables.
. .me tv (
dimensionless time T=-- (4)
dimensionless exchanger length Z=-
dimensionless tube-side temperature 0= T- iet (6)
dimensionless shell-side temperature 0 - Ts - Tnt (7)
The nondimensionalized energy balances are a pair of
partial differential equations (PDEs).
- + - = a - (0 - 0) 8)
- -=a .(0-0) (9)
The initial and boundary conditions become Eq. (10) in
0(Z,0)= 90 (Z)
s (Z,0)= 0,0 (Z)
S(0, T) = 1 cocurrent flow
S(1, T) = 1 countercurrent flow (10)
Chemical Engineering Education
The quantities a, a , and y are lumped parameters.
PsCps D2-D) V
Eqs. (8) and (9) are discretized into recurrence formulae in
the time dimension to provide an open-ended simulation. Note
that one could use standard, built-in solver utilities in either
MathCAD or in LabVIEW to obtain the numerical solution.
Both of these specify the end-times, however, and solve the
system as fast as the computational platform permits. For
real-time simulation of an experiment, it is necessary to run
an open-ended simulation that allows user interaction with
the simulator as time progresses. Hence, the need to write out
and solve the discretized model in extenso.
We employ Lax's modification to the FTCS method (for-
ward in time, centered in space) to numerically discretize and
solve Eqs. (8)-(10). The discretized forms of the PDEs become:
S 1 -(l+ c). + , I (1- c) . , a -(6s, - ,)(12)
t -1 s+gn cs.)
(1- sgn.c. 5)
j ., - ,)
The index i denotes time, andj denotes space. The quanti-
ties c and c are the Courant numbers for the two sides of the
unit operation, and a and a. are the dimensionless lumped
parameters. The compositions of these four quantities are
given as follows.
c=- =-, cs
a - - - = aAT, Ua
vAt v vAt L
Az v Az/L
psCps (D2 -D2)v
= aSAT (14)
We performed similar procedures using ammonia balances
on the gas and liquid phases of the gas absorber,[141 and energy
Figure 2. Heat exchanger algorithm coded in LabVIEW (block diagram programming mode).
Vol. 41, No. 2, Spring 2007
ly -- . . - - --
�, ", p'-m ,,
I. I - l
r * ""I, % o ~" I "i-.
� ' ' - ,B ^';-.;;.T " rE:: . --:
Iriic **�� I TW.~- _______
| `-1 .*$ .1 r~ , -S-T
I 1^ ^ | ^**"�"urt """" ^ [j^. 77"
1' I-� .. I'�~11�~ "�-I
,II *- � *'� A I4 C
balances on the air and water phases of the cooling tower. 
Interestingly and conveniently, the dimensionless models for
all three experiments have the same generic form as Eqs. (8)
and (9). Thus we obtained recurrence formulae in the time
dimension similar to Eqs. (12) and (13) for the absorber and
cooling tower as well.
Pilot Algorithms in MathCAD
Before coding the discretized mathematical model into the
distributable end product, the algorithm should be tested for
stability and accuracy. Particularly, dimensional consistency
must be assured when combining inputs expressed in different
unit systems such as the SI and English systems. For these
reasons, we piloted our algorithms in MathCAD software. In
the first coding of the algorithms, MathCAD's built-in units
handling was enabled, eliminating dimensional errors dur-
ing translation. Because the scripting language in LabVIEW
does not handle units automatically, the MathCAD pilot was
then modified with automatic unit handling disabled and ap-
propriate conversion factors introduced into the code. When
the results from the code without the units were the same as
the results from the code with the units, we were assured that
we had the correct conversion factors.
Validate Against Physical Laboratory Data
Once suitable dynamic models have been identified, the next
step is to validate the modeling results against experimental
data. In our department, we reconciled our mathematical
models with the results students obtained from the physical
experiments at Texas Tech University in prior years.
Convert MathCAD Code to C-like Scripting Language
As mentioned earlier, authenticity of the interface is an
important factor in maintaining a good learning experience
for the student from the computer-based lab. In the Virtual
Unit Operations Laboratory, we chose to use LabVIEW as the
front-end tool. Of the many factors that compelled us to choose
LabVIEW, the most important was the fact that the controls and
indicators provided by LabVIEW have a real "look and feel."
LabVIEW comes with a palette of indicators and controls that
are very much similar to the physical ones.
LabVIEW offers a C-like scripting language, in which one
can program the algorithms piloted in MathCAD. Figure 2
(previous page) shows the heat exchanger algorithm as coded
in the LabVIEW virtual instrument (block diagram program-
ming mode). LabVIEW was intended to provide a virtual
programmable interface to physical laboratory equipment.
With the scripting feature, however, one can incorporate a
mathematical model into the virtual instrument and use the
flexible user interface to design a facsimile of the simulated
equipment. The scripting block in Figure 2, in other words,
replaces the physical device.
Design the User Interface
The physical heat exchanger and its interface for the double-
pipe heat exchanger are illustrated in Figure 3. It is completely
interactive. The user is able to alter the hot- and cold-water
flow rates, as well as the temperatures of those streams.
Another parameter that can be changed is the direction of
flow, meaning the flow mode can be either countercurrent or
cocurrent. These parameters were chosen as the adjustable
ones because they are the same parameters that are adjustable
in the physical experiment.
The interface features two graphs. One shows the outlet
temperatures of both the shell and tube sides progressing
with time. These values change until the system reaches a
steady state; a change in an inlet temperature will affect both
the outlet temperatures. The magnitude of the change is de-
pendent on both the inlet temperatures and both the shell and
tube-side flow rates. The other graph shows the temperature
Figure 3. The Physical Heat Exchanger and its virtual counterpart.
Chemical Engineering Education
E*u Eb [j1.n fa b te -
-4 ilt * 1- I
am W "m N -! 1
Figure 4. Physical gas absorber (left) and the virtual gas absorber (right).
Mc 3ai-+ an
-, tfhn a wa
-9 >- iil *>*(
Figure 5. The physical and virtual cooling towers.
Vol. 41, No. 2, Spring 2007
Figure 6. The Introduction
Screen to the Heat Exchanger
profile along the length of the
exchanger for both tube and
shell side. With a countercurrent
flow mode, this graph has two
parallel lines along the length
of the exchanger, while in the
cocurrent mode, the graph has
two lines that converge toward
the steady state temperatures.
In both the physical and virtual
versions of the heat exchanger,
the students vary the shell-side
flow rate, and compare the re-
sulting Nusselt numbers with
the Sieder-Tate correlation.
All data displayed on the graphs
are also written to a text file for
Flr Cu ca t HX
The ammonia gas absorber
also has a completely interactive interface, which is a realistic
model of the physical experiment (Figure 4, previous page).
The user can change the flows of water, ammonia, and air as
well as the inlet ammonia concentrations in the two phases.
Using calibration data taken from the actual physical experi-
ment, the dials on the interface match those of the actual ro-
tameters found in the lab. This enables the student to emulate
the actual laboratory experiment.
The gas absorber interface has seven charts. The graphs
show the ammonia compositions of the liquid and gas along
the height of the column, and the ammonia compositions
in the exiting liquid and gas, all as a function of time. To
judge the approach to steady state, the expected steady state
concentration profiles in the two phases are plotted alongside
the dynamic profiles (screenshot not shown). The students
determine the height of a transfer unit (HTU) the number of
transfer units (NTU) the overall mass transfer coefficient based
upon the vapor phase (K a), and the overall column efficiency
(r1) as functions of the liquid to gas ratio (L/G). The simulation
is based on the model of Lakshmanan and Potter.[141
The cooling tower interface (Figure 5, previous page) is
similar to the heat exchanger and the gas absorber, and is also
completely interactive. The student adjusts the inlet air wet
and dry bulb temperatures, fan speed, and the temperature and
flow of inlet water, and observes the outlet temperatures of the
water and air. Two graphs record the exit temperatures of the
air and water as a function of time, as well as the development
of the temperature profiles along the flow path of the water
(screenshot not shown). The model displays breakthrough
behavior, as is often observed in packed vessels. The students
determine the heat transfer coefficient, mass transfer coef-
ficient, and the height of a transfer unit, as a function of inlet
liquid flow rate. We used the cooling tower model published
by Al-Nimr15] to simulate our cooling tower.
Integration into a Complete Suite
After choosing a model and validating it, the next step in the
process of creating the virtual module is to adapt the physical
lab procedures and pre-laboratory preparation to virtual use.
For the VUOL, we integrated the procedures, pre-lab prep,
and virtual experiments into a single self-guided computer
program. An excellent platform for creating such applica-
tions is Macromedia Flash. The student follows a series of
hyperlinks to move nonlinearly among the various sections
of the documentation.
A welcome screen briefs the user on the purpose of the
Virtual Unit Operations Laboratory. In the Using the Product
section, the user is given basic instructions on how to use the
virtual laboratory. At this point, the user must get into any one
of the three modules. Again using the heat exchanger as an
example, we describe the progression of performing a lesson.
At the start of each module, the user is presented with a brief
Introduction section regarding the experiment and the speci-
fications of the particular equipment being modeled (Figure
6). Then, the user is taken to the Pre-Laboratory Preparation
section, which details the various theoretical sections to be
reviewed, and the problems to be worked out (hyperlinked)
so that the student will be ready to understand and appreciate
the process and outcomes of the experiment.
Chemical Engineering Education
Next, the 0,, i.,,ir. Procedure for the particular virtual
experiment is presented. This section includes a thorough
explanation on how to start and stop the experiment, how to
adjust the input parameters and how to collect and analyze
the results. It also contains information regarding the various
process variables and their operating limits. This is also the
section in which the link to execute the LabVIEW simula-
After this section comes the Calculations section. This sec-
tion asks questions, relevant to the experiment module, that the
students must answer. Also, all relevant additional information
necessary to solve the problems (graphs, example figures)
is provided in this section (hyperlinked). The questions are
carefully chosen so that they will hone the skills of the student
attempting to answer them, with regard to the experiment.
Moreover, the Calculations section does not give information
regarding how to answer the problems. This is intentional,
because the Pre-Lab Preparation section already provided
information regarding the basic theoretical knowledge to
acquire and the student is expected to solve the problems in
the Calculations section by applying that knowledge. This
will help the student to develop the ability of applying general
theory to a specific case. The documentation for a module
ends with the References section, which contains citations
used for the compilation of the documentation.
Compilation and CD Authoring
The next step after the completion of the Flash shell is to
combine the modules and the documentation into one unit that
is easy to distribute. To do so, we compiled a CD containing
the modules and the documentation. Also, for increasing the
product's ease of use, the CD has the "autorun" feature, which
brings up a message window providing the user a choice of
two options: 1) Run the VUOL directly from the CD, or 2)
install the VUOL and support files to the hard drive. The CD-
ROM was distributed in August 2004 to 156 departments of
chemical engineering in the United States.
The task of developing the computer-based experiments
is not trivial. First, creating a computer-based lab is a time-
consuming process. The physical experiments under con-
sideration must be modeled mathematically. The modeling
requires careful analysis and study of the basic working of
the process to be correctly simulated. Once the ideal situa-
tion is modeled, the external factors that are prevalent in the
practical case must be identified and should be included in
the computer-based experiment. This analysis takes time and
effort, but it helps create an authentic tool that will familiarize
the student with the practical situations. Once a model that is
close enough to the physical case is obtained, the next step is
to realize the model using computing tools. This again needs a
careful scrutiny of the available tools to choose the best one.
Altogether, to create a computer-based version of a physical
Vol. 41, No. 2, Spring 2007
lab, a large amount of background work is needed.
Second, creating a computer-based experiment requires
skilled labor. In order to prepare the mathematical model of
the experiment, a person competent in the experiment being
modeled is needed. To transform the mathematical model
into a computer-based program, a person who is competent in
computer programming and who can "read" the mathematical
model is required. In only a few cases are these skills found
together in one person.
In replacing physical experiments with computer-based
analogs, the following question arises: How much is stu-
dent learning compromised by the reduction of tactile or
"hands-on" learning? In fall 2002, we conducted a rigorous
comparison between control groups performing physical ver-
sions of the experiments and groups performing the simulated
experiments, the results of which are published in Reference
17. We found no significant difference in learning between the
two delivery modes. The differential impact was measured by
performance on a comprehensive exam, by student feedback
on how well the objectives of ABET Criterion 3 (a-k) were
met, and by student recommendations in oral presentations.
The results indicate that student learning is not adversely
affected by the partial replacement of physical experiments
with computer-based laboratory exercises.
The tactile engineering laboratory should remain an integral
part of the engineering curriculum. Students gain confidence
from turning real valves and seeing real results in a real lab.
Real processes exhibit myriad unanticipated effects and
random influences, which students learn a great deal from
encountering. It is possible to program in some nonideali-
ties, but it is difficult for the course designer to anticipate all
possible effects and to simulate them faithfully.
Nonetheless, in view of the increasing use of computers
in the chemical process industries, the instructional mate-
rial can and should be adapted to the increasing use of
information technology in the manufacturing industries. It
is important that virtual experiments be devised that retain
high fidelity to their physical analogs. For example, if the
laboratory pedagogy requires students to write their own
experimental procedures for physical equipment, students
conducting virtual experiments should also have to write
experimental procedures. The software should be written to
allow this activity. From our experience in the development,
implementation, and assessment of a virtual unit operations
laboratory, we have concluded that a considerable up-front
time investment is required. This investment is justified, in
most cases, by the enhanced reliability and flexibility of the
lessons in conjunction with reduced instructional costs. Given
the many benefits of computer-based instruction and the
prevalence of computer tools in engineering practice, those
designing engineering curriculums should seriously consider
the use of computer-simulated experiments as an adjunct to
their laboratory instruction.
This work was supported by a Special Grant in the Chemical
Sciences from the Camille and Henry Dreyfus Foundation;
fund number SG-01-090.
1. Anido, L., M. Llamas, and M.J. Fernandez, "Labware for the Internet,"
Computer Applications in Engineering Education, 8(3-4) 201 (2000)
2. Hansen, E., "The Role of Interactive Video Technology in Higher Edu-
cation: Case Study and Proposed Framework," Education Technology,
p. 13, September (1990)
3. Cooper, D., and D. Dougherty, "Enhancing Process Control Education
with the Control Station Training Simulator," Computer Applications
in Engineering Education, 7(4) 203 (1999)
4. Murphy, T., V.G. Gomes, and J.A. Romagnoli, "Facilitating Process
Control Teaching and Learning in a Virtual Laboratory Environment,"
Computer Applications in Engineering Education, 10(2) 79 (2002)
5. Kadiyala, M., and B. Crynes, "A Review of Literature on Effectiveness
of Use of Information Technology in Education," J. Eng. Ed., 89(2)
6. Powell, R.M., H. Anderson, J. Van der Spiegel, and D.P Pope, "Using
Web-Based Technology in Laboratory Instruction to Reduce Costs,"
Computer Applications in Engineering Education, 10(4) 204 (2002)
7. Sorby, S.A., G. Walker, M. Yano, V. Glozman, K. Kochersberger, J.
Mathers, J. McKinney, J. Schulman, and M. Young, "Modernization of
the Mechanical Engineering Curriculum and Guidelines for Computer-
aided Engineering Instruction," Computer Applications in Engineering
Education, 7(4) 252 (1999)
8. TTU Chemical Engineering Advisory Board. personal communication
9. Hale, J.C., and H.L. Sellars, "Historical Data Recording for Process
Computers," Chemical Engineering Progress, 77(11) 38 (1981)
10. Anderson, M., and P Whitcomb, "How to Document Process-Moni-
toring-And-Control Systems," Chemical Engineering, 88(18) 117
11. Basta, N., G. Ondrey, and S. Moore, "The Latest Miniature Comput-
ers Let Engineers Access Data, on the Spot," Chemical Engineering,
12. Schiesser, W, and C.A. Silebi, Computational Transport Phenomena,
Cambridge University Press, Cambridge, UK (1997)
13. Pozrikidis, C., Numerical Computation in Science and Engineering,
Oxford University Press, New York (1998)
14. Lakshmanan, C.C., and O.E. Potter, "Dynamic Simulation of Packed-
Type and Tray-Type Absorbers," Industrial & Engineering ( t .....
Research, 28(9) 1397 (1989)
15. Al Nimr, M.A., "Dynamic Thermal Behavior of Cooling Towers,"
Energy Conversion and Management, 39(7) 631 (1998)
16. McCabe, W L., J. C. Smith, and P Harriott, Unit Operations of Chemi-
cal Engineering, 6th Ed., McGraw-Hill, Boston (2001)
17. Wiesner, T., and W Lan, "Comparison of Student Learning in Physical
and Simulated Unit Operations Experiments," J. Eng. Ed., 93(3) 195
Chemical Engineering Education