Citation
Chemical engineering education

Material Information

Title:
Chemical engineering education
Alternate Title:
CEE
Abbreviated Title:
Chem. eng. educ.
Creator:
American Society for Engineering Education -- Chemical Engineering Division
Place of Publication:
Storrs, Conn
Publisher:
Chemical Engineering Division, American Society for Engineering Education
Publication Date:
Frequency:
Quarterly[1962-]
Annual[ FORMER 1960-1961]
quarterly
regular
Language:
English
Physical Description:
v. : ill. ; 22-28 cm.

Subjects

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

Notes

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

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
01151209 ( OCLC )
70013732 ( LCCN )
0009-2479 ( ISSN )
Classification:
TP165 .C18 ( lcc )
660/.2/071 ( ddc )

UFDC Membership

Aggregations:
Chemical Engineering Documents

Downloads

This item has the following downloads:


Full Text


















c e education






























































STATEMENT OF OWNERSHIP. MANAGEMENT ANO CIRCULATION


CHIMIC. EBGIINEEIG EDUCATION 10 1 9 0 A 9/20/91

... I X ..










Raf A Fabien, CheN Eng DepartIOe R o 319 Un-I.erity f Florida, Ga nest Ie, FL 3


Carol C. o Chem Eng Department, o o. UnT l of Florida, Galnes lle, F 32011






















l TI.I NCM. w f2332 2080

-2
| zst bo It lfe B11 ab, o 2- rl
mA r.A A* A*rT oA..-rnel A. *AAAAAA.Aii.A VAT A/AAE











210 1982









A,..bAAAAAAAAAAAAAA~ildAA~TTAAAA CA.,5,/1











EDITORIAL AND BUSINESS ADDRESS:
Chemical Engineering Education
Department of Chemical Engineering
University of Florida
Gainesville, FL 32611
FAX 904-392-0861

EDITOR
Ray W. Fahien (904) 392-0857
ASSOCIATE EDITOR
T. J. Anderson (904) 392-2591
CONSULTING EDITOR
Mack Tyner
MANAGING EDITOR
Carole Yocum (904) 392-0861
PROBLEM EDITORS
James 0. Wilkes and Mark A. Burns
University of Michigan

PUBLICATIONS BOARD

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

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

MEMBERS *
George Burnet
Iowa State University
Anthony T. DiBenedetto
University of Connecticut
Thomas F. Edgar
University of Texas at Austin
Richard M. Felder
North Carolina State University
Bruce A. Finlayson
University of Washington
H. Scott Fogler
University of Michigan
J. David Hellums
Rice University
Carol M. McConica
Colorado State University
Angelo i. Perna
New Jersey Institute of Technology
Stanley I Sandier
University of Delaware
Richard C. Seagrave
Iowa State University
M. Sami Selim
Colorado School of Mines
James E. Stice
University of Texas at Austin
Phillip C. Wankat
Purdue University
Donald R. Woods
McMaster University


Chemical Engineering Education


Volume 26


Number 1


Winter 1992


EDUCATOR
2 Thomas F. Edgar, of the University of Texas, Austin

DEPARTMENT
8 Instituto Tecnologico de Celaya, Arturo Jimenez

AWARD LECTURE
52 Computing in Engineering Education: From There, To
Here, To Where? Part 2. Education and the Future
Brice Carnahan

CLASSROOM
12 Moments with Mathematica,
H. Binous, B.J. McCoy
20 Introducing Water-Treatment Subjects into Chemical
Engineering Education, L. Cdceres, E. Gruttner,
V. Monardes, R. Contreras, B. Gdmez-Silva
44 Chemical Engineering Design: Problem-Solving Strategy,
Richard Turton, Richard C. Bailie
50 Education in Process Synthesis: Application to Inorganic
Processes, J.M. Gutidrrez, J. Gimdnez, M.A. Aguado

OUTREACH
24 Introducing High School Students and Science Teachers to
Chemical Engineering,
Taryn Melkus Bayles, Fernando J. Aguirre

CLASS AND HOME PROBLEMS
30 More Applied Math Problems on Vessel Draining,
Jude T. Sommerfeld

CURRICULUM
34 Ideas About Curriculum, Donald R. Woods

LABORATORY
38 Computer Control of a Distillation Experiment, Carl T. Lira

RANDOM THOUGHTS
18 How About a Quick One? Richard M. Felder

7, 23 Letters to the Editor
33 Book Review



CHEMICAL ENGINEERING EDUCATION (ISSN 0009-2479) is published quarterly by the
Chemical Engineering Division, American Society for Engineering Education, and is edited at the
University of Florida. Correspondence regarding editorial matter, circulation, and changes of
address should be sent to CEE, Chemical Engineering Department, University of Florida, Gainesville,
FL 32611. Copyright 1992 by the Chemical Engineering Division, American Society forEngineering
Education. The statements and opinions expressed in this periodical are those of the writers and
not necessarily those of the ChE Division, ASEE, which body assumes no responsibility for them.
Defective copies replaced if notified within 120 days of publication. Write for information on
subscription costs and for back copy costs and availability. POSTMASTER: Send address changes
to CEE, Chemical Engineering Department., University of Florida, Gainesville, FL 32611.


Winter, 1992









Ire ILeducator


THOMAS F. EDG

of the

University of Texas at A


When Tom Edgar first joined the faculty at U'
he was given an excerpt from a 1966 speech
then Dean of Engineering, which summarized the
faculty member:
He/she should expect to write proposals for research,
special projects; to publish articles, reports, papers, and bi
to-date in his/her profession field; to serve on councils, b
mittees; to maintain the best possible relations with alu
and the business and industry of the region-in short, to
member of the community and to participate in many of i
we all know that these many activities must never oversh
est concern-the student. If our responsibilities to, and
student ever become secondary, we will be violating the ti
when we joined the faculty.
In his twenty years as a faculty member, Tom Ed
this creed, excelling in teaching, research, adminis
fessional service.
The influences that shaped Tom's career begai
young boy growing up in Bartlesville, Oklahoma (w
probably had the highest number of chemical engin


Becky, Jeff, Donna, and Tom on their favorite type oj


Copyright ChE Division ofASEE 1992


AR



lustin


T-Austin in 1971,
by John McKetta,
expectations of a

equipment, and
books; to keep up-
oards, and com-
mni, legislators,
be a responsible
ts activities. But
adow our great-
concern for, the
rust we accepted

gar has pursued
traction, and pro-

i when he was a
which at that time
eers per capital in
the U.S.). His academic instincts
were honed in the eighth grade when
he won the Oklahoma Spelling Bee
and received an all-expense paid
trip to the National Spelling Bee in
Washington, D.C. (this was his first
national meeting!).
Then, when Sputnik was launched
in 1959, thrusting America into the
space race, the resulting heightened
consciousness of science led Tom to be-
gin dabbling in various technical fields.
Encouraged by his father (a metallur-
gical engineer), his mother (a teacher
and a housewife), and an older brother
(an electrical engineer), he undertook
several science-fair projects (including
& one on fuel cells) which whetted his
appetite for science and engineering.
As a high-school senior he won the
ffamily vacation. Bartlesville Science Fair and a trip to
the National Science Fair.


Chemical Engineering Education









During his senior year, Tom's parents became
concerned about his choice of a career, so he took a
battery of special aptitude tests. He scored high on
memorization and vocabulary (the spelling-bee in-
fluence!), but low on spatial reasoning-which indi-
cated he should not become an engineer. Perhaps
organic chemistry would be a good field....
But Tom ignored the aptitude-test results, and in
1963 he enrolled in chemical engineering at the
University of Kansas. He was influenced to con-
sider a research career in his second semester when
his audition for a part-time job to become a radio
announcer for a classical music station was un-
successful. Tom mentioned the audition to J.O.
Maloney, who promptly arranged for a lab-assistant
job with Russ Mesler making high-speed movies of
nucleate boiling.
At KU, Tom's only "B" was in transport phenom-
ena (darn that Bob Bird!); it is ironic that transport
was the first course he was assigned to teach at
Texas. He became heavily involved in campus activi-
ties while at KU, which presaged the future level of
his professional service. Also of benefit were a num-
ber of summer jobs in industry and an NSF under-
graduate research project in process control.
Tom has fond memories of his first experience
with computer programming. The students in his
material and energy balance course were told to
write a computer program for flash vaporization (it-
erative calculation) within a two-week period of
time, and it was suggested that they could learn how
to write the computer program on their own. This
effectively forced Tom to learn how to use the com-
puter! The result was that computer decks took up
more than their share of his desk space during the
next ten years.
With this background, it was only natural for
Tom to study for the PhD under Leon Lapidus at
Princeton, who was probably the leading authority
on numerical analysis and optimal control at that
time. Since Tom had a NSF fellowship, Leon left him
alone to "do his thing," so in his free time Tom
became involved in several athletic clubs and even-
tually started as prop for the Princeton rugby team
for two years.
Tom's PhD dissertation dealt with the "minimum
time" control problem, which is consistent with his
mother's comment that he always wanted to be first
to finish any activity (unusual for a second child).
He managed to finish his PhD in the "minimum
time" (less than three years) even though he was
balancing child-care responsibilities (his wife Donna


was pursuing a Master's degree at Rutgers), athletic
pursuits, and research activities.
STARTING A CAREER
One of the worst times to graduate was in 1971-
there were probably only three or four academic
jobs available that year, and industry was not hir-
ing. As a college freshman, Tom had decided he

Tom's research in coal gasification and
combustion focused on developing
fundamentally based mathematical models in
a field noted for empirical approaches.


wanted to be a professor, so he was delighted when
the University of Texas offered him a faculty posi-
tion with a princely start-up package of $10,000. At
that time there was limited government research
funding (especially in control), few PhD students,
and a teaching load of four courses per year. Tom set
about to identify fundable projects where new mod-
eling and control techniques were needed to solve
the problems.
During the early 1970s, Tom decided to broaden
his background by exploring energy technology. He
was influenced by having an office next to John
McKetta, who was at the time gaining great atten-
tion for his views on the energy crisis. When Tom
came across an obscure reference to underground
coal gasification (UCG), he became interested in pur-
suing research on the problem. As he recalls, the
brief article stated that UCG was an economically
attractive synfuel alternative, especially for deeper
coal seams-but that a major difficulty existed in
that the process could not be controlled and was not
well understood. "A perfect application for modeling
and control," he thought. Since the State of Texas
was blessed with large reserves of deep lignite and
was accustomed to drilling for energy, Tom put to-
gether an interdisciplinary research project involv-
ing chemical, petroleum, and environmental engi-
neering, in addition to geology, which was funded by
NSF, DOE, and a ten-company consortium.
Tom's research in coal gasification and combus-
tion focused on developing fundamentally based
mathematical models in a field noted for empirical
approaches. The general thrust of the UCG research
was to determine those physical and chemical condi-
tions conducive to application of UCG. Between 1974
and 1984, the interdisciplinary group that he di-
rected developed computer-based methods for scale-
up and experimental methods to characterize a given


Winter, 1992









coal deposit for UCG. Mathematical models were
verified in small-scale reactors, combustion tubes,
and at the field scale. These techniques were used by
several oil companies and by government laborato-
ries and included computer models and experimen-
tal correlations for channel growth/sweep efficiency,
product gas composition, gas-solid reaction charac-
teristics, drying and mechanical properties, sulfur
reactions, combustion tube tests, block gasification,
flow characteristics, environmental impact, and tech-
nical and economic evaluation.
Many of these results, presented in some forty
papers, have also been applied to conventional gas-
ification and combustion processes. Of his papers on
UCG, a 1978 AIChE Journal review article and a
book chapter in Chemistry of Coal Utilization (1981),
stand as key chemical engineering references in
the field. Tom's research efforts were recognized by
the AIChE Colburn Award in 1980; it was probably
the only time this award has been given for "coal-
burning."
While the UCG work focused heavily on model-
ing, Tom did not leave the control field during this
period. His interests in multivariable control broad-
ened to include adaptive control in the 1970s. In
1977, he received one of two best-paper awards at
the Joint Automatic Control Conference. His 1981
work with a PhD student, Ernie Vogel, on an adap-
tive dead-time compensator solved a long-standing
problem inherent in many chemical processes, and
has been cited or used by a large number of subse-
quent investigators.

PROFESSIONAL INVOLVEMENT
Tom's leadership in professional activities began
when he and another engineering faculty member
founded the minority engineering program at UT-
Austin in 1974. At that time there were few minori-
ties in engineering and no engineering scholarships
for economically disadvantaged students at Texas.
Tom helped get the recruitment program started,
established a tutoring program, raised $60,000 for
scholarships from industry (a large sum in 1975!),
and served as director for two years. After a success-
ful start, the Equal Opportunity in Engineering Pro-
gram now has the fifth largest number of minority
students in engineering in the U.S.
Having learned how to balance a variety of out-
side activities with a heavy research and teaching
load, Tom became active in a number of groups,
including the CAChE Corporation, the AIChE CAST
Division, and the American Automatic Control Coun-
cil. He was selected by CAChE to edit the AIChEMI


Modular Instruction Series in Process Control and
an issue of Computers and Chemical Engineering on
the "Application of Computer Graphics in Chemical
Engineering." These modules are still being distrib-
uted by AIChE and have been important supple-


A bearded Tom with Dale Seborg at an authors'
meeting in Kuwait in 1983.

ments to the standard textbooks for process control.
Tom also worked with Duncan Mellichamp to de-
velop the eight-volume CAChE Monograph Series in
Real-Time Computing. His success in leading a vari-
ety of CAChE projects led the other CAChE trustees
to elect him as Vice-President of CAChE in 1980 and
as President from 1981-84, solidifying CAChE's fi-
nancial base so the non-profit educational group could
grow in influence during the 1980s.

Since 1981, Tom has served in a variety of posi-
tions in the AIChE CAST Division, and he thus
provided important leadership as the division grew
to the second largest division in AIChE. These of-
fices included Director, Vice-Chairman, and Chair-
man (in 1986). For the past three years he has been
the AIChE Council Liaison to the CAST Division,
having been elected a Director of AIChE in 1988.
Since 1974, Tom has attended the interdiscipli-
Chemical Engineering Education










Tom's leadership in professional activities began when he and another engineering faculty
member founded the minority engineering program at UT-Austin in 1974. At that
time there were few minorities in engineering and no engineering scholarships
for economically disadvantaged students at Texas.


nary American Control Conference every year and
has seen it become the premier automatic control
conference in the world during that time. His service
activities in that organization include Arrangements
Chairman (1974), Finance Chairman (1982), Pro-
gram Chairman (1979), and General Chairman
(1987). He was selected as Director (AIChE repre-
sentative) of the American Automatic Control Coun-
cil (AACC) in 1982 and in 1988 was elected by repre-
sentatives of the six other sponsoring societies as
Vice-President. He is just now finishing up a two-
year term as President and is leaving AACC in ex-
cellent shape for the future.
Tom has also been a driving force behind the
CPC (Chemical Process Control) series of confer-
ences and co-chaired (with Dale Seborg) the second
conference in 1981. He served on the organizing
committee and as session chair in 1986 and 1991
and was also the 1991 conference manager for CAChE
(held at South Padre Island). Perhaps Tom is best
remembered for his role as awards co-chairman in
1986 and 1991; CPC awards are like Harvard Lam-
poon awards and rely on a highly developed sense of
humor.
During the 1980s, Tom also took on a variety of
editorial board activities. He and other colleagues
founded a new journal on subsurface processing in
1978 (In Situ, published by Marcel Dekker), and he
served as General Editor until 1988. In addition, he
has served on a variety of journal editorial boards
including AIChE Journal, Computers and Chemical
Engineering, Chemical Engineering Reviews, Jour-
nal of Process Control, and most recently as process-
control editor of Chemical Engineering Education.
He serves on the advisory board to the chemical
engineering editor of John Wiley and Sons, and is
also highly sought after as a program evaluator (Utah
and Arizona State) and on departmental advisory
committees (Kansas, MIT, and Maryland).
BOOK-WRITING
The decade of the 1980s also resulted in a flurry
of books by Tom and his colleagues. In 1982 he had
three book projects in progress. His first book, a
professionally-oriented book on Coal Processing and
Pollution Control, (Gulf Publishing, 1984), brought
Winter, 1992


together a wide array of information on all aspects of
coal utilization, covering extraction, conversion, and
pollution control.
Tom taught an undergraduate elective course on
optimization since 1972, and as a result joined with
David Himmelblau to write Optimization of Chemi-
cal Processes (McGraw-Hill, 1988). This book contin-
ued the important role that UT-Austin has played in
this area of chemical engineering, beginning with
Beveridge and Schechter's successful book (Optimi-
zation: Theory and Practice) in 1966. The recent
textbook shows how the optimization field has ma-
tured in the past twenty years; its focus is on prob-
lem formulation (modeling) and it emphasizes only
those optimization methods proven to be the best
ones, while extensive coverage on various applica-
tions of optimization is also provided. Optimization
of Chemical Processes is highly student-oriented in
its presentation and sold nearly two thousand copies
in its first year of publication.
Perhaps Tom's best known book is Process Dy-
namics and Control, written with Dale Seborg and
Duncan Mellichamp. This book had its origins in a
five-day short course that Tom, Dale, and Duncan
first gave at the University of California, Santa Bar-
bara, in 1978. It had a very long gestation period due
to extensive class-testing at ten universities (at vari-
ous times) changes in process control technology,
geographical separation, extensive revisions by each
of the three authors, UCSB faculty governance as-
signments, building a house, etc (the excuses and
accusations are endless). In the California "spirit,"
Dale's philosophy was to "sell no wine before its
time." Tom wrote most of the first drafts for the
twenty-eight chapters, and Dale and Duncan then
tore them to shreds (Tom's version). In any event,
the book (published by Wiley in 1989) has received
excellent reviews and is now the number-one text-
book on process control. It won the ASEE Meriam-
Wiley Distinguished Author Award as the top new
engineering textbook in 1990.

RECENT RESEARCH ACTIVITIES
In spite of his many outside activities and re-
sponsibilities as Chairman at Texas since 1985, Tom
has managed to continue his strong research efforts









and is currently investigating modeling and control
applied to such diverse areas as chemical reactors,
distillation columns, pressure swing adsorption, and
most recently, microelectronics manufacturing. While
he was a "lone wolf' in his previous research efforts,
alliances with his UT colleagues Jim Rawlings, Ike
Trachtenberg, and David Himmelblau have permit-
ted him to continue supervising a large number of
graduate students. Tom still does much of the pro-
posal and publication writing, and his students have
learned to expect many corrections on their writing
efforts. During his twenty-year career at Texas he
has supervised forty-one MS and thirty-seven PhD
students, which is due in part to his many good ideas
and effective salesmanship techniques.
Tom's laboratories contain a wide array of highly
instrumented equipment, always connected to some
computer (PCs, workstations, commercial distributed
control systems). Asked about recent technical
achievements, Tom lists several examples:
demonstrating high sensitivity of some nonlinear
control schemes (with Chuck Alsop)
developing theoretically-based criteria for measure-
ment selection in distillation columns (with Wayne
Bequette)
developing a new computationally efficient scheme
for using nonlinear programming to compute model-
predictive controllers and demonstrating it on a
packed distillation column and packed-bed chemical
reactor (with Jim Rawlings, Ashu Patwardhan, John
Eaton, and Glenn Wright)
validating a fundamental reaction-transport model
for a commercial multiwafer low pressure chemical
vapor deposition reactor, the first highly documented
modeling study on polysilicon in the industry (with
Tom Badgwell and Ike Trachtenberg)

Over half of his current research group is focus-
ing on the modeling and control of microelectronics
processes, with support from NSF, the State of Texas,
Sematech, Semiconductor Research Corporation, and
Texas Instruments. Tom says the state of control
technology in the microelectronics industry today is
the same as it was for the chemical industry in 1970
and thus presents many opportunities to have an
impact. His goal is to see model-based feedback con-
trol developed and implemented for a wide variety of
unit operations, mainly in etching and deposition
reactors.
Tom believes strongly in teaching excellence; all
of the books he has written have arisen out of supple-
mental materials developed for his courses (he still
can't believe how he had the time to teach four
different courses for seven years). Students appreci-
ate his sense of humor, his gregarious manner, and


his genuine interest in them as individuals. Because
of his excellent memory, he is able to learn all of his
student's names in one week, aided by his semi-
Socratic classroom pedagogy. Students like the fact
that the door to his office is always open, even though
he is busy as department chairman and as an advi-
sor to many graduate students.
Tom has been extremely effective in influencing
a rise in national visibility for the Department of
Chemical Engineering at UT-Austin. He served as
Department Graduate Advisor from 1979-1985, a
time when PhD enrollment tripled. UT-Austin now
ranks third in the U.S. in PhDs produced and is also
third in research funding. The process control group
is the largest of its type among chemical engineering
departments. While the UT research focus has re-
cently become more fundamental and science-ori-
ented, it is still held in high esteem by industrial
practitioners, as is evidenced by the latest poll by
U.S. News and World Report.
Consistent with his personal juggling act, Tom
recognizes that the Department must have an atmo-
sphere where good teaching and good research are
valued, and where the students will receive a qual-
ity education. Evidence of this at Texas are the engi-
neering teaching awards won by Presidential Young
Investigators and the Outstanding AIChE Student
Chapter Award in 1990 (the last time they won was
when Tom was student chapter counselor in 1974).
Tom also places a high priority on alumni relations
and interactions with industrial supporters.
As Chairman, Tom spends a good part of his time
in setting new directions for the department and in
young faculty development. Adam Heller (who came
to Austin in 1988 from Bell Labs) states, "Tom Edgar's
leadership has allowed the Department to expand
beyond its previous national strengths in polymers,
separations, and process control into two new ar-
eas-microelectronics and biotechnology." Jeff
Hubbell comments, "Tom Edgar spends considerable
time so that young faculty have the opportunity to
become successful."
His friend Bob Seader (at Utah) sums up Tom's
impact on chemical engineering as follows: "Tom is
one of the best known and most well-respected young
engineering educators in the United States. During
a period of twenty years, he seems to have had one
major goal: to do everything he can do to advance his
profession and to help others to best utilize these
advancements. He is a tireless worker of almost
unlimited energy, an inspiration to his many col-
leagues, and a servant to his profession." 0
Chemical Engineering Education










to the editor


EQUILIBRIUM THERMODYNAMICS-REVISITED

Dear Editor:
I am pleased that Williams and Glasser[1,2' have out-
lined the course in thermodynamics that I introduced at
the University of the Witwatersrand in 1982 and lectured
there until the end of 1984. There are, however, a few
points that I feel need clarification.
One reason for the difficulties students often encoun-
ter in thermodynamics is the confusion over variables and
functions. One encounters references in the same context
to H, H(T,P) and H(T,V), for example, in which the single
symbol H means at least three different things: e.g., a
variable and two distinct functions.
An important objective of my approach is to attempt to
overcome the confusion to which this symbolism gives
rise. The symbol for any function is always constructed so
that it explicitly displays the independent variables (as
superscripts) and the dependent variable. Thus the func-
tion from T and P to H is written HTP. On the other hand,
the value of the function at T and P is written HTP(T,P).
One can write
H= HP (T,P) (1)
but not
H=HTP (2)
(at least for the present).
Williams and Glasser explain the symbolism and write
statements like Eq. (1). However, they then apparently
break the rules in a number of ways: a variable is equated
to a function and not its value (e.g., S = S'" in their Eq. 19);
the same symbol is used for a variable and a function (e.g.,
V = V(S)); they talk of "a new function A = U TS"; etc.
Having warned the reader of the importance of not confus-
ing functions and values of functions, they do just that
throughout both papers.
From the outset it is necessary to confine the use of the
term function to things that are functions. A function
always takes the form HTP:
HTP: T,P H (3)
Then A may be a variable, but it is certainly not a func-
tion. The term state function is not used-parameter or
property is better.
If one has particular values for the independent vari-
ables, then one can write

H= HP (T1,P) (4)
for example. Very often one does not, in which case the
presence ofT and P twice in Eq. (1) may appear unneces-
sary. When the students have become familiar with the
notation, and confident in its use, I draw attention to this
apparent redundancy (if the students have not already
done so). I then say that if the context makes clear that the
value of the function is meant, we can agree, from now on,
Winter, 1992


MWl letter


to an abbreviated notation: we may write HTP for HTP(T,P).
The convention is that HTP always means the function
unless the context shows that it must mean the value of
the function. In the latter case, the (T,P) is to be under-
stood even though it is not explicitly displayed. Wherever
there is any chance of confusion or doubt, (T,P) is not to be
omitted. Now, and only now, does it become legal to write
such equations as Eq. (2) above and Eq. (19) of the paper
by Williams and Glasser. I found it necessary to resort to
the sort of fuss and circumspection resorted to here.
If we are taking care to distinguish functions and their
values, we need to do so with derived functions or deriva-
tives as well. Thus
6USV or Us'v
as
is a derived function and
aUsv
aS (S,V) or USV(S,V)
its value (the above convention still applying). It now
becomes possible to agree with Eq. (18) of Williams and
Glasser

T= USV =TSV
The context implies an abbreviation for

T= USV(S,V)= TSV(S,V) (5)
Williams and Glasser write (their Eq. 25)

a (aaV (6)
This violates the rules of notation just as much as
ap
as
does. One needs to write

a v (7)

Otherwise one has no means of distinguishing the expres-
sion from

a rau^sv S(8)

for example. While the order may normally be reversed
with impunity in Eq. (7) (that is, when the superscripts
repeat), it cannot in Eq. (8) (that is, when they do not). On
the other hand
a2USV
asav
is unambiguous. There are hidden traps in Eq. (6) out of
which I have rescued more than one student. Students
easily obtain cross-differentiation identities that are sim-
ply not true.
My equation

dH_ aHTP
dP=0
(Eq. 9 of Williams and Glasser) I have preferred since
Continued on page 29.
























MA


INST

Ti



ARTURO JI
Institute Te
38010 Celay

M exico
her in
lems that w(
important tr
which indig
local process
Mexico need
The Inst
sixty techno
tem of Tech
kept under
Guanajuato
northwest o
department
dynamic de]
Tradition
a complete
grees, due ii
1980 only or
however, ou
PhD degree
thirty-seven


,,, r. Tri


ChE.


apartmentt )












OCEAN
PACIFIC
ain entrance to the Institute.



ITUTO

ECNOLOGICO

DE CELAYA

IENEZ
cnoldgico de Celaya
'a, Gto. Mexico

has been trying to improve her scientific progi
dustrial productivity ever since the heavy econo
ere experienced in the 1980s. Industry presently
transitional period from simple technology users
:enous resources must be used more intellige
ses must be more fully developed. To meet this c
Is qualified engineers.
ituto Tecnol6gico de Celaya, founded in 1958,
logical institutes in Mexico, all part of a Nati
nological Institutes. The size of each institute
3000 students. Celaya is a province of the
and is located in the center of Mexico, about
f the capital, Mexico City. The chemical en
in Celaya has been recognized as one of Mexi
apartments in recent years.
nally, few universities in Mexico have had facu
or nearly complete staff of professors holding
i part to a lack of available professionals at tha
ne faculty member in Celaya held a PhD degre
lr faculty has expanded to include seven mem]
s (see Table 1). The average age of the faculty
i, and their backgrounds include visiting appo
Copyright ChE Division, ASEE 1992


rams and
mic prob-
y faces an
to one in
ntly and
challenge,

is one of
onal Sys-
has been
state of
150 miles
gineering
co's most

ties with
PhD de-
t level. In
e. Today,
bers with
ty is only
intments


Partial view of the ChE laboratory
Chemical Engineering Education


Students controlling one of the reactors.










at places such as the Universities of Wisconsin, Texas
at Austin, California at Davis, Texas A&M Univer-
sity, and CSIRO in Australia. We hope to establish a
solid faculty by sending our professors abroad for
further graduate studies in the future, and possibly
by adding new members to the staff. Eight faculty
members are currently involved in PhD studies (see
Table 2), and our eventual goal is to have a staff of
fifteen professors with PhD degrees by the mid-point
of this decade.

THE UNDERGRADUATE PROGRAM
The undergraduate program began in 1961 and
has grown to a current figure of 300 enrolled stu-
dents (see Figure 1). Female students have been
enrolling at higher rates in recent years; from six



TABLE 1
1991 ChE Faculty, Instituto Techn6logico de Celaya

Rafael Ch6vez (PhD, University of Utah)
Alejandro G6mez (PhD, University of Utah)
Arturo Jim6nez (PhD, University of Wisconsin)
Gustavo Iglesias (PhD, Texas A&M University)
Alberto Ochoa (PhD, University of California, Davis)
Pedro Quintana (PhD, University of Texas, Austin)
Antonio Rocha (PhD, University of Texas, Austin)
Maria-Guadalupe Almanza (MS, Tecnol6gico de Celaya)
Carlos Cardenas (MS, Tecnol6gico de Celaya)
Eleazar Escamilla (MS, Universidad de M6xico)
Teresa del Carmen Flores (MS, University of California)
Irma-Concepci6n Galindo (MS, University of North Texas)
Salvador Le6n (MS, Portland State University)
David Trigueros (MS, Tecnol6gico de Celaya)
Ma. de los Angeles Vazquez (MS, Tecnol6gico de Monterrey)
Angel Vazquez (MS, Tecnol6gico de Monterrey)
Juan Ledesma (MS, Universidad de Guanajuato)
Emma Torres (MS, Universidad de Guanajuato)


TABLE 2
ChE Faculty Currently Pursuing Ph.D. Degrees


Professor
Francisco Javier Alvarado
Cristina Coronado
Guillermo Gonzalez
Gloria Maria Martinez
Ramiro Rico
Julio Rocha
Fernando Tiscarefio
Rodolfo Trejo


Yea
Termi


University


Texas A & M University
Universidad de Barcelona
University de Salamanca
Universidad Aut6noma Metropolitana
Princeton University
Imperial College
University of Wisconsin
Universidad Aut6noma Metropolitana


1'
1I
1!
1I
1I
1
1!
1!


We try to give our students good exposure
to laboratory activities, with the result that the
chemical engineering laboratory is used in three
courses on unit operations in addition to one
in kinetics and reactor design

percent in 1980 to twenty-nine percent in 1989. About
forty chemical engineers graduate each year from
our program.
The undergraduate curricula comprise typical
chemical engineering courses together with addi-
tional courses emphasizing industrial engineering.
We try to give our students good exposure to labora-
tory activities, with the result that chemical engi-
neering laboratory is used in three courses on unit
operations in addition to one in kinetics and reactor
design. The laboratory includes several units on a
pilot-plant scale for operations such as distillation,
absorption, extraction, evaporation, and drying.
There are also sections for chemical reaction,
solids treatment, and transport phenomena, and
students must take two courses on analytical chemi-
cal analysis. Recently, we have also incorporated
computer-aided techniques more significantly into
our teaching.

THE GRADUATE PROGRAM
Our graduate program (one of the youngest in
Mexico) began in 1980 and offered an MS degree in
chemical engineering. Initially the MS program con-
sisted of nine courses and a thesis project. There was
a set of core courses on thermodynamics, applied
mathematics, transport phenomena, reactor design,
and separation processes. Elective courses included
process design, process control, design of experiments,



400



300 N

rof
nation 200 -
992
992
993 100
992
993
0


993U
992
992


80 81 82 83 84 85 86 87 88 89 90
Year

FIGURE 1. Students enrolled in B.S. program.


Winter, 1992










An interesting and vital challenge for us is to offer a solid PhD
program that emphasizes research activities of interest to Mexico while at the
same time keeping the solid technical basis that characterizes many of the graduate
programs in the first-world universities.


Operation of a
double-effect evaporation system
and selected topics on thermodynamics and trans-
port phenomena.
We revised our program in 1985, giving more
emphasis to research by reducing the academic load
by two elective courses. This change yielded positive
results. It shows the history of admissions and gradu-
ating students for the Master's program and demon-
strates that our graduating efficiency has improved
significantly in the past few years. The low gradua-
tion rate noted in earlier years was not unique to our
graduate program, but in fact has been a problem of
major concern to most Mexican universities offering
graduate programs in chemical engineering.
One system we have implemented for supervis-
ing our students' research progress is to assign a
committee (in addition to the major advisor) from
the beginning of the thesis project. At the end of
each semester, the student must present a seminar
before the committee, and based on the progress
reported, a grade for the research work is assigned
by the committee. The student also benefits from
committee feedback at each step.
We have also made special efforts to recruit top
students for our graduate program, and this fact
accounts in part for our recent good results. Some of
our MS graduates have gone for further PhD work
at universities such as the University of Wisconsin,
University of Texas at Austin, Texas A&M Univer-
sity, and Princeton University.
In Mexico, graduate programs offering PhD de-
grees are new and at present there are only three


universities where one can be obtained. The Autono-
mous Metropolitan University (since 1984), the Na-
tional Autonomous University of Mexico (since 1988),
both located in Mexico City, and just recently, the
Technological Institute of Celaya. It was not until
1991 that the first PhD degree was granted by the
Autonomous Metropolitan University.
An interesting and vital challenge for us is to
offer a solid PhD program that emphasizes research
activities of interest to Mexico while at the same
time keeping the solid technical basis that charac-
terizes many of the graduate programs in the first-
world universities. For this challenge to be success-
fully met, we must build a proper infrastructure for
research activities.

RESEARCH ACTIVITIES
As a natural consequence of our pioneering ef-
forts to establish a solid graduate program in Mex-
ico, part of the job has been devoted to obtaining
financial support for research activities, a task that
has been complicated by the economic situation in
our country. We have, however, made significant
progress.
Until 1982, federal support for research activi-
ties was weak, but in spite of our economic problems
the situation has improved steadily since then. For
instance, CONACYT (the National Council for Sci-
ence and Technology) initiated a special program
(Programa de Apoyo al Posgrado) that offers support
for improving the infrastructure at selected univer-
sities. Celaya has received support through this pro-
gram since 1984, and as a result our computing and
laboratory facilities have expanded significantly.
Support for individual research projects has also
been received from COSNET (the Council of the
National System of Technological Education). The
support from these two agencies has particularly
influenced our growth.
Our research programs have centered on com-
puter-aided design, experimental work on separa-
tion processes, process control, and production of
specialty chemicals. The computing facilities in the
department are based on a MicroVax II system, which
is also used as a network server for twenty PCs.
Some process simulators and software for process
Chemical Engineering Education









control studies are available. To support experimen-
tal work, we have some of the newer analytical equip-
ment such as GCs, atomic absorption, TGA, etc.,
along with the more conventional equipment.
Some major results of experimental work con-
cerned two projects for the production of ferrous
fumarate and oxalic acid. The work was undertaken
due to an interest expressed by industry, and as a
result there are now industrial plants for the pro-
duction of these two specialty chemicals. The eco-
nomic benefit for our institution and for the research
team, however, has been marginal (a fact that has
produced strong criticism from our politicians!). The
experience has been part of a learning process for us.
These two projects are indicative of our efforts to
bridge the traditional gap between academia and
industry in our country. We have also provided serv-
ices to industry in the form of instrumental analysis
and other specific research projects. In addition (and
taking into account our previous experience), we are
exploring routes for the manufacture of specialty
chemicals that are currently being imported by
Mexico, and a multi-purpose pilot plant is nearing
completion.
Celanese Mexicana has recently given additional
scholarships for four graduate students who will
develop their research projects on topics of general
interest to the company.

OTHER ASPECTS AND OUTLOOK
Professor J.M. Smith (University of California,
Davis) was associated with our graduate program
between 1984 and 1987. He twice taught a graduate
course on reactor design and helped conduct a re-
search project on the thermodynamics of three-phase
reactor systems. In our effort to stimulate excellence
in the department, in 1984 we instituted the J.M.
Smith award, given each year to the graduate stu-
dent with the highest GPA. Seven students have
received the award and four of them are presently
pursuing the PhD degree.
A similar award, given in honor of Professor Jos6
Martinez-Avella, one of the first lecturers in our
institute, was instituted in 1989. The recipient of
that award is the student graduating from our under-
graduate program with the highest GPA.
In order to stimulate research activities, in 1984
the Mexican government established a National Sys-
tem of Investigators. Eleven professors from the
Celaya department have been inducted into the sys-
tem, the highest representation of any of the techni-
Winter, 1992


cal institutes in Mexico.
Each week a lecturer is invited to address the
faculty and graduate students on some topic of inter-
est. On occasion the speaker will be one of our own
graduate students talking about his or her research
project. These seminars have been held since 1980.
Each year for the past eleven years, we have
organized a week of chemical engineering, the Semi-
nario Anual de Ingenieria Quimica, an event that


Students undertaking an experiment
on chemical reaction analysis.


has now become a tradition in Mexico. The main
part of the event is a series of courses given by
recognized professionals from both Mexico and abroad
to an audience of national professors, graduate stu-
dents, and practitioner engineers. Among the past
participants in this Seminario have been J. M. Smith,
Octave Levenspiel, Charles Hill, Brice Carnahan, J.
D. Seader, James Fair, Warren Stewart, Edwin Light-
foot, Carlos Smith, Richard Felder, Charles Holland,
and George Stephanopoulos.
We also attempt to share nonacademic activities
with our students. Thus, activities such as our "Thank
God It's Friday" basketball games have become tra-
ditional events enjoyed by everyone.
Recent developments in Mexico seem to indicate
that those who are in government positions are ac-
quiring a better awareness of the importance of aca-
demic and research activities. Industry, too, is show-
ing more interest in academia. Scholarship programs,
such as CONACYT, are essential for providing
greater numbers of qualified professionals who will
be capable of meeting the challenges of the future.
Consolidating a modern chemical engineering
program in Mexico will not be an easy task, but
those of us at Celaya hope to take a leadership role
in effecting the needed changes. 0










classroom


MOMENTS WITH MATHEMATICS*


H. BINOUS, B.J. McCoY
University of California
Davis, CA 95616

M athematica, an interactive software system that
does computer algebra and other mathemati-
cal computations, is a valuable tool for solving large
problems involving symbol manipulation. Mathe-
matica symbolically performs algebraic and calculus
operations such as factorization, substitution, linear
algebra, solution of polynomial equations, limits, and
evaluation of integrals and derivatives. Numerical
computations and visualization of advanced func-
tions by two- or three-dimensional contour plots are
also attractive features of the system. Although more
efficient packages (e.g., Matlab) are available for
large-scale numerical tasks such as matrix inver-
sion, Mathematica can algebraically perform ma-
nipulations of this type. Many potential applications
of computer algebra systems in chemical engineer-
ing research and education are possible to imagine,
and in this article we discuss some examples.
The developer of Mathematica, Stephen Wolf-
ram, is the author of a helpful manual called Mathe-
matica: A System for Doing Mathematics by Com-
puter (Addison-Wesley Publishing Company, 1988).
The book is readable and readily understood by those
with a general familiarity of mathematics and com-
puters. Our experience is that studying the manual
with the software at hand to work the numerous
illustrations is helpful in learning the system. Op-
erations with Mathematica can also be programmed
(e.g., do-loops for iterative computations); see, for
example, Programming in Mathematica.J11
Several computer algebra systems are available
on the market and promise to automate mathemati-
cal computations. Mathematica is particularly at-
tractive because it can be installed in a desktop
computer, although speed is compromised. In its
workstation configuration, however, Mathematica is
fast and user-friendly. Computer algebra systems
probably will radically change problem solving in
applied mathematics; tedious mathematical compu-

Mathematica is a Trademark name owned by Wolfram Research
Inc., PO Box 6059, Champaign, IL 61821-9902


stations will be much less important than the funda-
mental problem formulation.
Opportunities are plentiful in chemical engineer-
ing research and teaching for the application of com-
puterized symbolic manipulation. As an example of
its use, we have applied Mathematica to the compu-
tation of moment expressions for several problems
in separations and chemical reaction engineering.
MOMENTS IN CHEMICAL ENGINEERING
The computation of moments is useful for (1)
interpreting experimental concentration profiles, (2)
determining mass transport parameters from ex-
perimental data, (3) predicting concentration histo-
ries, and (4) designing and scaling-up separation
processes.
The temporal moments are given by

mn(z)=(-1)n lim dnc/dsn = tnc(t,z)dt (1
s-->0 J
o
where the Laplace transform of the concentration is
defined by



o
c(s,z)= fc(t,z)e-stdt (2)
0
Numerical or analytical integration provides values
or expressions for the moments when c(t,z) is known
or when c(s,z) can be derived from a model of the


Housam Binous, born
ceived a general engine
technology electives) fro
Superieure des Mines d
cently received his MS
neering from UC Davis.
are in separation science
puters in engineering.


Ben McCoy received his BS in chemical engi-
neering from the Illinois Institute of Technology
(1963) and his MS and PhD from the University
of Minnesota (1964, 1967). He has been at UC
Davis since 1967, serving as Department Chair
from 1980-88. He is currently Associate Dean for
Research in the College of Engineering. His re-
search interests are in chemical reaction engi-
neering and separation processes.


in Tunis in 1964, re-
eering degree (with bio-
om the Ecole Nationale
le Paris in 1988. He re-
legree in chemical engi-
His research interests
e and application of com-










AaAV


Copyright ChE Division, ASEE 1991
Chemical Engineering Education









process. Thus, we avoid inverting c, a task that
often is not feasible. The algebraic manipulations to
obtain the moments from a mathematical model are:
1. solution of differential equations to
determine c
2. derivatives with respect to the Laplace
transform parameter s
3. limits as s -4 0
The development of c in a series provides an easy
way to find the lower-order moments. Expanding e st
in Eq. (2) leads to

C(s,z)= sn (-1)nmn (z)/n! (3
n=o
The reduced moments are defined as
pn(z)=mn/m0 (4
and the central moments are given by

Pn (z)= (1/ m)(t- g)n c(t,z)dt (5:
o
The binomial expansion of Eq. (5), written in terms
of the binomial coefficients, yields

n(z) 1/mo) ()mni (6:

The first four moments have an important physi-
cal significance. The zeroth moment, m0, is the area
under the concentration plot c(t) and is proportional
to the mass of the species. The mass balance ensures
that mo(z) = mo(z=0) for species that do not react.
The first reduced moment i',, is the position of the
peak and gives the average position of the species.
The second reduced central moment, 92, measures
the spread of the peak. The third reduced central

TABLE 1
Hermite Polynomials and Coefficients of
Gram-Charlier Expansion, Eq. (12)

Ho(x)= 1 (1- 1)
Hi(x) = 2x (1-2)
H2(x)= 4x2-2 (1-3)
H3(x) = 8x3 12x (1-4)
H4(x)= 16x4 -48x2 +12 (1-5)
ao = 1/22iT2 (1-6)
a = (1-7)
a2= 0 (1-8)
a3 = 3/(3!22,Fi2) (1-9)
a4 =(4/2 3)/(4!4 2~12) (1-10)

Winter, 1992


For many cases, mathematical models provide
the moment expressions in terms of mass
transport parameters and geometric properties.
From experimental methods, we obtain the
concentration c(t). Then the moments can be
obtained by numerical integration...

moment, P, is a measure of the skewness of the
peak.
For many cases, mathematical models provide
the moment expressions in terms of mass transport
parameters and geometric properties. From experi-
mental methods, we obtain the concentration c(t).
) Then the moments can be obtained by numerical
integration using Eq. (1). Equating the experimen-
tally determined moments to the moment expres-
sions based on a model of the process represents a
method for obtaining equations where the mass trans-
port parameters are the unknowns. Schneider and
Smith'2' obtained estimates for the intraparticle dif-
) fusion, axial dispersion, and adsorption coefficients
using this method. Mitchell13' used probabilistic ar-
guments to obtain the spatial distribution, mean
position, and variance about the mean of macro-
molecules moving in an external field and undergo-
S ing reversible isomerization. By comparing these re-
sults to experiments, the author was able to provide
values for the forward and backward switching rates.
The same problem was treated by Killalea and
McCoy,'4' who derived expressions of spatial moments.
The concentration profiles can be constructed us-
ing the moments in a Gram-Charlier expansion"51
i.e.,
c(t,z)=moe-X2 a anHn(x) (7)
n=0
where
x=(t- )/ 2u

Expressions of ao, a1, a2, a3, and a4 are depicted in
Table 1. The coefficients of the series, a., depend on
the reduced central moments and are obtained using
the orthogonality condition satisfied by the Hermite
polynomials, Hn,

fHn (x)H (x)e-2 dx=2nn! nm (8)

where e- 2 is the weighting function. The first five
Hermite polynomials are given in Table 1. This pro-
cedure is useful for predicting concentration pro-
files when values of the model parameters
are known. Breakthrough curves for fixed-bed ad-
sorbers and reactors can be represented using mo-
ments of the impulse response."6 As an alternative
to Hermite polynomials, the Laguerre polynomials
13









are frequently used."'
Mehta, et al.,18' were able to obtain agreement
between experimental elution curves and Hermite
polynomial series representation. In addition, fitting
the two results provided a criterion for the trimming
of tails and leading edges. Thus, it is possible to
account for phenomena such as end effects. Elution
curves in chromatographic columns are nearly Gauss-
ian when dispersion is small relative to convection.
For such cases, the first term in the series, the
Gaussian approximation, will provide an accurate
representation of the pulse.
Chromatography, a widely used separation proc-
ess, utilizes differences in species behavior, e.g., solid-
fluid adsorption, to effect the separation. Reliable
ways that make use of the moment theory to analyze
and optimize chromatographic separations have been
developed. For two solutes A and B, separation is
usually satisfactory when


tion, the adsorbed molecules undergo a reaction that
is assumed first-order in the adsorbate concentra-
tion. The Laplace transform of c(t), c(s), is readily
found from the governing equations listed below.
* Mass balance in the fluid
Eac / at= 0(co c)- Apkp(c- ci(R)) (11:
* Mass balance on an individual particle

paci /3t=(Di /r2)a(r2 ci /3r)r-kadsCi +kd a (12:
Reversible adsorption with first-order surface
reaction
ac /at= k dsci -kdc -krc (13

Initial conditions


c=c.=c =0
1 a


for t=0


* Boundary conditions
(aci /ar) r=00 and Di(c;i /Ir)=R =kf(c-ci)r= (15)

* The inlet concentration is considered an impulse


s= A -1A + 1
The resolution, Rs, provides a
criterion for separation of two
species when narrow pulses
are injected. We can define the
Height Equivalent to a Theo-
ci;ital P1at o Pont' v by('-


HETP=L4i2 '/2


(10)


Contrary to R,, HETP depends
only on one input sample size
and is an important charac-
teristic for evaluating chroma-
tographic processes.

EXAMPLE PROBLEMS
To illustrate how Mathe-
matica can be used to assist in
finding moment expressions
for problems of interest to
chemical engineers, we discuss
three examples. A list of es-
sential commands to solve
these problems is available
from the authors.
Fluid-Solid Adsorption and
Reaction in a CSTR
We consider a continuous-
flow stirred tank reactor
(CSTR) fed with an input pulse
of adsorbable species. As a rep-
resentation of a catalytic reac-
14


Es+e+Apkp-RApk /(D[bRcoth(bR)+kpR/Di -1])

b= Ps+kads -kadskd/(S+kr +kd)
D,
0c0
mo Oco

b kads -kadskd/(kr +kd)
b= Di
v=[boRcoth(b0R)+kpR Di-1]
= +Apkp -RApk /(Div)

0C E+RAPk R o R2(1coth(boR) R2 thboR)) / Dv2
ml- W2

(=(P+kadskd/(kr +kd)2)

c2c + RAp coth(bR) R2 (1-coth2 (boR)) (D2)
m2 cRA pk 2Dibo 2DR2 J Di

29coRAk 2Rcoth(boR) R2 1-coth2 (boR))2
+ 2 p kP + !q -


TABLE 2
The Laplace Transform and Temporal Moments for Fluid-Solid
Adsorption and Reaction in a CSTR


(2-1)

(2-2)

(2-3)

(2-4)
(2-5)
(2-6)


(2-7)

(2-8)


Y2Div3 2Dibo 2Di )
coRApk2~ R2coth(boR) R22(1-coth2(boR)) 2R32coth(boR)(1-coth2(boR))
4DDi2b2 4Db3 4D2b2 4D jbo
OcoRApkp Rkdskd coth(boR) R2kadskd(-coth2bR)) (2-9)
P P1 1(2-9)


l2Div2 (Dibo(kr +kd)) (Di(kr+kd)) J
Chemical Engineering Education


re l P a t u vonvc 7 I









Co(t)=C08(t) (16)
The complex nature of the expression for c(s),
determined from the Laplace transformed solution
of Eqs. (11) to (16), makes using Mathematica to
compute mo, ',4, P2 very attractive. We provide ex-
pressions for c(s), and for mn, m,, and m2 in Table 2.

Chromatographic Separation Based on
Fluid-Solid Adsorption

Let us consider separation processes based on
fluid-solid adsorption such as the adsorption of hy-
drocarbons on silica gel in a chromatographic col-
umn.'[2 The concentration of adsorbing gas, c(z,t), is
the solution of the following equations:

* Mass balance of adsorbable component in the gas
phase

(EA 2c/ 2 vac/ z-ac/ t- 3D (1- a)/(R(x)- =0

(17)
* Mass balance of adsorbable component in the par-
ticle
(D. /p)(r2Ci 2 +2/raci /ar)-aci l/t-(pp /pB)ca /t=o
(18)
* Linear rate of adsorption
aCads /t=kads(Ci -Cads /KA) (19)
* Boundary conditions


TABLE 3
The Laplace Transform and Temporal Moments for
Chromatographic Separation Based on Fluid-Solid Adsorption


c(s,z) = (C /s)(1-e -tOA )e-

y=-(va/2EA)+ va/2EA )2+(sa/EA)(1+h(s))
I /s-
h(s)= (3kf/R)((1-a)/a )/s-
1/ ((sDi/kf )v-coth(RV)


-= (sp/Di)(l+pKAkads (P(KAS+kads)))
mo = mo(z= 0)
I1 = (z/v)(l+80)+I*(z=0)
2 = (2z/v)[1+(EA/a)(1+/o)2(1/v2)+12(z=0)
0 = ((1-a)I/a)(1+(KAPp/P))
SKAPp/(Pkads)+
S=R2p/ 15)(+(ppKA )) (/D -5/(Rk,))

When c (t = 0)=c for 0 5 t toA we have the following results:
mo =cotOA 11(z=O)=t0A/2 92(=O)= t2A/12 (


* Initial conditions
c=0
c.=0


at z>0 for t=0
at r>0 for t=0


* Bed-inlet condition
c=co(t) at z=0 for 0 c=0 at z=0 for t>tOA (22)
Kubin19 and KuceraE101 solved this system of equa-
tions and obtained c(s,z), the Laplace transform of
c(t,z), given by Eq. (3-1) in Table 3. Expressions for
mIo, pi, and 92 were also presented by Schneider and
Smithl' (see Table 3). The computations done by
hand require many hours of tedious labor. Simple
and fast computations with Mathematica provide
an identical result. The procedure is straightfor-
ward when a few special techniques are applied.
Instead of X defined by Eq. (3-4) in Table 3, we use
its development in series around the point s = 0 to
order s7. Around the point k = 0, we apply the series
itanh(R-)=Rx (R3x2)/3+2(R5X3)/15+0(X3) (23)
To avoid taking the derivatives of X1/2 around k = 0,
which presents some problems, we substitute the
left-hand side of Eq. (23) into the definition of h(s),
Eq. (3-3) in Table 3. This allows one to obtain a
development in series of h(s) around the point s = 0
to order s4, which is used in Eq. (3-2) to develop the
series for y.


(3-1)
(3-2)

(3-3)

(3-4)
(3-5)
(3-6)
(3-7)
(3-8)

(3-9)


3-10)


Finally, we are able to compute the
derivatives of c(s,z) and obtain j',, and ,2.
Unfortunately, rearranging the result into
a neat form requires human judgement
and is not convenient with Mathematica.
However, this task is readily performed by
hand.

Spatial Moments of Moving and
Interchanging Isomers
We are interested in finding the spa-
tial moments of two isomers A1 and A2,
which are moving, and switching back and
forth between the two isometric states as
a first-order reaction. This system describes
electrophoresis, gel filtration, or sedimen-
tation if the species are moving respec-
tively in an electrostatic, velocity, or cen-
trifugal field. The system is a type of chro-
matographic reactor.
The governing equations of the concen-
tration are


Winter, 1992


D. (c. /ar)rR =k,(c-c.)
(aci/ar)ro =0 for t>0


+s(l-(Di /Rk,)))










ac, 2c ac
-t=D1 -v -k 1 +k2c2 24
-2=D2 -V 2 -k c +k c
2t aX2 X 2 ax 2 2 1

where D. is the diffusion of axial dis-
persion coefficient, v. is the species
velocity, and k is the rate constant
for isomerization.

* The boundary conditions are
c (t,x-->=0) for j=land2 (25)
* The initial conditions are infini-
tesimally narrow distributions
c.(t=0,x)=f6.(x) for j=land2 (26)
with the total initial concentration
given by f, + f2 = 1.
Mathematica provides (a) the Fou-
rier transform of c.(x,t) defined by


cj(k,t)= c (x,t)e-ikx dx


and (b) the limits of the successive derivatives of c..
In terms of c.(x,t) and c.(k,t), the nth spatial moment
of species Aj is given by

m (t)= c (x,t)xndx=(i)limdn cj(k,t)/dkn (28)

Thus we are able to compute expressions for moj, mj,
and m,. (see Table 4). Similar results were presented
by Killalea and McCoy'41 and can be used to con-
struct the concentration profiles as explained in sec-
tion 1.3. Mathematica can handle readily such alge-
braic calculations.
Isomerization in a Countercurrent
Chromatographic Reactor
Based on the movement in opposite directions of
a sorbent and a fluid, a countercurrent chromatogra-
phic reactor can carry out separation and reaction
simultaneously. This can push to completion a reac-
tion limited by equilibrium. Thus, such a device can
enhance the conversion of the product in reversible
reactions of type A1, <- A2. Experimental investiga-
tion of such a system was provided by Takeuchi, et
al.,1111 who studied xylene isomerization.
The governing equations are given by a special
form of Eq. (24)
dc2
-Ui- -kic+k2c2=0
--ud+klc -k2c2 0(29)


where
u=u-uusK and u2=u-usK2

We have made the following assumptions: (1) steady
state, (2) adsorption equilibrium, (3) no axial disper-
sion, (4) linear adsorption isotherm, (5) isothermal
operation, (6) first-order reversible reactions, and (7)
constant linear velocities.
The solid adsorbent, moving countercurrently to
the fluid at velocity us carries the adsorbates A, and
A2, which have adsorption equilibrium K1 and K'.
We select the feed, sorbent, and species velocities in
such a way that (1) A, is continuously fed at the
bottom of the reactor, (2) sorbent is supplied at the
top of the reactor, (3) A, is less strongly adsorbed
than A2, (4) A1 is moving upward with the fluid and
A2 is transported downward by the sorbent, and (5)
A1 is completely converted in the reactor and the
exiting stream is free of A2. Thus, the boundary
conditions are
Chemical Engineering Education


TABLE 4
Spatial Moments for Moving and Interchanging Isomers

mol=(k2 /k)(f -f2/k+)e-k t (4-1)
mi= Avk2(kl-k2 1k )i1l-e't /k3
-t[(k2 -fik+)(v k +v2k2)e-kt -k2(vk2 +v2k)]/k2 (4-2)
m21=(2k2 /k)(D -D2)[k2(l-f1) k(l+f,)](e-k+t -
-(2k2(Av)2 /k5)[k(1-fl)(e-kt -1)+k (1+2f,)(e-kt-1)+ kk(4-f)( -e-kt)]
+(2t/k)[D1k2 +Dkk(l+fe -kt)+Dkfle t-k2 (-fe- kt(Dk +D2k2)
-(2t(Av)2/k )klk2[kfe-k+t -k21+ 1-fe-k 1)]
-(2t(Av)/k4)(v k,+v,2k2)2kk2f1 +k2(lf)(k2 -k1)]e-kt
3 1 2-kki)]t-
-(t2e-k /k )(fl-k2/k+)(vk+v2k2)2 (t2k2/k3)(lk2 +v2k)2 (4-3)
where
Av=v -v2 (4-4)
k+=k +k2 (4-5)
Tofindm02,m 2, andm 22,interchange subscripts land 2.


TABLE 5
Concentration Profiles for Isomerization in a
Countercurrent Chromatographic Reactor
c1=Ae-r+B (5-1)
c2 = -(u/u2)Ae-" +(k1/k2)B (5-2)
where
A= (cok1u2erL)/(k2ui +klU2erL) (5-3)
B=(ciokiu +c20+u2k)/(kiu2+k2ul) (5-4)
C20+ = (cioklU(1-erL))/(k2u1 +ku2erL) (5-5)
r=kl/ul+k2u2 (5-6)











(30
el(0 )---C10

The concentrations present discontinuities at the
inlet and outlet of the reactor. The flux conservation
equations provide values for these jumps.

c(L+)= c(L -)u/u
(31)
c2(0)-c-C2(0+)U2 /us

Expressions for the concentrations c1 and c2 are
depicted in Table 5. With Mathematica, one can first
find these expressions, then perform numerical simu-
lations, and finally plot the results.


CONCLUSIONS
We have described the computation of expres-
sions of moments using Mathematica for chemical
reaction or separation processes. These computa-
tions are exceedingly tedious to perform and to con-
firm when done by hand. Mathematica is an inter-
esting tool for solving problems with algebraic ma-
nipulations because it is user-friendly, powerful, and
fast. This is particularly true when the software is
run on a Unix machine rather than a smaller, slower
personal computer. It should be clear that Mathe-
matica has limitations, some of which can be
overcome by skillful organizing of the computa-
tional steps. The numerical, graphic, and program-
ming capabilities of Mathematica are exciting po-
tentials that can be applied in multiple areas of
chemical engineering.


NOMENCLATURE

a = interparticle void fraction in the adsorbent bed
A = 3 n/Rp
S= interparticle void fraction (internal porosity) of
the adsorbent
c = concentration of the adsorbable fluid in the
interparticle space
c = input concentration of the adsorbable fluid
c1,c2 = concentration of species A,, A2
ad = concentration of the adsorbed fluid
c. = concentration of the adsorbable fluid in the
pore space


D0'1,
D1,D2


= defined by Eqs. (3-8) and (3-9)
= diffusion or axial dispersion coefficient of


species A1, A2
D = effective intraparticle diffusion coefficient
Winter, 1992


EA
fhf2

h(p)
K,K2


= 1- np/p
= effective axial dispersion coefficient
= rate constant of species A1, A2
= defined by Eq. (3-2)
= function given by Eq. (3-3)
= adsorption equilibrium constant


k,,k2 = rate constant of species A1, A2
KA = adsorption equilibrium constant
k., = adsorption rate constant


kd = desorption rate parameter
kf = mass transfer coefficient in column
k = mass transfer coefficient in the CSTR


k
L

m
n
ni


= surface reaction rate
= length ofchromatographic reactor
= defined by Eq. (3-4)
= nth moment
= nth moment of species A where j = 1 or 2


n = number of particles (grams/vol. of reactor)
0 = ratio of volumetric flow rate to CSTR volume
R = radius of the spherical particle of adsorbent
pp = apparent particle density
s = variable in the Laplace transformation
t = time
tOA = time of duration of the injection of adsorbate
u = superficial fluid velocity
,u2 = velocities of species A,, A2 in the countercur-
rent column
uu = superficial solid velocity
v = linear velocity of the carrier gas in the inter-


vlv2 =
X =
Z =


particle space
velocities of species A,, A,
axial coordinate in the countercurrent column
length coordinate of the bed of adsorbent


REFERENCES

1. Maeder, R., Programming in Mathematica ', Addison-Wesley
Publishing Company (1990)
2. Schneider, P., and J.M. Smith, AIChE J., 14, 762 (1968)
3. Mitchell, R.M., Biopolymers, 15, 1717 (1976)
4. Killalea, M.K., and B.J. McCoy, Biopolymers, 19, 1875 (1980)
5. Kendall, M., and A. Stuart, The Advanced Theory of Statis-
tics, Griffin, London, Chap. 6 (1977)
6. Linek, F., and M.P. Dudukovic, The Chem. Eng. J., 23, 31
(1982)
7. McCoy, B.J., Chem. Eng. Comm., 52, 93 (1987)
8. Mehta, R.V., R.L. Merson, and B.J. McCoy, J. Chromatog-
raphy, 88, 1 (1974)
9. Kubin, M., Collection Czechoslov. Chem. Commun., 30, 1104
(1965)
10. Kucera, E., J. Chromatography, 19, 237 (1965)
11. Takeuchi, K., T. Miyauchi, and Y. Uraguchi, J. Chem. Eng.
Japan, 11, 216 (1978) 0


III










Random Thoughts...





HOW ABOUT A QUICK ONE?


RICHARD M. FIELDER
North Carolina State University
Raleigh, NC 27695-7905

Of all instructional methods, lecturing is the most
common, the easiest, and the least effective.
Unless the instructor is a real spellbinder, most
students cannot stay focused throughout a lecture:
after about ten minutes their attention begins to
drift, first for brief moments and then for longer
intervals; they find it increasingly hard to catch up
on what they missed while their minds were wan-
dering; and eventually they switch the lecture off
altogether like a bad TV show. McKeachie11' cites a
study indicating that immediately after a lecture
students recalled 70% of the information presented
in the first ten minutes and only 20% of that from
the last ten minutes.
There are better ways. Actively involving stu-
dents in learning instead of simply lecturing to them
leads to improved attendance, deeper questioning,
higher grades, and greater lasting interest in the
subject.'112' A problem with active instructional meth-
ods, however, is that they sound time-consuming.
Whenever I describe in workshops and seminars the
proven effectiveness of in-class problem-solving, prob-
lem-formulation, trouble-shooting, or brainstorming
exercises, I can always count on someone in the
third row asking-usually sincerely, sometimes bel-
ligerently-"If I do all that, how am I supposed to
get through the syllabus?"
I have a variety of answers I trot out on such
occasions, depending on my mood and the tone of my
questioner, but they mostly amount to "So what if
you don't?" Syllabi are usually made up from the

Richard M. Felder is Hoechst Celanese Pro-
fessor of Chemical Engineering at North Caro-
lina State University. He received his BChE
from City College of CUNY and his PhD from
Princeton. He has presented courses on chemi-
cal engineering principles, reactor design, pro-
cess optimization, and effective teaching to vari-
ous American and foreign industries and insti-
tutions. He is coauthor of the text Elementary
Principles of Chemical Processes (Wiley, 1986).

18


standpoint of "What do I want to cover?" rather than
the much more pertinent "What do I want the stu-
dents to be able to do?": when the latter approach is
adopted, it often turns out that large chunks of the
syllabus serve little educational purpose and can be
excised with no great loss to anyone. But never mind:
let's accept-for the remainder of this column, at
least-the principle that it is critically important to
get through the syllabus. Can I (asks my friend in
the third row) use any of those allegedly powerful
teaching techniques and still cover it all?
Yes (I reply), you can. Here are two techniques
for doing it.

In-Class Group Problem-Solving

As you lecture on a body of material or go through
a problem solution, instead of just posing questions
to the class as a whole and enduring the subsequent
embarrassing and time-wasting silences, occasion-
ally assign a task and give the class one or two
minutes to work on it in groups of three to five at
their seats. For example:
Sketch and label a flow chart (schematic, force
diagram, differential control volume) for this
system.
Sketch a plot of what the problem solution should
look like.
Give several reasons why you might need or
want to know the solution.
What's the next step?
What's wrong with what I just wrote?
How could I check this solution?
What question do you have about what we just
did?
Suppose I run some measurements in the labo-
ratory or plant and the results don't agree with
the formula I just derived. Think of as many
reasons as you can for the discrepancy.
What variations of this problem might I put on
the next test? (This and the last one are particu-
larly instructive.)
Copyright ChE Division ofASEE 1992
Chemical Engineering Education










You don't have to spend a great deal of time on such
exercises; one or two lasting no more than five min-
utes in a fifty-minute session can provide enough
stimulation to keep the class with you for the entire
period. The syllabus is safe!
Warning, however. The first time you assign group
work, the introverts in the class will hang back and
try to avoid participating. Don't be surprised or dis-
couraged-it's a natural response. Just get their at-
tention-walk over to them if necessary-and re-
mind them good-naturedly that they're supposed to
be working together. When they find out that you
can see them* they'll do it, and by the time you've
done three or four such exercises most of the class
will need no extra prodding. Granted, there may be
a few who continue to hold out, but look at it this
way: in the usual lecture approach, 5% of the stu-
dents (if that many) are actively involved and 95%
are not. If you can do something that reverses those
percentages or comes close to it, you've got a winner.

In-Class Reflection and Question Generation
The one-minute paper is an in-class assignment
in which students nominate the most important and/
or the most confusing points in the lecture just con-
cluded.13,41 Variations of this device can be used to
powerful effect. About two minutes from the end of a
class, ask the students-working individually or in
small groups-to write down and turn in anonymous
responses to one or two of the following questions:

What are the two most important points brought
out in class today (this week, in the chapter we just
finished covering)? Examination of the responses
will let you know immediately whether the students
are getting the essential points. Also, when the stu-
dents know beforehand that this question is coming,
they will tend to watch for the main points as the
class unfolds, with obvious pedagogical benefits.
What were the two muddiest points in today's class
(this week's classes, this section of the course)?
Rank the responses in order of their frequency of
occurrence and in the next class go over the ones
that came up most often.
The responses to this question will surprise you.
What you would have guessed to be the most diffi-
cult concepts may not show up on many papers, if
they show up at all; what will appear are concepts
you take for granted, which you skimmed over in
your lecture but which are unfamiliar and baffling
to the students.
What would make this material clearer to you?
You also never know what you'll get in response to
this one-perhaps requests for worked-out examples

* Students in a class of more than fifteen always imagine they're
invisible.
Winter, 1992


The first time you assign group work,
the introverts in the class will hang back
and try to avoid participating. Don't
be surprised or discouraged-
it's a natural response.

of solution procedures or concrete applications of
abstract material, or pleas for you to write more
clearly on the board, speak more slowly, or stop
some annoying mannerism that you weren't aware
you were doing. Responses to this question can pro-
vide valuable clues about what you could do to
make your teaching more effective.
SMake up a question about an everyday phenom-
enon that could be answered using material pre-
sented in class today (this week). (Optional:) One
or two of your questions will show up on the next
test.
I used the last exercise-including the zinger
about the next test-at the end of a course segment
on convective heat transfer and got back a wonderful
series of questions about such things as why you feel
much colder in 200C water than in 20C air; why you
feel a draft when you stand in front of a closed
window on a cold day; why a fan cools you on a hot
day and why a higher fan speed cools you even more;
why a car windshield fogs up during the winter and
how a defogger works; and why you don't get burned
when you (a) move your hand right next to (but not
quite touching) a pot of boiling water; (b) touch a
very hot object very quickly; (c) walk across hot
coals. I typed up the questions (sneaking a few addi-
tional ones onto the list) and posted them outside my
office-and in the days preceding the test I had a
great time watching the students thinking through
all the questions and speculating on which one I
would put on the test. (I used the one about the fan.)
There are other short, easy, and effective instruc-
tional methods, but these should do for starters.
Check them out and let me know how they work for
you. If I collect some good testimonials (positive or
negative) I'll report them in a future column.

REFERENCES
1. McKeachie, W.J., Teaching Tips, 8th Edn., D.C. Heath &
Co., Lexington, MA (1986)
2. Bonwell, C.C., and J.A. Eison, Active Learning: Creating
Excitement in the Classroom, ASHE-ERIC Higher Educa-
tion Report No. 1, George Washington University, Washing-
ton, DC (1991)
3. Wilson, R.C., "Improving Faculty Teaching: Effective Use of
Student Evaluations and Consultants," J. Higher Ed., 57,
196 (1986)
4. Cross, K.P., and T.A. Angelo, Classroom Assessment Tech-
niques: A Handbook for Faculty, National Center for Re-
search to Improve Postsecondary Teaching and Learning,
Ann Arbor, MI (1988) O










classroom


INTRODUCING


WATER-TREATMENT SUBJECTS INTO

CHEMICAL ENGINEERING EDUCATION


L.CACERES, E. GRUTTNER, V. MONARDES, R.
CONTRERAS, B. GOMEZ-SILVA
Universidad de Antofagasta
Casilla 170Antofagasta, Chile

traditionally, water-treatment subjects are cov-
ered extensively only in civil engineering, even
though several aspects can be analyzed without dif-
ficulty under the light of current chemical engineer-
ing theoretical tools. Severe damage to the environ-
ment caused by discarded wastewater, alternatives
for treatment, and theoretical considerations can all
be efficiently introduced in courses such as the intro-
duction to chemical engineering, transport phenom-
ena, reactor design, and unit operations.
Including water subjects in engineering educa-
tion is particularly important in arid regions of the
world where water is a limiting factor for industrial
and urban development. In Northern Chile, a region
of extreme aridity,'1121 every industrial activity is af-
fected by a limited supply of water, and chemical
engineers are often faced with industrial processes
susceptible to be upgraded, e.g., water recycling. Also,
the trend of ever-increasing environmental regula-
tions in industry justifies the need for introducing

Luis Caceres V. is an assistant professor of chemical engineering at the
Universidad de Antofagasta (Chile), and he is a researcher at the Institute
del Desterto. He received his Master's Degree in 1984 from the University of
Toronto (Canada).
Erika Gruttner D. is an associate professor of chemical engineering at the
Universidad de Antofagasta (Chile), and is a researcher at the Institute del
Desierto. She finished her studies in chemical engineering at the Universidad
Tecnica del Estado, Santiago, Chile, in 1978.
Vinka Monardes V. received her Bachelor's Degree in 1989 from the
chemical engineering department at the Universidad de Antofagasta (Chile),
and since then she has been actively involved in research at the Institute del
Desierto.
Ren Contreras N. s an associate professor at the Universidad de An-
tofagasta (Chile) and is a researcher at the Institute del Desierto. He
received his PhD from the University of Perpignan (France) in 1984.
Benito Gdmez-Silva is a biochemist, associate professor, and director of
the Instituto del Desierto, Universidad de Antofagasta. He received his PhD
degree in photobiology at Brandeis University in 1985.


Including water subjects in engineering education
is particularly important in arid regions of the
world where water is a limiting factor for
industrial and urban development.

wastewater treatment subjects into the chemical en-
gineering curriculum.
Most educators agree that environmental-related
issues need to be incorporated into the chemical
engineering curriculum, but there is no agree-
ment on the procedure for doing so. To achieve
this, some U.S. universities have considered the fol-
lowing options:
including some environmental-related issues in the
capstone senior design course
consideration of elective courses containing environ-
mental-related issues
coordinating environmental-related-issue training
through examples and homework problems in the
core chemical engineering coursesl"
The chemical engineering department at the
Universidad de Antofagasta has recently incorpo-
rated environmental-related water issues into the
curriculum with good results and without a signifi-
cant increase in the students' academic load. The
aim of this activity is: 1) to promote a working knowl-
edge of both the basic indexes used in the descrip-
tion of water quality and water treatment technolo-
gies, 2) to advance an understanding of the need to
rid the water of microorganisms and chemicals, and
3) to enhance the students' perception of water and
its relationship to environmental issues. Our ap-
proach exposes the students to exercises with incom-
patible or incomplete sets of data in order to encour-
age discussion and creativity. This non-traditional
approach is a well-recognized practice that has the
effect of improving the student's insight into chemi-
cal engineering processes.'41 We also introduce key
water-treatment concepts through exercises and dis-
Copyright ChE Division ofASEE 1992
Chemical Engineering Education










cussions, and we emphasize both water contamina-
tion problems and water reclamation during routine
visits to local mines and factories.
Water recycling and wastewater reclamation are
two environmental-related issues discussed in classes
at the Universidad de Antofagasta, and they are
described in detail below.
EXAMPLE 1
Water Recycling in a Copper Concentrating Plant
During the course Introduction to Chemical En-
gineering, a once-a-year visit to the nearby Mantos
Blancos mine is planned. Figure 1 shows the recy-
cling water system used there.E51 Chemical engineer-
ing students are asked to determine all the informa-
tion necessary to complete Table 1 by interviewing
supervisors and operators at the plant. Both the
table and the figure are given to the student in
advance of the plant trip. They must also inquire
about the details of the process strategies which
minimize water content in the band filter cake.
The final field report should contain a discussion


Data Compatibility: in order to find out how
connected all the information collected during
field trip is, a mass balance and the correspol
degree of freedom analysis must be performed
Water recycling: From the analysis above and
the fresh-water cost, the importance of maxin
cake dewaterability in the filter band must be
deduced.


COPPER
COENCET ATE

ORE





1) Cog pF




DUMPII NG
SI TE



Figure 1. Diagram flow of the recycling water
system used at Mantos Blancos copper mine, II
Region, Chile. 1) Concentrating plant.
2) Hydrocyclone. 3) Thickener. 4) Mixing tank.
5) Band filter. 6) Conveyor belt. WW) Discarded
tailings. FF) Fine fraction. CF) Coarse fraction.
S) Sludge. M) Sludge. BF) Band filter. F) Filter
cake. R) Recycled water.
Winter, 1992


well-
g the
ending
d[61
given
Rum


As a result of these exercises and
discussions, the students achieve a
straightforward understanding of the basic
operating principles of wastewater
stabilization ponds.

At this point students should "translate" the in-
formation given by the plant personnel, i.e., the use
of percentage solid index, over or under 200 mesh,
when standards for cake dewaterability are estab-
lished. Thus, compounding elements of mass bal-
ance, integral process control, and concern for water
are all involved in this example. Some field reports
contain interesting suggestions on the processes stud-
ies, e.g., experimental measurements of cake dewa-
terability under different particle size distribution.
Water recycling is also partially reviewed in
courses such as unit operations, where bench scale
filtration tests are computed.
EXAMPLE
Water Treatment in Stabilization Ponds
Theoretical aspects of water treatment can be
discussed in a single exercise in the reactor design
course.
Wastewater stabilization ponds are open lagoons
where aerobic bacterial and microalgae growth and
metabolism transform organic matter into stable com-
pounds and pathogenic microorganisms are gradu-
ally eliminated.171 Water can then be safely reused,
either for agriculture or industrial applications, e.g.,


TABLE 1
Process Flow Values at Mantos Blancos Concentration Plant

Inflows) and Effluent(s)
Equipment Purpose % solids* and % solids* Remarks

Hydrocyclone Fast slurry class- WW = 1 FF= Particle size distri-
ification into fine CF= bution in WW can
and coarse Xww = 15-40% Xff= be partially modi-
fractions Xcf= fied, if necessary
Thickener Separation of FF= S=
settleable BF= R=
particles Xff= Xs=
Xbf= Xr=
Mixing tank Mixing thickener S= M=
and hydrocyclone CF=
underflows Xs= Xm=
Xcf=
Filter band Solid separation M= BF=
by filtration as FC=
to obtain cake Xbf=
with high solid Xf=62-64% Xfc
content

* Blank spaces must be filled in by students.









ore leaching. Therefore, municipal wastewater can
be treated by this relatively inexpensive method in
places where both low cost land and high solar irra-
diation are available. 81
Since continuous reactors and stabilization ponds
operate similarly, the latter are also known as
bioreactors. Thus, the same efficiency concept and
equations studied during the reactor design course
can be applied to bioreactors.
Organic matter in raw wastewater is a complex
mixture of different chemicals and microorganisms.
Therefore, it is necessary to find a single variable or
index in order to use a single kinetic expression
properly, either for total biodegradable organic mat-
ter or microorganisms population. Examples of these
indexes are Total Coliform Bacteria per 100 ml of
water (TCB) and Biochemical Oxygen Demand
(BOD). The latter indicates the total amount of oxy-
gen required for the biological breakdown of carbon-
and nitrogen-containing material, i.e., the amount of
oxygen required to sustain biological activity in an
aerobic pond.191
To better expose the students to wastewater treat-
ment through wastewater stabilization ponds, the
following topics are emphasized:
a. Due to the complexity of the process, a realistic model
must consider that organic matter bioconversion and
microorganism deaths are both linearly dependent on
sunlight intensity, and also that the kinetics of the
process can be expressed as -dA/dt = k*A*I, where A
represents BOD or TCB, I is sunlight intensity, and k is
the temperature-dependent kinetic constant.
b. Since solar energy is absorbed by water, a vertical
temperature profile is reached in the pond. Therefore,
there should be a discussion on how BOD or TCB is
affected by such a temperature profile (assuming that
k has an Arrhenius-type temperature dependency) but
the kinetic expression is not affected by sunlight inten-
sity.
c. An analysis should be made on the assumptions nec-
essary for obtaining a single k value at any place in
the wastewater stabilization pond.
d. If Fick's equation, with analytical solution, is to be
used to account for the overall efficiency, the implicit
assumptions involved in Fick's equation must be ex-
plained. Also, the student should try to answer the
question of what happens with Fick's equation when
a completely mixed or plug reactor is considered.
e. Finally, the class discusses the beneficial uses of water
reclamation in Northern Chile.
Temperature, pH, and absorbance measurements
at different depths and times of the day demonstrate
the complexity of the biological processes occurring
in a wastewater stabilization pond, and therefore
its modeling. If time and depth are not considered


during the measurements, a simpler model can be
reached.
In topics a and b above, the discussion is cen-
tered on the need to work out mass and heat trans-
fer equations simultaneously, including velocity
terms and using the assumptions required. Further-
more, the proposed analytical form for dA/dt can be
questioned since organic matter can also be degraded
by anaerobic bacterial activity and TCB can also
decrease by changes in pH. A discussion on the clas-
sical dispersion model for a non-ideal reactor is in-
volved in topics c and d, but using the wastewater-
stabilization-pond terminology.
As a result of these exercises and discussions, the
students achieve a straightforward understanding
of the basic operating principles of wastewater stabi-
lization ponds. In addition, they are able to practice
key kinetic reactor concepts with a minimum addi-
tional academic load. Example 2 is complemented by
a field trip to the Universidad de Antofagasta's waste-
water stabilization pond pilot plant. Finally, a simi-
lar discussion can be carried out on activated sludge
as an alternative process where oxygen concentra-
tion replaces sunlight intensity in the analysis.

CONCLUSION
Water treatment subjects can be successfully in-
troduced in the core chemical engineering courses as
practical examples of standard concepts. It requires
a coordinated effort in several courses in order to
discuss similar wastewater aspects through various
approaches.
The students can obtain a critical view of water
treatment processes for recycling purposes through
field trips, where real processes are seen in action.
Introducing wastewater treatment subjects in
the chemical engineering curriculum provides direct
access to biotechnological and ethical issues. The
students are exposed to environmentally related is-
sues and treatment alternatives which require an
interdisciplinary approach in both industrial and
academic research.

ACKNOWLEDGEMENT
Part of this work was funded by the United Na-
tions Development Program project CHI-87-024.

REFERENCES
1. Geografia de Chile. Tomo II Regi6n, Edited by Instituto
GeogrAfico Militar, 1st Edition, Santiago, 221 (1990)
2. Klohn, W., "Hidrografia de las Zonas Des6rticas de Chile,"
J. Burz, Ed., United Nations Development Program,
Santiago, 188 (1972)
3. Lane, A.M., "Incorporating Health, Safety, Environmental,
Chemical Engineering Education










and Ethical Issues into the Curriculum," Chem. Eng. Ed.,
23(2), 70 (1989)
4. Felder, R.M., "The Generic Quiz: A Device to Stimulate
Creativity and Higher-Level Thinking Skills," Chem. Eng.
Ed., 19(4), 176 (1985)
5. Narvaez, A., and R. Contreras, "Reciclaje de Agua en Planta
Concentradora de Cobre," Innovacidn, 2(1), 2 (1989)
6. Reklaitis, G.V., Introduction to Material and Energy Bal-
ances, John Wiley & Sons, New York (1983)
7. Mara, D., and H.W. Pearson, WHO ERO, "Waste Stabiliza-
tion Ponds," EUR/ICP/CWS 053 7384V (1987)
8. Arthur, J.P., "Notes on the Design and Operation of Waste
Stabilization Ponds in Warm Climates of Developing Coun-
tries," World Bank Technical Paper No. 7 (1982)
9. "Wastewater Stabilization Ponds: Principles of Planning and
Practice," WHO EMRO Technical Publication No. 10 (1987)
O


letter to the editor

LANGMUIR'S ISOTHERM:
Kinetics or Thermodynamics?

Dear Sir:
The Langmuir isotherm for adsorption equilibrium of
m solutes is of the form


a.c.
J m
1+ b.ici
i=l


where the q, and c, are the concentrations of the solutes i
on the adsorbent and in the fluid phase, respectively, and
the a. and bi are constant coefficients. Langmuir' 1 derived
his equation in 1916 with a kinetic argument: that the
rates of adsorption and desorption must be equal at equi-
librium.
To this day almost every textbook follows his reason-
ing, even though it does not stand up too well to close
scrutiny. It postulates the adsorption rate to be propor-
tional to the concentration of the solute in the fluid phase
and the unoccupied surface area; and the desorption rate
to the amount adsorbed.
While this is plausible, there is no proof. Also, in real-
ity, adsorption and desorption rates often are controlled
by mass transfer and so may obey different laws. The
derivation should therefore retreat to declaring its rates
as only those of attachment to and detachment from the
surface,121 and even then the rate law may differ if attach-
ment and detachment were to involve more than a single
step.
More seriously, where Langmuir's equation is not
obeyed, do we really believe the deviation is caused by a
kinetic anomaly? Lastly, have we not warned our students
on other occasions that intermingling kinetic and equilib-
rium arguments is fraught with pitfalls?
Although its original derivation is shaky, the Langmuir
isotherm has proved to be more successful than any other
of comparable simplicity. So why not show our students a
derivation that is more convincing and free of merely
Winter, 1992


plausible assumptions?
The isotherm describes adsorption equilibrium, and
the finest tool we have for equilibria is thermodynamics.
We used thermodynamics to derive the mass-action law.
To recapitulate: The Gibbs free energy of a (closed) system
is a function only of temperature, pressure, and conver-
sion and is at its minimum at equilibrium; with m partici-
pants, stoichiometric coefficients vi, and at constant tem-
perature and pressure


dG= -G dn =0 (2)
i=1 ani
at equilibrium, where n; is the number of moles of species
i. Introducing the chemical potential gp and its concentra-
tion dependence
3G.
-n. i =L +RT in ci

provided the system is ideal (activities equal concentra-
tions). The reaction stoichiometry requires
dn. v.
so that
m m m

i=l i=l i=l
at equilibrium. Since the term involving the standard
potential, p2, is constant

m
icv --K=const. (3)
i=1 '
at equilibrium. This is the mass-action law, with K as the
equilibrium constant.
Nowhere in this derivation was it necessary to assume
chemical bonds to be formed or broken, or even to identify
the participants as molecules or chemical species. The
derivation covers any kind of process in which something
is changed into something thermodynamically distinguish-
able. Its result can therefore be applied to adsorption as
well, and it yields the Langmuir isotherm.
An easy way of showing the long-known mathematical
equivalence of the mass-action law and Langmuir iso-
therm is as follows. For the reversible "reaction" of a
solute molecule j with free adsorbent surface to form occu-
pied surface, the mass-action law gives
80
j =Kj =const. (4)
Ofreecj
where the free and occupied surface are expressed as
fractions 6 of total surface. Also


m
free = 1- .
i=l


From Eq. (4)


Continued on page 51.


0. K.c.
0i Kici
8. K.c.
J Jj


(j= 1'...,m)










I outreach


INTRODUCING HIGH SCHOOL

STUDENTS AND SCIENCE TEACHERS

TO CHEMICAL ENGINEERING1


TARYN MELKUS BAYLES,2
FERNANDO J. AGUIRRE
University ofNevada, Reno
Reno, NV 89557

Engineers and scientists will be key professionals
in the future development and implementation
of new technology programs that will serve to keep
our country in its leadership role. Recent enrollment
trends in engineering programs11l show a decrease in
the number of high school students who are inter-
ested in technical careers, and many who are inter-
ested do not have an adequate background for com-
pletion of an engineering program because they do
not take advanced science and math courses while in
high school.
Following the national trend, the Mackay School
of Mines and the Chemical and Metallurgical Engi-
neering Department at the University of Nevada,
Reno, have also experienced significant enrollment
reductions for the past few years. Figure 1 depicts
the number of freshmen and total enrollment for the

i Freshmen .- Total


Number of students


80 81 82 83 84 85
YEAR


8. 87 88 89 90
86 87 88 89 90


FIGURE 1. ChE enrollment at the University of
Nevada, Reno
SPaper presented at AIChE Annual Meeting, Chicago, IL, 1990
2 currently with Westinghouse Electric Corporation, Resources
Energy Systems Division, Pittsburgh, PA
24


Taryn Melkus Bayles is currently a Senior En-
gineer with Westinghouse Electric Corporation.
She was formerly an assistantprofessor ofchemi-
cal engineering at the University of Nevada,
Reno, where she was active with the recruiting
and scholarship programs. She received her
BSChE from New Mexico State University, and
her MSChE, MSPetE and PhD in chemical engi-
neering from the University of Pittsburgh


Fernando J. Aguirre is an associate professor
of chemical engineering at the University of Ne-
vada, Reno. He received his MS and PhD in
chemical engineering from the University of Pitts-
burgh and worked for several years as a re-
search engineer for Bethlehem Steel Corpora-
tion. His major research interests are mathemati-
cal modeling, simulation, neural networks, and
computer applications.

last ten years. After reaching a low of one freshman
in 1986 and a total of twenty-four in 1987, we began
an active recruiting program which involved visiting
high schools and telling students about careers in
chemical engineering. We visited high schools in
Reno, Las Vegas, northern Nevada, and neighboring
counties in California. These visits proved to be very
useful, with the result that students are now more
aware of careers in engineering and that teachers
are now better informed and prepared to advise stu-
dents who have technical interests.
As a complement to the active recruiting pro-
gram, a "Summer Institute" was developed. Again,
the primary purpose was to introduce high school
students and teachers to chemical engineering. The
approach here, however, was through hands-on ex-
perimentation and exercises which were conducted
with the help of university students, a more involved
step than the recruiting seminars.
Active participation was the vehicle through
which students could discover the challenges, the
usefulness, and even the fun that is associated with
this scientific discipline. The students also acquired
a better feel for what types of preparatory math and
Copyright ChE Division ofASEE 1992
Chemical Engineering Education









science courses they would need to pursue an educa-
tion in chemical engineering.
The program which was developed for this sum-
mer instituted included an introduction to chemical
engineering, a description of typical chemical engi-
neering jobs, and general instruction in material
balances, fluid mechanics, heat transfer, mass trans-
fer, and process control. Of course, these topics could
only be covered superficially in such a short time,
but it was enough to provide some insight into the
problems that chemical engineers solve and the cor-
responding scientific background and preparation
that is required.
The main goal of the program was to increase
enrollment in engineering and to encourage women
and minority groups to increase their representation
in the engineering workforce.

STUDENT SELECTION
The science teachers in several local high schools
(within a 75-mile radius of Reno) were invited to
attend the summer institute, and each of the teach-
ers was asked to select three or four of their better
students who had an interest in math and science to
also attend the institute. They were also asked to
encourage females and minorities to participate.
Five high schools responded to the invitation,
with a total of five teachers and nineteen students
participating. The student group consisted of nine
women and ten men, with four of them classified as
minorities. Eleven of the students had just com-
pleted the freshman year, seven had just completed
the sophomore year, and one had completed the jun-
ior year. Therefore, most of the students had two
remaining years of high school in which to prepare
themselves in math and science in the event they
wished to pursue an engineering degree.

PROGRAM ORGANIZATION AND SCHEDULE

The Summer Institute in Chemical Engineering
was one week in duration, with sessions lasting from
9:00 a.m. until 4:00 p.m. each day. The overall sched-
ule of activities for the week is presented in Table 1.
The program began on Monday morning with a gen-
eral introduction and slide presentation about chemi-
cal engineering, attended by all students. For the
remainder of the Monday session, the group of
twenty-four students was divided into three groups
of eight students each.
For the remaining days (Tuesday through Fri-
day) the students were divided into two groups, "A"
and "B," each of which was divided into two sub-

Winter, 1992


The program ... included an
introduction to chemical engineering,
a description of typical chemical engineering
jobs, and general instruction in material
balances, fluid mechanics, heat
transfer, mass transfer,
and process control.



TABLE 1
General Schedule for Summer Institute
DAY TOPIC
Monday Introduction
Careers in chemical engineering
(videotapes and personal experiences)
Material balances (lecture, problems,
and computer exercises)
Tuesday Fluid mechanics session: Group A
Mass transfer session: Group B
Wednesday Mass transfer session: Group A
Fluid mechanics session: Group B
Thursday Heat transfer session: Group A
Process control session: Group B
Friday Process control session: Group A
Heat transfer session: Group B

groups of five to seven students each. Dividing the
students into smaller groups provided for a more
personal level of participation and understanding by
all students involved, which fulfilled the course cri-
teria of active participation in experiments rather
than simply listening to a lecture or watching a
demonstration.
Each group of approximately twelve students
(groups A and B) had a common one-hour lecture in
their first morning session. This was necessary in
order to provide background information and to in-
troduce the students to the subject. Some of the
activities and experiments scheduled for the day
were also discussed in this session, and explanations
about the phenomena they would be observing or
measuring were provided.
After the morning lecture the groups separated
and participated in activities such as laboratory ex-
periments, calculations to analyze experimental data,
and computer exercises. Each of the laboratory ses-
sions was directed by a teaching assistant, who pro-
vided guidance to the students. A rotation procedure
was developed so that each of the smaller groups
had an opportunity to participate in all laboratory
sessions. A list of activities for each daily session
(fluid mechanics, heat transfer, mass transfer, and









process control) is presented in Table 2.
Lunch and refreshments were provided for all
participants. These periods provided a more infor-
mal atmosphere in which the students could inter-
act with the instructors and the teaching assistants,
as well as make friends with each other.
A comprehensive notebook was prepared and dis-
tributed to each of the student participants. It con-
tained all of the lecture notes, a description and
explanation of all the experiments and demonstra-
tions, and worksheets for performing calculations
and other exercises. This material enabled the stu-
dents to follow along more easily, and hopefully it
will serve as a reference for them in the future. It
should also be a useful guide for science teachers
who might later integrate some of the material into
their teaching.
The Summer Institute was taught by two profes-
sors of chemical engineering (the authors) with the
help of four teaching assistants who were seniors or
graduate students in the Chemical and Metallurgi-
cal Engineering department.

DESCRIPTION OF TOPICS
A list of the topics covered in the Summer Insti-
tute, with a brief description of their content, fol-
lows.

Introductory Topics
Material Balances This topic introduced the
students to chemical engineering. They learned how
to formulate and solve material balance problems
which are the basis in the design of process plants.
The topic was covered in three different sessions:
introductory lecture, solution of problems, and dem-
onstration of applications using a microcomputer. In
this last session the students used the CAAPS (Com-
puter Aided Analysis for Process Systems) softwareE2'
which is a linked system of menu-driven, compiled
BASIC programs for the elementary steady-state
analysis of chemical processes. The software was
demonstrated for balancing chemical reactions, con-
verting units, and performing some simple steady-
state material balances.
Biotechnology A ten-minute videotape which
explores the variety of careers open to chemical en-
gineers in the field of biotechnology was shown.
Advanced Materials A fifteen-minute video-
tape which explores the variety of careers open to
chemical engineers in the field of advanced materi-
als was shown.


TABLE 2
List of Activities
Fluid Mechanics Introductory lecture
Reynolds number experiment
Reynolds number calculations
Pressure drop experiment
Pressure drop calculations
Computer design of slurry pipelines

Heat Transfer Introductory lecture
Temperature profiles in solid rods
(experiment)
Temperature profiles in solid rods
(calculations)
Double-pipe heat exchanger experiment
Double-pipe heat exchanger calcula-
tions
NASA space shuttle tile demonstration

Mass Transfer Introductory lecture
Diffusion experiment
Diffusion calculations
Distillation column experiment
Distillation column calculations
Flotation demonstration

Process Control Introductory lecture
Valve characterization experiment
Valve characterization calculations
Liquid level control experiment
Liquid level control calculations
Computer simulation of level control



Environmental Protection A twenty-five min-
ute videotape exploring the variety of careers open
to chemical engineers in this field was shown. It
described many of the current environmental prob-
lems and the role chemical engineers have in their
solution.
Chemical Engineering Job Experience A de-
scription of typical job assignments given to chemi-
cal engineers was presented. The two instructors
have had prior work experience in the petroleum
refining and steel industries, and therefore a pres-
entation of each of those industries was made, in-
cluding a description of the petroleum refining proc-
ess and the steelmaking and cokemaking processes.
In addition, each instructor presented examples of
his/her own job assignments while employed in those
industries.


Chemical Engineering Education









Fluid Mechanics Topics
Reynold's Number Experiment Visual obser-
vations of various types of flow laminarr, turbulent,
and transition) were made by allowing water to flow
through various sized and shaped conduits, and in-
jecting a small stream of dye. These visual observa-
tions were compared to predicted values via calcula-
tion of the Reynold's number.
Pressure Drop in Pipes and Fittings Labora-
tory experiments were performed to measure the
frictional losses of water flowing through various
sized pipes, valves, and fitting. These frictional losses
were compared to theoretically-predicted pressure
drops.
Computer Design of Slurry Pipelines Per-
sonal computers were used to demonstrate a pro-
gram that is used for designing slurry pipelines.131
Use of the computer enabled students to investigate
problems associated with slurry pipeline design
through some specific design problems.

Heat Transfer Topics
Double-Pipe Heat Exchanger Three steam-
jacketed, copper, circular tubes of different diame-
ters were used to demonstrate the operation of
double-pipe heat exchangers. Water was passed
through the inner tube, and steam was passed
through the annular space. The heat transfer coeffi-
cients, both as a function of flow rate and tube di-
ameter, were determined for comparison with litera-
ture values.
Temperature Profiles in Solid Rods Three
cylindrical rods of varying diameters and construc-
tion materials were heated at one end. The tempera-
tures along the rods were measured to determine
the temperature distribution of each rod. These re-
sults were then compared to theoretical predictions.
The use of a thermocouple as a temperature-meas-
uring device was also demonstrated.
NASA Space Shuttle Tile Demonstration An
actual NASA space-shuttle tile was used to demon-
strate the thermal insulating power of modern ce-
ramic materials. The tile was heated to 900C in a
furnace, and then, due to the thermal conductivity of
the material, the students found they were able to
hold it without burning their hand.

Mass Transfer Topics
Diffusion Experiment The diffusion coefficient
of acetone in air was determined at various tempera-
tures. Acetone was placed in a graduated cylinder
Winter, 1992


and maintained at a constant temperature while air
was blown through the top. The rate of diffusion was
determined by measuring the change in acetone liq-
uid level as a function of time. The experimental
results were compared to theoretical predictions.
Distillation Column Operation A pilot-plant
size distillation column was operated to demonstrate
the principles underlying separation of two liquid
components in solution. Liquid samples and tem-
perature measurements were taken during steady-
state operation while separating an acetone-water
solution. The students were taught how to plot equi-
librium data and to calculate the number of stages
required for a given separation by using the McCabe-
Thiele method. They used this information to com-
pare with their experimental measurements and to
calculate the overall efficiency of the distillation col-
umn.
Flotation Demonstration A Denver labora-
tory cell was used to demonstrate the underlying
principles of flotation. This process uses differences
in the surface properties of particles in an aqueous
pulp to affect a separation. Hydrophobic particles
are floated to the surface by finely dispersed air
bubbles and are collected as froth concentrate. Hydro-
philic particles do not adhere to the air bubbles, but
remain in suspension in the pulp and are carried off
as underflow.

Process Control Topics
Valve Characterization The valve-sizing coef-
ficient of a miniature control valve was determined
as a function of valve-diaphragm pressure. The stu-
dents also determined whether the control valve had
linear or equal percentage inherent characteristics.
The procedures used for specifying control valves in
practical applications were also discussed.
Automatic Control ofLiquid Level An analog
controller was used to maintain the liquid level in a
tank at a desired value. The principles of automatic
process control were described, and the student was
exposed to the actual instrumentation used in the
field. Performance of the level controller was evalu-
ated for various operating conditions. The problems
which can arise by improper design or through selec-
tion of the wrong settings was also demonstrated.
Computer Simulation of Level Control A
computer program that simulates the operation of
three non-interacting tanks in series was used to
demonstrate some of the principles of process con-
trol. The program runs on a microcomputer and
graphically displays the level of one of the tanks.









This level can be monitored on-line as the user
changes various controller parameters. The program
runs interactively and the student can become the
plant operator who has the capability of making
changes to the process. The student can open or
close the control valve, select manual or automatic
operation, change the setpoint, or change controller
settings. As a result, he or she can observe the re-
sults of any given change immediately on the screen.

COURSE EVALUATION
At the end of the week the students were asked
to complete a course evaluation to help us improve
the course in future years. The questions included in
the form are shown in Table 3.
The response to the Summer Institute was very
favorable; many of the students and teachers indi-
cated that they would be interested in attending the
next session also. The students selected the com-
puter sessions and the laboratory experiments (which
involved more participation) as their favorites. The
videotapes and some of the calculation sessions
seemed to be less attractive. However, in response to
the question about the session they liked the least,
the answer that students repeated the most was
"none." Several students indicated that they would
include "all" sessions as those they liked the most,
and one student wrote, "I liked almost all of them
the same. They put into perspective the thoughts
installed into our brains in the morning." The diffi-
culty level of the material covered was considered
appropriate by 74% of the students, while the re-
maining 26% thought it was somewhat high. One
student commented, "A few early-morning sessions
were hard to understand at first but were cleared up
by the afternoon. Having daily periods build on each
other was very effective."
The majority of the students (62%) indicated that
the time spent on lectures should remain the same,
and 33% of them would like this time reduced. The
opposite occurred for the time spent in the labora-
tory doing experiments or computer work: 62% sug-
gested an increase and 33% no change. More mixed
results were obtained for the time spent doing calcu-
lations: 52% suggested no change, 38% indicated
that the time should be decreased, and 10% sug-
gested an increase. The notebook that was prepared
received very favorable comments, as did the in-
structors and teaching assistants. One student said
about the notebook, "It was nice to have notes com-
pleted so we could look back and read what we
missed. It will be nice to keep and look back at."
Another student wrote, "I plan to study this book


TABLE 3
Questions on the Course-Evaluation Form
1. Has this course been of help in understanding something
about chemical engineering?
2. What sessions) did you like the most? Comments.
3. What sessions) did you like the least? Comments.
4. Was the overall material covered and discussed
too difficult?
too easy?
about right?
Circle your answer and provide any comments. If any
specific session was too difficult or too easy, please
indicate below.
5. How do you feel about the time spent on each activity?
Mark your suggestions and provide any comments.
Increase Decrease No Change
Time spent on lectures
Time spent in laboratory
Time spent doing calculations
6. Did you find the notes useful and clear? Comments.
7. What would you suggest changing to make the course
better?

and use it for future reference."
The students made several suggestions on how
to improve the course. Their suggestions ranged from
making the course and/or days shorter to making it
a four-week course. Even though the course empha-
sized hands-on laboratory experimentation by the
students, they would like to see more experiments in
a variety of areas. One general suggestion was that
students should have had a course in chemistry be-
fore attending the institute since some of those who
had not been exposed to chemistry had trouble under-
standing the material.
In summary, the Summer Institute in Chemical
Engineering proved to be a very enjoyable and use-
ful experience for everyone involved.

ACKNOWLEDGEMENT
The financial support of the University of Ne-
vada System Chancellor's Office and Nevada Mining
Association is gratefully acknowledged. We would
also like to thank Thomas Lugaski and the teaching
assistants Timothy Burchett, David Castillo, Bain-
ian Liu, and Carl Nesbitt, for their help.

REFERENCES
1. Donaldson, T., "Chemical Engineering Enrollments,"
AIChExtra, pg. 2, September (1990)
2. Cadman, T.W., CAPPS: Computer Aided Analysis for Proc-
ess Systems, User's Guide, ENSCI, Inc. (1988)
3. Provine, W., B. Freeman, G. Dow, and M. Denn, "Design of
a Slurry Pipeline," CACHE IBM PC Lessons for Chemical
Engineering Courses Other Than Design and Control O
Chemical Engineering Education










LETTER: Equilibrium Thermodynamics
Continued from page 7.

1984 to write
dH T if dP = 0 (10)
(dP = 0 does not have any special attachment to the ratio
that precedes it.) It is extremely useful, but I have seen it
nowhere else. dH and dT are differentials, smallness not
being implied.
Williams and Glasser give the following definition of
an integral:

STSVdS= im TS(Si,V(Si))ASi (11)
S, i=1
N-)-
(A missing comma has been added on the right.) I believe
the definition is faulty. The main point about the proposed
notation is that the superscript variables are independent.
In DUsV/dS, the variable V has nothing to do with the
differentiation as such. One can, of course, define a new
function Us by
S= Us (S, VS) (12)
for some function (or "path") Vs and then obtain the de-
rivative 3Us/dS. Because
TSV (13)
aS (13)
one logically writes the corresponding anti-derivative or
integral as
STSvdS=US +BV (14)
and
s2
ITSVdS=USV(S2,V)-USV(S1,V) (15)

The latter is a function of the single variable V, say I. It is
clear that the definition should in fact be

TsvdS= lim -TSV(Si,V)ASi (16)
S, A -->o i=1
N->-
The definition Williams and Glasser give is the definition
for the constant
82
JTSdS (17)
S1
where
TS TSV(SVS) (18)
again for a particular function (or "path") Vs. Of course,
the integral then depends on the function VS and is, there-
fore, function (or "path") dependent. (It is commonly termed
a functional.) The integral of Eq. (16), on the other hand, is
variable dependent, not function dependent. The descrip-
tion "path-dependent function" is clearly ill-advised. Both
of the integrals given by Eqs. (16) and (17) are useful.
Incidentally, expressions of the form
Winter, 1992


Q= TdS (19)
are justifiable in this context only as abbreviations for
integrals like those above.
In their treatment of my approach to conservation of
energy, Williams and Glasser employ what I think is a
little unnecessary circularity. They use the concept "well-
insulated enclosure" to discover the concept commonly
called "heat." But it is not clear what a "well-insulated
enclosure" means when one has yet to meet "heat." My
approach is to argue that experiments are found to fall
into two classes: those for which
AU= w (20)
holds, and those for which it does not. The former are
called anything we like-say, adiabatic (but it could as
well be well-insulated for that matter)-or are said to be
surrounded by an adiabatic wall; the latter, non-adiabatic,
or surrounded by a diathermal wall. At this stage of the
argument these terms mean nothing except that Eq. (20)
is or is not obeyed. (It is easy to add terms on the left
where necessary to account for kinetic and potential en-
ergy.) Now one is ready to invent a new quantity which
can be given the symbol q; it is merely the quantity that
allows Eq. (20) to be modified so that it always holds. The
modified equation (and definition ofq) is
AU = q + w (21)
The confused thinking one encounters elsewhere over
"work" and "heat" is quite remarkable. Suggestions that
the gerunds "working" and "heating" be used instead'2' do
not seem helpful. I use "mechanical" and "thermal trans-
fer of energy." I also quote Callen's'[3 analogy of the pond
and the modes of transfer to it of water by rain and by
stream which, when I encountered it, gave me just the sort
of aha!-insight into thermodynamics that I had badly
needed.
I borrowed (and modified) the postulates from Callen
that now appear in the latter half of Williams and Glasser.12'
Much of what follows there is also based on Callen. But
there are other sources and influences; for example, the
early ideas of researchers such as Georgian'4'51 with whom
I had corresponded about the units of temperature and the
"universal gas constant" R.
Williams and Glasser describe what Callen calls the
basic problem of thermodynamics and use his method of
solving it when the internal adiabatic constraint is re-
moved: one maximizes entropy over all constrained states.
At the end of Part 2 they suggest that the student try
three ostensibly similar problems in which other internal
constraints are removed. Their first problem requires re-
moval of an internal rigidity constraint and no other. It
turns out, however, that the method does not work in this
case (as it is not difficult to show). Solution by maximiza-
tion of entropy always assumes removal of the adiabatic
constraint at least. Earlier in Part 2, Williams and Glasser
raise the practical difficulty of how to relax the permeabil-
ity constraint without relaxing the adiabatic constraint as
well. The answer is that one does not.
Continued on page 37.









class and home problems


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



MORE APPLIED MATH PROBLEMS ON

VESSEL DRAINING


JUDE T. SOMMERFELD
Georgia Institute of Technology
Atlanta, GA 30332


P processing and storage vessels in the chemical
and allied industries come in a large variety of
shapes. There are almost as many reasons for this
variability as there are shapes; these reasons can
include convenience, insulation requirements, land
and material costs, safety considerations, tradition,
etc. Drainage of such vessels through an orifice-type
hole at the bottom of the vessel represents a class of
non-linear, ordinary, first-order differential equa-
tions, amenable to analytical solution. Thus, from
an academic standpoint, this category of practical
applications provides engineering educators with a
wide variety of useful problems in the area of ap-
plied mathematics.
Solutions to these drainage problems have ap-
peared for many of the geometrical configurations
that typically occur in practice. These solutions nor-
mally appear in trade journals or similar outlets.
For example, in one of the earlier such articlesE'" on
this subject, formulas were summarized to compute
the time requirements for emptying vessels of four
different shapes: vertical cylinder, cone, horizontal
cylinder (with flat ends), and sphere. Later articles
gave similar formulas for draining elliptical vessel
heads at the bottom of vertical cylinders,'21 elliptical
saturator troughs (horizontal elliptical cylinders with
flat ends),l3' and horizontal cylinders with elliptical
dished heads or end.'4'
One can conceive of a number of other geometri-
cal shapes for vessels or tanks. Admittedly, they
might not occur often in the real world, but such


configurations may be of some use for academic pur-
poses, e.g., examination or homework problems. Thus
(and also in the interest of completeness) this brief
article presents tank-drainage formulas for five new
configurations: parallelepiped (or box), vertical ellip-
tical cylinder, regular tetrahedron, pyramid (in-
verted), and paraboloid.
GENERAL CONSIDERATIONS
There are two fundamental engineering equa-
tions which must be invoked in the solution to any of
these tank drainage problems. The first of these is a
dynamic material balance for the liquid in the tank,
which in this rather simple case merely reduces to
the rate of accumulation being equal to the negative
of the output rate
dV
t= -q (1)
or, more specifically

Ad= -Aov2 (2)

For the simpler geometric configurations (e.g.,
vertical cylinders [circular or elliptical] and box), the


Jude T. Sommerfeld is a professor in the School
of Chemical Engineering at Georgia Tech. He
received his BChE from the University of Detroit
and his MSE and PhD degrees, also in chemical
engineering, from the University of Michigan. His
twenty-five years of industrial and academic ex-
perience have been primarily in the area of com-
puter-aided design, and he has published over
one hundred articles in this and other areas.


Copyright ChE Division, ASEE 1992
Chemical Engineering Education









cross-sectional area (A) of the liquid surface in the
tank is a constant quantity and not a function of the
variable liquid level (z). This results in a very trac-
table, non-linear differential equation.
The effluent liquid velocity (v2) through the drain
hole or orifice is determined from a mechanical en-
ergy balance, specifically between elevation points 1
and 2 in Figure 1. This results in the classical Ber-
noulli equation
2 2
2 + Z + v2 + h
P1g + ,Z1_2+h (3)
P g 2g p g 2g
Two assumptions are conventionally made at this
point. The first of these assumes that the vessel is
1) vented to, and 2) drains to, the atmosphere, and
thus P1 = P2. The second assumption asserts that the
rate of change of the liquid level in the vessel (v,) is
negligible at all times in comparison with the liquid
velocity through the drain hole (v2). After replacing
the variable elevation difference, Z1 Z2, with the
liquid level (z) in the tank, we have

v2 = 2g(z-he) (4)
When the orifice discharge equation is used, the
fluid head loss (he) due to friction is not explicitly
calculated. Rather, an orifice discharge coefficient
(Co, generally less than unity) is introduced to at-
tenuate the fluid head and compensate for this head
loss

v2 = C 2gz (5)
As shown in most unit operations'51 and fluid flow
textbooks, this quantity CO is a function of the fluid
velocity (as incorporated in the Reynolds number)
and the downstream (orifice)/upstream diameter ra-


Vent


h
f dz
0


Y


x


q,V 2

FIGURE 1. Sketch of a vessel of arbitrary shape
with a drain hole or orifice located at the bottom.
Winter, 1992


Solutions to these drainage problems
have appeared for many of the geometrical
configurations that typically occur in practice.
[They] normally appear in trade journals
or similar outlets.

tio, although a constant value is generally assumed
for a given application. Typical values of C are be-
tween 0.60 and 1.0. Indeed, a value of 0.75 for this
coefficient is reported in a recent article'61 describing
an undergraduate experiment on efflux times through
a drain hole at the bottom of a horizontal cylinder
with flat ends. By way of information, civil engi-
neers171 know Eq. (5) (with C = 1) as the Torricelli
theorem. Insertion of this latter expression into Eq.
(2) then yields

A =-CAo2gz (6)
dt- 0
as the non-linear differential equation to be
integrated.

SPECIFIC CASES

The general integrated form of Eq. (6) can be
written as
h
t =- 1 f dz (7)
COAV 2g 0 Vz

It should be emphasized at this point that the inte-
gral in the above expression cannot be replaced by
the total volume (V) to be drained divided by the
square root of the average fluid head over the total
height (h) to be drained. That is


SAdz
0
h
f z dz


V
23J
3


where the denominator represents the average fluid
head. When A = const, the integration of Eq. (7)
yields

2A h
t= 2 h (9)
CoAo22g 9
as the general equation for the efflux time to com-
pletely drain a tank whose cross-sectional area is
not a function of height.









Parallelepipeds
This first simple case will employ Eq. (9). We
consider a rectangular parallelepiped with a length
and width (both in the horizontal plane) of a and b
units, respectively. In this case, A = ab, and the
efflux time from Eq. (9) is

2ab'jh
t= 2CoA (10)
CoA.oV2g
Clearly, in the case of a square (a = b) parallel-
epiped, one merely replaces the product ab in Eq.
(10) with a2.
Vertical Elliptical Cylinder
This is another case wherein the surface area (A)
formed by the liquid level is a constant, namely nab,
where a and b are the lengths of the major and
minor axes (again both in the horizontal plane), re-
spectively, of the elliptical cross-section. Equation
(9) in this case then becomes

2 tab h (11)
CoAo 2g
and in the special case of a vertical circular cylinder
(a = b = D/2), Eq. (11) reduces to the equation pre-
sented earlier by Foster."'

Regular Tetrahedron
We consider here only the case of a regular tetra-
hedron with four equilateral triangular surfaces
and with the drain hole located at a bottom vertex
opposite the top triangle in the horizontal plane.
The length of any edge of this figure is denoted by a.
By application of the Pythagorean theorem, the
height of any one of these triangles is equal to aw3 / 2.
The total height of this figure is determined to be
a2 /3 from a second application of this theorem,
and then, from similar triangles, the cross-sectional
area (A) of the liquid surface at any level z is given
as (3V3 / 8)z2. Lastly, insertion of this result into Eq.
(7), followed by integration, then yields for the efflux
time


5/2
t 33 h 5/2
20 CoAo2g


(12)


Note the interesting result here that the edge
length (a) does not appear in the above formula. It is
clear, however, that the initial liquid level (h) cannot
be selected in a manner inconsistent with a given
such length; that is, h cannot exceed a2 / 3. There
is also the not-surprising result ofh appearing to the


power of 5/2 in the integrated expression of Eq. (12),
which is consistent with Foster's result"1 for a coni-
cal tank.
Inverted Pyramid
This case of an inverted pyramid, like the tetra-
hedron above, is also similar to the case of an in-
verted cone. The drain hole is at the bottom vertex of
the inverted pyramid, the total height of which is
equal to c. A rectangular cross-section is assumed
for generality. Thus, let a and b (both in the horizon-
tal plane) represent the length and depth, respec-
tively, of the pyramid at its top. Two successive
applications of similar triangles yield (ab/c2)z2 as the
expression for the area (A) of the liquid level at any
elevation z. Integration of Eq. (7) with this expres-
sion for the area then gives

2abh5/2 (13)
t=52 (13)
5c CoAo 2g
as the expression for the time for complete drainage.
We note again the appearance of h to the power of
5/2 in Eq. (13), as in the preceding case. The product
ab in this equation is merely replaced by a2 in the
case of a regular pyramid with a square side length
of a.
Paraboloid
This last case examines an elliptical (again for
generality) paraboloid of total height c. As with the
vertical elliptical cylinder, a and b here represent
the lengths of the major and minor axes, respec-
tively, of the ellipse in the horizontal cross-section at
the top of the paraboloid. The equation for this fig-
ure then becomes


2 2
a b2 c


(14)


It should be noted that in many mathematical
handbooks and textbooks the right-hand side of the
above equation is written as cz or 2cz, in which case
c would have the units of reciprocal length. In any
event, any horizontal cross-section of this figure is
elliptical, and from Eq. (14) it follows that the lengths
of the major and minor axes of any such intermedi-
ate ellipse at an elevation of z are equal to


a z/c


and bVz/c


respectively. Thus, the area of this ellipse becomes
nabz/c, and the resulting efflux time formula is

t= 2abh3/2 (15)
3cCoAo 2g
Chemical Engineering Education









Finally, Eq. (15) becomes


t ED2h3/2
6cCoAo 2g


(16)


as the expression for the efflux time in the special
case of a circular (a = b = D/2) paraboloid.
NOMENCLATURE
A cross-sectional area of the liquid level in a tank at
any time, L2
Ao cross-sectional area of the drain hole or orifice, L2
a length of a rectangle, edge of a regular tetrahedron,
or major axis of an ellipse, L
b width of a rectangle or minor axis of an ellipse, L
C orifice discharge coefficient
c height of a pyramid or paraboloid, L
D diameter of a circle, L
F force unit
g acceleration due to gravity, L/T2
g conversion factor, ML/FT2
h initial height of liquid in a tank, L
h fluid head loss due to friction, L
L length unit
M mass unit
P pressure, F/L2
q liquid volumetric flow rate out of a tank, L/T
T time unit
t time, T


book review

CHEMICAL KINETICS AND DYNAMICS
by Jeffrey I. Steinfeld, Joseph S. Francisco, and
William L. Hase
Prentice Hall, New York; 548 pages, $48.75 (1989)

Reviewed by
Robert W. Carr
University of Minnesota

This book is a text intended for use in courses on
chemical kinetics at the advanced undergraduate
and graduate level. It covers a broad range of sub-
jects in empirical macroscopicc) chemical kinetics,
the kinetics of elementary reactions, the quantum
state (microscopic) approach known as chemical dy-
namics, and the connections between them. Some
background in thermodynamics, quantum and sta-
tistical mechanics, and kinetics at the level of an
introductory course in physical chemistry is assumed,
making it a suitable text for chemical engineering
students.
The book consists of fifteen chapters and three
appendices. After each chapter there is a list of ref-
erences, a bibliography, and a number of problems.
Winter, 1992


v liquid velocity, L/T
x length coordinate in the horizontal plane, L
y width or depth coordinate in the horizontal plane,
L
Z vertical elevation, L
z variable elevation of the liquid level in a tank, L
Greek Letters
n number pi (3.14159...)
p liquid density, M/L3
Subscripts
1 liquid surface in the tank at any time
2 bottom of tank (at drain hole)

REFERENCES
1. Foster, T.C., "Time Required to Empty a Vessel," Chem.
Eng., 88, 105, May 4 (1981)
2. Koehler, F.H., "Draining Elliptical Vessel Heads," Chem.
Eng., 91,90, May 14 (1984)
3. Sommerfeld, J.T., "Inventory and Drainage of Saturator
Troughs," Amer. Dyestuff Reporter, 79, 44, April (1990)
4. Sommerfeld, J.T., "Tank Draining Revisited," Chem. Eng.,
97, 171, May (1990)
5. Brown, G.G., et al., Unit Operations, John Wiley & Sons,
New York, NY, p. 158 (1953)
6. DeLozier, K.R., K.A. Platner, and J.T. Sommerfeld, "An
Inexpensive Laboratory Experiment on Fluid Dynamics,"
Intl. J. Appl. Eng. Ed., 6, 377 (1990)
7. Vennard, J.K., and R.L. Street, Elementary Fluid Mechan-
ics, 5th Ed., John Wiley & Sons, New York, NY, p. 131
(1976) J



The references and bibliography will be useful for
those seeking entry into the literature of various
topics in chemical kinetics.
The book is unusually broad in its coverage, deal-
ing with a number of subjects that are usually in-
cluded only in more specialized texts, but which
have now become commonplace in kinetics practice.
Chapter 1 is conventional in its treatment of ele-
mentary concepts and definitions, but Chapter 2,
dealing with complex reactions, goes beyond the usual
presentation of analytical solutions to coupled sets
of ordinary differential equations to discuss applica-
tions of Laplace transforms, matrix methods, nu-
merical methods (Euler, Runge-Kutta, predictor-cor-
rector) and stochastic methods. Computer programs
for Runge-Kutta integration and Monte Carlo simu-
lation are included. Chapter 3, on kinetic measure-
ments, emphasizes modern instrumental methods
for direct detection of reactive intermediates and the
treatment of kinetic data. Sensitivity analysis, an-
other subject not normally covered in introductory
texts but which is of enormous help in understand-
ing a mechanism, is introduced in this chapter.
Chapter 4 deals with reactions in solution and
Continued on page 49.










curriculum


IDEAS ABOUT CURRICULUM


DONALD R. WOODS
McMaster University
Hamilton, Ontario, Canada L8S 4L7


What a challenge it is to develop a curriculum
that will prepare our students for the years
2020 to 2040! Should we focus on breadth, or depth?
Produce a generalist, or a specialist? Besides subject
knowledge, how much explicit training should be
included to develop problem-solving skills, commu-
nication skills, interpersonal and team skills, and
the ability to learn-how-to-learn? What is the role of
projects?
Our department at McMaster is relatively small
and does not have expertise in some of the "emerg-
ing technologies.""' We feel that the design project is
the capstone course, that synthesis activities should
be spread over the whole curriculum, and that com-
puter skills should be well developed. In this con-
text, what should our curriculum be and how should
we go about developing it?
GOALS AND ANALYSIS OF OPTIONS
To address the issues of curriculum design sub-
ject to our unique constraints, we examined and
defined our goals, and consulted with many others
during our two years of deliberations. As a depart-
ment, we set goals so that our graduates
would be technically sophisticated, with an emphasis
on the fundamentals
would not only be computer literate, but also be able
to program effectively and efficiently and would be
familiar with a wide range of computer executive
programs such as FLOWTRAN, PROCESS, HYSIM,
and various spreadsheets


could learn on their own
would have good communication, problem-solving,
interpersonal, and group skills
would know how to create hypotheses, perform
experiments, and draw valid conclusions
would have enriched experience through senior
electives drawn from our research expertise
Acceptance of these goals immediately highlighted
some directions: depth of the core fundamentals,
generalist with electives available in our areas of
expertise. Because the students cannot learn in class
all that they will need to know in their career, we
will have explicit activities to develop the student's
confidence in their own how-to-learn ability.
We held a number of faculty retreats for identify-
ing and streamlining the core concepts in chemical
engineering and removing the overlaps and the in-
consistencies. Concurrently, a team of students per-
formed the same tasks from their perspective. Spe-
cial workshops were held on such topics as the labo-
ratory (led by O.M. Fuller'21) and on the results of our
four-year research project to discover methods of
improving the development of problem-solving and
interpersonal skills. 3M
We discovered, among other things, that using
open-ended problems and modeling how we solved
problems did not develop the student's problem-solv-
ing skills, just as giving them good trouble-shooting
problems did not improve their trouble-shooting abil-
ity. If they entered the activity with the ability to
think of several causes (whereas we hoped they would
be able to think of, say, ten), they left the activity
still able to think of only several.
To develop the student's ability to learn-how-to-
learn, we chose as a model the problem-based, small-
group, self-assessed, self-directed learning model used
by McMaster Medical School.'41 However, we did not
have the advantage of its admission policies (that
eliminated students who did not have well-devel-
oped problem-solving and group skills), the time-
table flexibility, or the resources to provide a faculty
tutor for small groups of five students each. So we
had to rephrase the problem to "how can we adapt
their approach to our constrained system?" and chose
Chemical Engineering Education


SDon Woods is a professor in the department of
chemical engineering at McMaster University. He
is a graduate of Queen's University and the Uni-
versity of Wisconsin. His teaching and research
interests are in surface phenomena, plant de-
sign, cost estimation, and developing problem-
solving skills.
0 Copyright ChE Division ofASEE 1992










We discovered, among other things, that using open-ended problems and modeling how we solved
problems did not develop the student's problem-solving skills, just as giving them good
trouble-shooting problems did not improve their trouble-shooting ability.


to use a thread of courses throughout the curriculum
which would develop problem-solving and interper-
sonal skills and small-group learning skills.['5,6 The
development of problem-solving and interpersonal
skills was an important goal in its own right as well
as an important component needed for the design
project.
Another issue was the integration of previous
courses with the capstone design course so that the
team-approach project course could focus on synthe-
sis. Our approach was that the principles of engi-
neering economics, ethics, professionalism, process
improvement, and process design would be handled
in a separate "course"17' and to have industrial cli-
ents with real projects.'"8 Teams of about ten stu-
dents and two faculty would tackle a problem. For
example, a recent project was with Esso/Texaco on
assessing an industrial wastewater-treatment facil-
ity. Difficulties arose since the students entered the
project without sufficient experience with computer
executive programs and without having had some of
the core courses (such as reaction kinetics and reac-
tor design, mass transfer and process control).
In rethinking the approach for the new curricu-
lum we were not trying to replace the design experi-
ence-we wanted to enrich it. Four important changes
pertaining to the design course were:
1. Introducing computer programming (especially the
executive programs PROCESS and HYSIM) in the
sophomore and junior years so that the students
would enter the project with this expertise already in
place.
2. Providing all of the basic fundamentals before the end
of the junior year.
3. Shifting the course load in the design course so that it
was two tutorials per week in both semesters.
4. Providing explicit training ahead of time on how to
work effectively in groups and how to be a team
leader.
Thus, in addition to a "stream" of core fundamen-
tals that would be completed by the end of the junior
year, we envisioned a series of learning activities
on computer programming, numerical analysis, and
executive programs
on "how to experiment"
to develop communication skills
to develop problem-solving and interpersonal skills
These activities would culminate in and be inte-
Winter, 1992


grated through the capstone design course. The skill
of learning-how-to-learn would be acquired through
problem-based, small-group, self-directed learning
in the course "engineering economics."
CURRICULUM
For the core fundamentals the former senior-
level courses were altered as follows:
the topics of reaction kinetics and reactor design,
stagewise operations, diffusional transport phenom-
ena, and separations were split into two phases, with
the junior level offering the basics and a senior level
offering enrichment. This actually increased the
amount of coverage of this material
process control was shifted to the junior year.
Thus, by the end of the junior year the students had
the core fundamentals in all areas.
To provide expertise in the computer executive
programs that would be needed for the design project
course, a string of three courses in the sophomore-
junior years focused on computer programming, nu-
merical methods and modeling, and the use of execu-
tive packages. For example, PROCESS or HYSIM
was introduced in the sophomore course, was used
to enrich other courses in the sophomore-junior years,
and was used explicitly in a computer-simulation
course in the junior year that considered a relatively
well-structured problem on a small section of a pro-
cess. Since the content in this string of courses is not
ambitious, we decided to combine the problem-solv-
ing/interpersonal skills development activities with
each of these three courses. In principle this would
provide a ready-made "problem" and "group activ-
ity" where the problem-solving skills could be ap-
plied. Thus, the problem-solving activities would not
be developed in a vacuum. Rather, they would be
embedded and bridged into the computer program-
ming, model building, and process simulation activi-
ties. An example of the latter is given by Woods,
Wood, and Gallinger.'91
The problem solving/interpersonal skills/self-di-
rected learning components were dovetailed into the
three team-taught computer programming courses
described above and into a senior-level course on
process economics and synthesis. In addition, the
first sophomore course has as a co-requisite the mass
and energy balances course. In this way we hoped to
ensure that we had sufficient variety and opportu-
nity to apply the problem-solving skills in solving










homework problems. Details of this approach are
available.1101
A revised series of activities for the laboratories
and on experimentation were added. In the sopho-
more year the emphasis was on familiarity with the
hardware"" and on measurement techniques. In the
junior year the emphasis was on doing a series of
medium-size, medium-duration experiments to il-
lustrate principles. In the senior year, the students
do several in-depth experimental projects. A lecture
component is included with the laboratories to con-
sider activities on data treatment, hypothesis test-
ing, and providing individual feedback about the
written reports.
Another stream of experience focused on commu-
nications skills. The Faculty of Engineering has a
policy that before students are admitted to the sopho-
more year, they must pass a Test of Writing Compe-
tence. So the students in our sophomore course on
communication (see Woods and Feuerstein'121) had
the necessary basic fluency in English. We then for-
mally reinforced this skill with private reviews and
feedback on the laboratory reports in the junior and
senior years and in the presentation of the design
project.
Many of our introductory, graduate-level courses
were modified so that, although the lecture compo-
nent was common, a graduate and undergraduate
offering of the course would be presented in areas of
our research expertise: biomedical, polymer engi-
neering, process control, statistics, environmental
engineering, particle processing, and surface phe-
nomena. Senior students could elect to take three of
these courses.
Figure 1 shows the resulting new curriculum.
The freshman year is common to all branches of
engineering, so we had no direct control over the
content in that year. Accreditation requirements dic-
tate the eighteen credits of English and humanities
(in the freshman, sophomore, and senior years). Fac-
ulty requirements dictate the engineering courses in
mechanics/structures and electricity/magnetism.
design project: senior year, 4 credits
enrichment chemical engineering electives: senior
year (compared with general technical electives), 9
credits
computer/simulation stream: introduction, freshman
year, 4 credits: sophomore year, 1 credit; junior year,
1.5 and 2 credits
communication stream: sophomore year, 1.5 credits;
part of 2-credit lab in both junior and senior years
and part of the design project
problem solving/interpersonal skills/self-directed
learning: sophomore year, 2 credits; junior year, 1.5


and 1 credit; Senior year (with special co-requisite
requirements to embed the skill development"), 2
credits
experimentation and laboratory: sophomore year,
about 1/2 credit; junior year, 2 credits; senior year, 2
credits
mass and energy balances: sophomore year, 8 credits
thermodynamics: junior year, 3 credits
reaction kinetics/reactor design: junior year, 3 credits;
senior year, 3 credits
fluid mechanics: sophomore year, 4 credits
heat transfer: junior year, 4 credits
stagewise operations and mass transfer: junior year, 4
credits; senior year 3 credits
process control: junior year, 3 credits
engineering economics and process engineering:
senior year, 2 credits
HOW WELL HAS IT WORKED?
In two words-extremely well. Employers com-
ment very positively on the quality of our entry-level
graduate. Alumni are enthusiastic about the cur-
riculum and the preparation they have received. The
most unique innovation has been in the problem-
solving/interpersonal skills/self-directed learning
components. For this reason, we have conducted a
variety of pre- and post-tests as well as control-
group assessments. They have shown a statistically
significant increase in performance in the main-
stream chemical engineering courses and a 1 to 2
standard deviation shift in confidence that they now
possess skill in problem solving. The self-directed


EBM.


Figure 1
Chemical Engineering Education










learning has been more difficult to quantify. Anec-
dotal evidence abounds as to the student's prefer-
ence for this mode of learning and to instructors who
feel that the students learned more and performed
as well, if not better, on written exams than did
students who previously "learned" the material via
traditional lectures.15,61

We are changing some components since they did
not work out as we had hoped. In particular, the
blend of problem solving with computer program-
ming in the sophomore year did not provide the
synergy we anticipated. The students viewed them
as two separate parts. So we have integrated com-
puter programming into the mass and energy bal-
ances course and maintained a separate 2-credit
course on problem solving that applies the skills in
the required co-requisite mass and energy balances
course. This latter bridge has worked extremely well.

The extra course in reactor design in the senior
year, likewise, has not lived up to its expectations.
So we combined the non-ideal behavior with the
junior course (now 4 credits) and created a richer,
elective course in reactor design. Also, the students
found that the sophomore fluid mechanics course
was very challenging, so we have now switched this
to the junior year.

The design project is much more effective in the
new format. We have shifted from PROCESS to
HYSIM as the computer executive program of choice.

These changes have been minor. All in all, we
feel that this new format offers a very viable curricu-
lum model for the year 2000.

ACKNOWLEDGEMENT
I am pleased to acknowledge the efforts and in-
put of my colleagues. To Bob Moore, Leslie Eubanks,
Cam Crowe, Terry Hoffman, Joe Wright, Andy
Hrymak, Phil Wood, Paul Taylor, Bob Marshall, and
Ian Doig for nurturing the problem-solving courses;
to Derek Ryder, Steve Kelly, Joe Laricchia, Craig
McDougall, Bret Cousins, and Sandra Allen; to Bob
Anderson, Les Shemilt, John MacGregor, Irwin
Feuerstein, Malcolm Baird, Marios Tsezos, Andy
Benedek, Keith Murphy, Archie Hamielec, John
Brash, John Vlachopoulos, and Jim Dickson. The
Ontario Universities Program for Instructional De-
velopment and McMaster University funded the four-
year program to understand and develop the prob-
lem solving/interpersonal skills components. In par-
ticular, I thank Alvin Lee, Vice-President Academic,
and Les King and Alan Blizzard, Instructional De-
velopment Center, for ongoing support.
Winter, 1992


REFERENCES
1. Baird, M.H.I., "Things are Humming at McMaster," Chem.
Eng. Ed., 4, 112 (1970)
2. Fuller, O.M., "On Experimentation: A Workshop," McMaster
University, Hamilton, May 6 (1981)
3. Woods, D.R., J.D. Wright, T.W. Hoffman, R.K. Swartman,
and I.D. Doig, "Teaching Problem Solving Skills," Annals of
Eng. Ed., 1, 238 (1975)
4. Barrows, H.S., and R.M. Tamblyn, Problem Based Learn-
ing: An Approach to Medical Education, Springer, New
York (1980)
5. Woods, D.R., "Implementing Self-Directed, Problem Based
Learning in a Traditional Curriculum," Chapter 4 in The
Challenge of Problem Based Learning by D. Boud and G.
Feletti, Kogan Page (1991)
6. Woods, D.R., "Self-Directed Learning," MPS Unit, McMaster
University, Hamilton, Canada (1990)
7. Woods, D.R.,"Innovations in a Design and Development
Course," Chem. Eng. Ed., 2, 162 (1968)
8. Woods, D.R., T.W. Hoffman, and A.I. Johnson, "Teaching
Experience with Design and Simulation Projects," Chem.
Eng. Ed., 7, 96 (1973)
9. Woods, D.R., P.E. Wood, and F.H. Gallinger, "Exploiting the
On-Campus Boiler House," Chem. Eng. Ed., 20, 28 (1986)
10. Woods, D.R., C.M. Crowe, P.A. Taylor, and P.E. Wood, "The
MPS Program for Explicitly Developing Problem Solving
Skills," 1984 ASEE Conference Proceedings, Salt Lake City,
UT, 1021 (1984)
11. Woods, D.R., R.W. Dunn, J.J. Newton, and D.J. Webster,
"The Travelling Circus as a Means of Introducing Practical
Hardware," Chem. Eng. Ed., 12 116 (1978) and "Touching
Technology," Chem. Tech, Oct., 586 (1981)
12. Woods, D.R., and I.A. Feuerstein, Eng. Ed., 70, No 7, 745
(1980) J


LETTER: Equilibrium Thermodynamics
Continued from page 29.
The adiabatic constraint is removed in their second
and third problems, which should, therefore, cause the
student no difficulties.
(Many of the points discussed here I raised in the
literature a decade ago or more.'6-111)
W.F. Harris
Optometric Science Research Group
Department of Optometry
Rand Afrikaans University
P.O. Box 524
Johannesburg 2000 South Africa

REFERENCES
1. Williams, D.F., and D. Glasser, Chem. Eng. Ed., 25, 74 (1991)
2. Williams, D.F., and D. Glasser, Chem. Eng. Ed., 25, 164
(1991)
3. Callen, H.B., Thermodynamics, John Wiley & Sons, New
York, NY (1960); 2nd ed. (1985)
4. Georgian, J.C., Nature, 201, 695 (1964)
5. Georgian, J.C., J. of Chem. Ed., 43, 414 (1966)
6. Harris, W.F., ChemSA, 6, 195 (1980)
7. Harris, W.F., Int. J. of Mech. Eng. Ed., 9, 181 (1981)
8. Harris, W.F., Int. J. of Mech. Eng. Ed., 9, 317 (1981)
9. Harris, W.F., ChemSA, 7, 259 (1981)
10. Harris, W.F., ChemSA, 8(7), 82 (1982)
11. Harris, W.F., Int. J. ofMech. Eng. Ed., 10, 287 (1982) 0










laboratory


COMPUTER CONTROL

OF A DISTILLATION EXPERIMENT


CARL T. LIRA
Michigan State University
East Lansing, MI 48824-1226

U sing the GENESIS environment, software has
been developed for computer control of a 10-tray
column in the chemical engineering department at
Michigan State University. The software permits
sequenced start-up of the column, transition to
steady-state cascade control, and data acquisition.
During runtime, process data are displayed on an
equipment schematic or a graphical trend window.
The GENESIS software is a powerful authoring sys-
tem for writing control and data acquisition pro-
grams and requires minimal background in com-
puter control.
EQUIPMENT AND INSTRUMENTATION
Our department developed the control program
to help introduce industrial-style computer control
to undergraduate students. The system has two dif-
ferent columns which may be used independently
and a single reboiler, condenser, and tank system.
Flows through the desired column are obtained
by moving blind flanges just above the reboiler and
just below the reflux splitter. The columns are Pyrex,
which permits the students to observe upsets in the
columns as well as normal tray operation. The six-
inch diameter columns have ten trays. One column
consists of five valve trays and five bubble cap trays.
The other column consists of sieve trays. The trays
are constructed from stainless steel and Teflon. (Fur-
ther reference to the system will refer to "the col-
umn" since only one column is used at a given time.)

Our department developed the control
program to help introduce industrial-style
computer control to undergraduate students. The
system has two different columns which may be
used independently and a single reboiler,
condenser, and tank system.
@ Copyright ChE Division ofASEE 1992


Carl Lira is an assistant professor in the chemical
engineering department at Michigan State Univer-
sity. He received his BS degree from Kansas State
University and his MS and PhD from the Univer-
sity of Illinois, Urbana-Champaign. He joined the
MSU faculty in 1986. His current research inter-
ests are in the area of processing with super-
critical fluids.


A schematic of our equipment is shown in Figure
1. The input and output signals are summarized in
Table 1. All of the wiring from the column passes
through conduit to a wall panel box. An identical
panel box is installed on the back of the computer
cart. BNC and computer cables are used to connect
the two panels. This provides flexibility in computer
location as well as in use of the computer with other
experiments.

INTERFACE HARDWARE
An IBM AT is used for computer control. We
purchased interface hardware from Analog Devices
Inc. We use both RTI815 and RTI817 boards in the
computer. These boards require signal conditioning
which must be performed with additional boards.
The combination of boards and configuration are
dependent on the large number of signals which
were interfaced in our system, and details will be
supplied upon request.

SELECTION OF SOFTWARE
Several different commercial programs are avail-
able which could equivalently fulfill the objectives of
this project. GENESIS was selected because of the
following capabilities:
1. A user-friendly interface which requires minimal
expertise in hardware/software interfacing and data
input
2. Graphic display of the process flowsheet and
process variables


Chemical Engineering Education















Device
Thermocouple
DP cell
Float Level
Pressure
Feed Flow
Valve Position


TABLE 1
Summary of Computer I/O

Computer Inputs


# of
Signal Type Signals
Analog (V) 13
Analog (mA) 2
Digital (V) 8
Analog (mA) 1
Analog (mA) 1
Digital (V) 6


SComputer Outputs
Reflux Splitter Digital (V) 1
Bottoms Overflow
Solenoid Valve Digital (V) 1
Steam Control
Valve Analog (mA) 1
Valve Position Digital (V) 6


3. Graphic-based control-loop strategy to provide easy
editing and visualization of the control loop
4. Sequencing to permit stepwise start-up of the
column
5. Security to limit the control parameters and
setpoints the students can change
6. Capability of "on-the-fly" alteration of control
parameters.

In addition to the above advantages, MSU had
previously purchased the interface hardware and
therefore did not need a hardware/software combi-
nation. We also knew that GENESIS software had
received favorable reviews.I",2

PROGRAMMING WITH GENESIS
There are two main components of GENESIS-
the strategy builder and the runtime system. The
strategy builder is used to "write" the control pro-
gram. Actually, no code needs to be written since the
control loops are entered graphically on the screen.
As part of the strategy builder, GENESIS provides a
powerful graphics program (the graphics builder) to
construct schematics of the process which are to be
viewed during process control.

An example of a simple strategy is shown in
Figure 2. The strategy is constructed using a mouse
to select icons from pictorial menus. By clicking on
an icon, the algorithm blocks are selected and may
be placed on the screen, dragged, and sized with the
mouse.

Each of the algorithm blocks is labeled with an
Winter, 1992


BV Ball Valve
DP-Differential Pressure Cell
FI- Mass Flow Meter FC- Flow Control Input
LI- Float Level Indicator PR- Pressure Regulator
PS- Pressure Sensor RS- Reflux Splitter
SV- Solenoid Valve TI- Thermocouple


Figure 1. Schematic of the instrumentation utilized for the
distillation column.


acronym which identifies the type of block. In Figure
2, DEV represents the interface board (hardware
device). AIN, AOUT, and DOUT are analog inputs,
analog outputs, and digital outputs, respectively.
DGAP and PID represent a gap controller and PID
controller. Many other blocks may be used, includ-
ing an F(x) block which may contain up to five lines
of calculations using several inputs and registers.
Each algorithm block is given a tag name to
make the block unique and easily identifiable. In
this example, temperature input blocks are indi-
cated by PLATE1 and PLATE7. The DGAP control-
ler and PID controllers are used to control the re-
boiler level and steam flow. Specification of details
of a block is achieved by selecting the block with the
mouse and entering information in parameter win-


Figure 2. Example control strategy as constructed by
strategy builder.











dows which pop up on the screen. The actual control
loop is more complex than Figure 2 because it is
desirable to use direct manipulation of the steam
flow setpoint while keeping the temperature control-
ler disabled during start-up of the column or for
manual override of cascade during operation, as
shown in Figure 3.

An example of our control screen designed with
the graphics builder is shown in Figure 4. As the
students step through the start-up procedure, differ-
ent control screens are displayed. For each of our
screens, the equipment schematic remains the same,
but the control panel in the upper left contains dif-
ferent instructions. Figure 4 is the cascade control
screen. The control screen is displayed using colors
which are not reproduced here. During runtime, the
graphics screens provide instructions and data dis-
play. Depending on how the graphics are designed,
process data may be displayed numerically, graphi-
cally, or pictorially using a change in the size or color
of an object. In our schematic the valves are dis-
played in brown when closed and bright green when
opened.

On the control screen, all process data are nu-
merically displayed near the sensing device in the
schematic and are updated each second. The secu-
rity of cursor positions and the range of valid entries
is controlled. The only valid cursor locations are the


data-entry positions in the
per left of the screen.
Alarm messages in the
boxes on the right side
of Figure 4 are displayed
only if triggered by a spe-
cific condition. Different
alarm levels may be
used to identify the se-
verity of an event. In our
application, all of the
alarms are warnings ex-
cept for a low reboiler
level which triggers the
emergency state and
shuts off the steam line.

RUNTIME

Once a strategy and
associated graphics dis-
play have been pro-
grammed, the process
may be executed. Trend
windows permit graphic
display of the data dur-


control panel in the up-


Screen B
Control Variables
Plate 7
Setpoint CC) 85 5
Measured CC) 85 4
Ref lux
Cycle TimeCsec) 40
D/L Ratio 0 5B
Man Valve Control' N
Man Steam Setpoint
CO-500 Ibs/h) 200
Valve PosItions
Feed Tank 1 BV-1 Y
Feed Tank 2 BV-2 N
Column Feed BV-3 Y
Emergency Shutdown? N
FT-1 FT-2
Flow
c -1 1-- 10.2

BV-1 BV-2
P1
To enter data,
use tab key. make -T
changes, press Alt-D


ing runtime to follow the controller performance
and the approach to steady state. Trends may be
displayed over time intervals of 1, 6, or 30 minutes
only. Up to 20 variables may be trended and are
displayed on the screens in groups of up to 5. The
screens may be selected using the "Page Up" and
"Page Down" keys. An example of a trend window is
shown in Figure 5, illustrating the effect of a tem-
perature setpoint change on the steam flow rate.
The trend windows utilize colors to differentiate be-
tween variables and labels. For Figure 5, different
fonts have been used to differentiate variables for


Feed


Temp
Error "1 Controller Output
Signal ,
Manual
Temp Steam Flow
Setpoint Setpoint


Steam Flow
Controller Output


Signal


Figure 3. Schematic of the cascade control loop used for
the MSU distillation column.


CNDSR
PLATE
PLATE
PLATE
PLATE
PLATES
PLATES
PLATE
PLATES
PLATE
PLATE10
REBOILER
FEED


SHUTDOWN


SOUTH COLUMN FEED OPEN
10 FEED TANK FULL
--I-- FEED TANK 1 LOW
FEED TANK 2 LOW
BOTTOMS RECEIVER FULL
PRODUCT RECEIVER FULL
EMPTY RECEIVERS !
LOW STEAM PRESSURE
LOW REBOILER, EMERG SHUTDOWN!
Pressure
Steamb h1 % 9 pig
Seto 145
Meas 17 O -rn
--LX>-STEAM
JReboilr CV- BV-5
I 27 5 n Procrammer C T Lire


Figure 4. Example control screen for the MSU control program. This screen is the screen
for cascade control
Chemical Engineering Education










purposes of this publication. During runtime, ac-
quired data may be written to the hard disk at time
intervals ranging from 1/10th-second upwards.

STUDENT USE

The objectives of the use of the distillation equip-
ment are:
to expose the students to the industrial equipment
and sensors
to run the column at steady-state and collect and
analyze tray samples to calculate tray efficiencies
to expose the students to an industrial-style control
system.

The students are provided with a short problem
statement. Typically, they are to separate an equimo-
lar feed of ethanol and propanol to a product of 90%
ethanol and bottoms of 90% propanol and to deter-
mine Murphree tray efficiencies. They must obtain
literature data to construct the x-y diagram and
then perform a McCabe-Thiele analysis to predict
the composition on tray 7 at steady-state, which
determines the setpoint temperature. Also, the ap-
propriate reflux ratio must be calculated.
During the first 3-hour laboratory period the stu-
dents familiarize themselves with the experiment by
watching a video tape which illustrates the start-up
procedure and use of the computer. The teaching
assistant demonstrates the use of the gas chromato-
graph used for sample analysis. Students read
through the laboratory manual which contains por-


JUN 11/91
14:34:43

1


500
500
100
100







0
0
15
15


-TREND -
S I I I


I I


I I I I I I


I- -I

30 Minutes
STMFLW OUT 120.2 LBS/HR
STMSET OUT 122.0 LBS/HR
TEMPSET OUT 84.12 C
PLATE7 OUT 83.93 C


Figure 5. Example trend window which may be viewed
during experimental runs.
Winter, 1992


During the first 3-hour laboratory
period the students familiarize themselves with
the experiment by watching a video tape
which illustrates the start-up procedure
and use of the computer.


tions of the data sheets and documentation which
the manufacturers supply with the instrumentation.
Upon student request, the column is started to fur-
ther illustrate correct operation.
The students submit a written project proposal
twenty-four hours before the second laboratory pe-
riod which summarizes their McCabe-Thiele calcu-
lations and proposed laboratory procedure. Upon ap-
proval, the students may begin the experimental
work.
A six-page instruction summary in the labora-
tory manual provides the necessary guidance for the
use of GENESIS with this experiment. Although an
experienced operator can start the column without
the instructions, the printed information provides
useful reference regarding the steps performed at
each screen during the sequence, hints on steam
flow during start-up, explanation on handling alarms,
and important keystroke combinations to plot trend
data on the control screen.
The sequencing capability of GENESIS is uti-
lized during the start-up to display instruction
screens to the students. At each screen the students
must take some action, such as verifying that the
cooling water is flowing, or draining condensate from
the steam line, or turning on the pump. When they
are ready to proceed, a keyboard command takes
them to the next sequence step.


Feed is prepared by mixing alcohols from the
product and bottoms tanks. The composition is de-
termined by gas chromatography. Usually, students
use 30-45 minutes to drain the tanks and prepare
the correct feed composition. Start-up of the column
is achieved using the steam-flow controller only. The
setpoint is entered on the screen by the students.
Although the written instructions provide sugges-
tions on the steam flowrate, the final choice is made
by the student. Typically, the students will encoun-
ter some type of difficulty during start-up, such as
flooding due to vapor flow up the downcomer of the
bottom tray
flooding on the feed tray
reboiler level too high or too low
slugs of vapor in the reboiler thermal siphon, causing
incorrect readings in the reboiler level.









Usually, we permit the students to encounter
these difficulties because they help to illustrate the
limitations of control systems. In the absence of com-
plications, vapors reach tray 7 in about fifteen min-
utes and the condenser in thirty minutes.
After vapors have reached the condenser the stu-
dents proceed to the cascade control screen shown in
Figure 4. Two factors are important in determining
the approach to steady-state. First, the temperature
controller has very little integral control due to the
sluggish behavior of the system. Students are en-
couraged to use the manual steam flow setpoint
until the temperature of plate 7 is within three de-
grees Celsius of the desired setpoint before proceed-
ing to cascade. The second factor is that GENESIS
provides bumpless transition of the steam flow set-
point upon transfer to cascade control, which means
the steam setpoint will be adjusted gradually from
the value existing at transfer to cascade control.
Students are encouraged to anticipate the sluggish-
ness and adjust the steam flow setpoint up or down
before transferring to cascade control. This adjust-
ment requires them to make a judgement as to
whether the system will overshoot if operating at
the current settings. An overshoot of a couple of
degrees will result in stabilization at setpoint taking
up to fifty minutes. A well-planned transfer to cas-
cade will result in a stabilization in about twenty
minutes. The trend screen data plots are useful in
watching the approach to setpoint and steady-state.
After tray 7 has reached the setpoint value, tray
temperatures are monitored for at least twenty min-
utes to assure that temperature drifting (composi-
tion changes) are not present on any trays. The
students then collect tray samples.
Shutdown of the system is also performed using
GENESIS. Shutdown uses the same procedure as an
emergency shutdown, so the student operator enters
a "Y" in the cursor location next to the "EMER-
GENCY SHUTDOWN?" prompt in Figure 4. (This
prompt is present on all screens.) The record of col-
umn history is printed from the hard disk following
completion of the experiment and is submitted with
the student's laboratory report.
Typically, students will repeat the experimental
run in a third laboratory period, making slight modi-
fications to the reflux ratio or setpoint temperature.
With our current laboratory objectives, the students
do not have adequate time to fully explore the wealth
of control experiences feasible with GENESIS. Pos-
sible control exercises include studying the effects of
sudden changes in feed flowrate or reflux ratio. An-


other possibility is to further explore the tuning of
the controllers. Typically, the students do not fully
appreciate the sluggishness of the system. An ex-
ample exercise would permit them to increase the
integral control to improve the rate of approach to
setpoint and then observe that the change results in
continuous oscillations about the setpoint value. An-
other possibility is to permit the students to perform
the Ziegler-Nichols tuning of the controllers, or to
attempt control without using cascade methodology.
We do not utilize these unsteady-state exercises be-
cause, with the available laboratory time, it is not
possible to restabilize the system and collect tray
samples for meaningful comparison with McCabe-
Thiele analysis.

CRITIQUE AND SUMMARY
Student response to the use of GENESIS is ex-
tremely positive. We have chosen to install the con-
trol system on equipment that illustrates the use of
a large number of data inputs. GENESIS would also
be extremely useful for a smaller scale experiment
with a single control loop. The flexibility of software
permits the instructor to prepare exercises and con-
trol screens which require little knowledge of con-
trollers or which are very advanced. The capability
for watching unsteady-state behavior is excellent.
It would be easy to plot the integral, derivative
and proportional components of a controller on the
same trend window as the overall controller and
loop response. The use of logged data from the spread-
sheet compatible files permits the students to fur-
ther manipulate their data and prepare plots. The
data files are compatible with LOTUS 1-2-3 and
may be manipulated further, although we do not use
this feature.
We have found the application of GENESIS to
provide a powerful method for design of control and
data acquisition programs, especially for those with-
out experience in computer control. Because of the
graphic-based strategy builder, the programmer does
not need to develop or maintain expertise in reading
or writing to registers. Once an understanding of the
software has been developed, it becomes easy to
write new control loops. The control program does
not greatly change the instructional time that must
be spent with the students, but provides exposure to
an important aspect of control technology.
ACKNOWLEDGEMENTS
I would like to thank Al Paulsen of Dow Chemi-
cal for his assistance in specification of instrumenta-
tion, Dow Chemical for funds for the instrumenta-
tion, and Bruce Wilkinson of the Chemical Engi-
Chemical Engineering Education










neering Department at MSU for planning and su-
pervising the installation of the instrumentation.

REFERENCES
1. Bottali, P., and M. Kelly, "Software Monitors Electrical
Loads at IBM,"" Control, March (1989)
2. Software Engineering Consultants, "GENESIS: Human Fac-
tors Approach to SCADA," Control Engineering, November
(1989)

APPENDIX 1: Equipment Description
The level indicators are floats which indicate only whether a
tank is nearly full or empty. All of the ball valves are air actuated,
with a sensor to indicate valve position. The air supply is switched
to the valves using solenoid valves in an air-supply line. The feed
pump is a small bellows pump. Feed rate is sensed with a
Micromotion mass-flow meter. The oscillations from the recipro-
cating pump cause fluctuations in the flow-meter readout which
are dampened by adjusting the electronic filter furnished on the
flow meter. The flow-meter signal is furnished to the computer as
a 4-20 mA signal. The feed enters the column after passing
through a single tube in shell preheater constructed from two
different sizes of stainless-steel tubing.
The steam is regulated from a supply pressure of 90 psig to
20 psig before entering the system. A steam pressure sensor
provides a 4-20 mA signal to the computer. Steam flow is sensed
with a differential pressure (DP) cell across an orifice plate. The
DP cell provides a 4-20 mA signal on a single pair of wires using a
24V supply. Steam flow is regulated by a 4-20 mA signal from the
computer. An I/P (current to pressure) converter provides the air
pressure to drive the control valve. The relative placement of the
orifice plate and control valve have provided satisfactory control
of the column. In the current configuration, the upstream pres-
sure on the orifice plate depends on the control-valve position.
The reboiler consists of a stainless-steel heat exchanger. The
reboiler level is sensed using a DP cell. This DP cell operates on
the same principles as the steam-flow DP cell. The reboiler level
overflow to the bottoms receiver is controlled with a solenoid
valve. The computer controls the solenoid with a bistate gap
controller signal based on the reading from the DP cell. A needle
valve in the overflow line serves as a restriction to limit the flow
rate. The gap controller is set to open the solenoid at hydrostatic
pressure of 25.1" of water and to close at 24.9".
The reflux splitter is driven by a 120 VAC solenoid. The
computer may be used to close a relay to provide the AC power or
a separate electrical/mechanical timer may also be used.

APPENDIX 2: Interface Hardware and Software
Analog signals are isolated and converted with an Analog
Device 3B signal conditioning system. Analog input signals from
the thermocouples and current inputs are converted to 0-10V
signals which the interface boards then converted to numerical
values of 0-4096. GENESIS reads the registers and converts the
register values to engineering units. The analog output for con-
trol of the steam flow is passed through the conditioning system
with a reverse of this procedure. Digital signal I/O is handled
through optical isolator modules and the signals are passed
through either the RTI815 or the RTI817.
GENESIS is available from ICONICS (phone 508-543-8600).
A variety of control versions is available in addition to a wide
range of device drivers to communicate with hardware from many
manufacturers. Data Translation and MetraByte boards, which
are popular in many university laboratories, are supported. GEN-
ESIS is priced competitively with other control packages such as
LABTECH Control. ICONICS offers a university discount. A very

Winter, 1992


inexpensive demo version is available which is limited to 20
algorithm blocks.

APPENDIX 3: Using the Strategy Builder
The start-up of the column is achieved using a sequencing
algorithm block which is not illustrated in Figure 2. The se-
quencer is designed with a series of states. Eight states are used
in our start-up sequence, along with an emergency state. Control
events may be associated with each state. Sixteen bits are associ-
ated with each state which may also trigger events. The status of
each bit may change with a given state of the sequencer. In our
application, the sequencer is used to control the display of control
screens which provide instructions for start-up, enable alarms
and warning messages which alert the students when a step has
not been completed, and provide restrictions on the type of steam-
flow control for a given state.
After an algorithm block has been placed on the screen, the
details regarding the block may be entered by clicking the mouse
on the block. When a block is selected, a parameter window pops
up on the screen. Parameter values for input ranges, output
ranges, alarms, controller settings, etc., are entered in this win-
dow. All of the values entered in the blocks are in engineering
units, making the programming quite simple. The instrument
range and operating range may differ. For example, our thermo-
couple inputs provide an instrument range of 0-500 C, but the
measurements are only expected in the 15-110C range. By speci-
fying the operating range, the graphical use of the acquired data
is kept on a meaningful scale. GENESIS scales all data internally
for communication between blocks.
Following specification of the blocks, arrows are added be-
tween blocks, using the mouse, to specify the flow of information.
For example, in Figure 2 the reboiler level (REBLVL) is obtained
from the interface card (RTI815). At the time the arrow is con-
structed, windows pop up to facilitate specification of the input
channel of the card which carries the signal. REBLVL serves to
convert the raw signal to engineering units. OVERFLW provides
a digital signal to OVR, which communicates with the hardware
card.
TMPSET is an example of a block which does not obtain its
value from the interface card. The initial value of TMPSET can be
entered at the time the strategy is constructed, but the value may
be changed during runtime from the computer screen. This is
done by associating a particular cursor location on the control
screen with TMPSET during graphical construction of the control
screen. The graphics builder is used after the strategy has been
developed to design the computer-control screens which display
data from the strategy during runtime.
Careful design of the control loop and graphic display should
precede programming to minimize modifications. Alteration of
the strategy block connections or graphic display requires leaving
the runtime system and reentering the strategy-builder program.
Our system runs on an older 286 IBM AT which was slow in
loading the large strategy builder and in saving the compiled
control program "code." On the other hand, most of the param-
eters for the algorithm blocks may be changed during runtime,
eliminating the need for reentering the strategy builder if the
structure of the strategy is correct. This makes controller tuning
relatively simple. The revised parameters are stored on the hard
disk, and by simply renaming a file after completion of the run
the revised parameter file becomes the default file. The details of
some of the strategy algorithm blocks are not described com-
pletely in the GENESISdocumentation, but many of the capabili-
ties exceed the needs for most applications. In general, a phone
call to the company quickly resolves any confusion. GENESIS
offers a training course which is helpful in learning the capabili-
ties of the software. O










classroom


CHEMICAL ENGINEERING DESIGN

Problem-Solving Strategy


RICHARD TURTON, RICHARD C. BAILIE
West Virginia University
Morgantown, WV26506-6101

A significant fraction of our undergraduates ac-
cept industrial employment that requires deal-
ing with problems associated with an operating sys-
tem (production). These responsibilities depend on
an ability to both predict problems before they occur
and to diagnose problems after they occur, in addi-
tion to designing and developing new processes. Prob-
lems must be solved under the constraints imposed
by physical equipment and the existing processes.
Neither course work nor the typical capstone design
course makes a conscious effort to develop such abili-
ties in the student.
To address this problem, the chemical engineer-
ing department at West Virginia University has ini-
tiated a process in the senior design course that
focuses on the development of the necessary skills
for solving problems which production engineers may
expect to confront.rl,,31 The program introduces a
series of comprehensive problems, extracted from a
single process flowsheet, into the two senior design
courses. The problems are worked out individually
under the same rules that apply in the AIChE stu-
dent-contest problem wherein the student presents


Richard Turton received his chemical engineer-
ing degrees from Nottingham University (BSc,
1977) and Oregon State University (MS, 1979;
PhD, 1986). He is currently an Associate Profes-
sor at West Virginia University and his primary
interests lie in fluidization research and under-
graduate and graduate education.


Richard C. Bailie holds ChE degrees from Iowa
State University (PhD), Wayne State University
(MSChE), and Illinois Institute of Technology
(BSChE), and is currently a Professor at West
Virginia University. His interests are in fluidization
I and energy utilization, and he has published many
articles and a book in these areas. He is currently
working on a second book on the teaching of
design in the undergraduate curriculum.
Copyright ChE Division, ASEE 1992


TABLE 1
Unit Function: Intellectual vs. Industrial
UNIT INTELLECTUAL CONCEPT INDUSTRIAL APPLICATION
1 Visualization plus Analysis Trouble Shooting
2 Analysis plus Synthesis Retrofit
3 Synthesis plus Optimization New Process Design

a written report and a one-hour oral report on each
problem. The student is "escorted" through the tran-
sition from a detailed consideration of a single unit
operation to management problems that can include
several interrelated unit operations. This is accom-
plished by providing intensive individual and group
feedback after each assignment.
Major adjustments in student attitudes are some-
times necessary. It is difficult for the student to go
from solving narrow, focused, single-answer prob-
lems that involve no creativity to global, multi-
answer problems that require considerable creativ-
ity. They find that the well-entrenched strategies for
problem solving that worked for them in the past are
no longer effective.

DISCUSSION
The general content of the comprehensive prob-
lems are summarized in Table 1. It shows the skills
that are necessary for solving the problem and the
area of its industrial application. The order of the
problems is significant. The skills required in a pre-
vious problem must be mastered in order to solve
each new problem. Each problem develops a single
skill while introducing a new skill that is the basis
for the following problem. For example, the first
problem develops an ability to visualize-and intro-
duces analysis. The second problem assumes the
ability to visualize, develops an ability to analyze-
and introduces synthesis. In terms of industrial ap-
plication, the three problems will cover trouble-shoot-
ing, retrofit operations, and new process design.
The problems are introduced to the student


Chemical Engineering Education










through industrial "role-playing." The student (a nov-
ice engineer) joins a company and is assigned to a
plant site where three assignments are given that
cover the areas shown in Table 1. The student is
given a packet of materials that includes a process
flowsheet and flow table, a description of changes
that have been made to the process, thermodynamic
data, and various other items of information describ-
ing the process and the company that may be useful.
The process flowsheet is special. It represents a
poor process design (it was built many years ago)
under current conditions and includes several poor
design areas that could be improved. Some defects
have been distorted (like big ears and long noses) to
make them stand out and be easy to detect. The
special flowsheet represents a "caricature flowsheet"
and is an aid to the early development of visualiza-
tion skills. In addition to poor design, several proc-
ess modifications (operating conditions) have been
made in response to changes in constraints and
availability of new technology.
In the process of solving problems, students 1)
will identify all of the "caricature problems" and
offer solutions, 2) will identify many other problems,
not highlighted for identification, and offer solutions,
and 3) will create new processes that improve the
original process and eliminate the poor engineering
of the original system. All solutions must be justified
by economic considerations.
All three projects are associated with the produc-


The assignments and experiences of
our 1988 class are used here as an example of
how the comprehensive problems, along with
the "caricature flowsheet," are used to
introduce and develop skills.


tion of the same chemical commodity. Since each
successive problem requires mastery of the previous
problem, it has been necessary to provide extensive
individual and group feedback during the oral and
after each project.

STUDENT ASSIGNMENTS
The assignments and experiences of our 1988
class are used here as an example of how the com-
prehensive problems, along with the "caricature flow-
sheet," are used to introduce and develop skills. Fig-
ure 1 is a "caricature" flowsheet for the production of
acrylic acid. It was created from a 1986 AIChE stu-
dent contest problem on the production of acrylic
acid via partial oxidation of propylene.
A packet of material (about twenty pages) was
provided that contained information concerning the
instability of the acrylic acid with temperature (even
with the addition of an inhibitor, it will rapidly po-
lymerize at above 90C when present in large con-
centrations). Some additional information on cur-
rent operations is also given.


EQUIPMENT LIST
AB-1 Quench Tower
AB-2 Off Gas Absorber
C-1 Inlet Air Blower
E-1 Reactor Coolant Exchanger
E-2 Absorber Product Cooler
E-3 Solvent Separator Condenser
E-4 Solvent Separator Reboiler
E-5 Waste Water Separator Condenser
E-6 Waste Water Separator Reboiler
E-7 Product Separator Condenser
E-8 Product Separator Reboiler
E-9 Incinerator Waste Heat Boiler
H-1 Off Gas Incinerator
M-1 Solvent-Product Mixer
S-1 Solvent-Product Liquid Separator
T-1 Solvent Separator
T-2 Waste Water Separator
T-3 Product Separator
UTILITY LIST
BFW Boiler Feed Water
CW Cooling Water
CWR Cooling Water Return
CR Condensate Return
DI Deionized Water
NG Natural Gas
STM Steam


Figure 1. Acrylic Acid Process Flowsheet


Winter, 1992


To Stack STM


Acrylic Acid










ASSIGNMENT 1

The plant faces a serious problem. Our largest
customer cancelled its order. We shipped off-spec
material (a slight discoloration in the normally clear
acrylic acid). It damaged their process and their
product was worthless (colored). The potential loss
to our company is in excess of $500,000/day, and this
is the economic incentive that calls for an immediate
resolution. The assignment, to be completed in
one week, is 1) to identify the cause of the off-spec
material, 2) to recommend action to be taken to
correct the situation, 3) to present a written report,
and 4) to make an oral presentation to the company
president.

There are several potential explanations for the
"off-spec" material. There is evidence to support some
of the explanations, while others are no more than
speculation. Table 2 gives a partial list of some of the
explanations. They can be separated into three
groups: 1) global, 2) numerical, and 3) research. An
example from each group follows and possible solu-
tions are included in Table 2.

Global problems require that the student con-
sider items outside of the process unit for an expla-
nation. For example, the off-spec material may have


resulted from low-
quality feed. There
is evidence in the ma-
terial packet given to
the student that sup-
ports this explanation:
the amount of feed
has increased (feed
may not be pure pro-
pylene); the amount of
steam generated in the
waste heat boiler in-
creased (the non-pro-
pylene material in the
feed may not react in
the reactor and burns
in the afterburner); the
non-propylene mate-
rial was colored.

This problem was
intentionally set up for
the student to dis-
cover. Poor feed is, in
fact, the most common
cause of an off-spec
product. We shipped
our customer tainted


feed and ruined their product. This could be the
same problem we have. We could have received
tainted feed material. We have evidence to support
this explanation. Students not only fail to make a
connection between these two situations but they
also are not ready to accept the explanation.
A limited number of explanations can be verified
numerically (the purpose of this problem was to limit
the number of necessary calculations). The refriger-
ated water (at 100C) to E-7 (Figure 1) was replaced
with cooling water (at 300C). Many of the students
noted this and suggested that it might have caused
the temperature of the acrylic acid to reach +900C,
causing polymerization. When asked to determine
this temperature, many of the students had a vague
idea of how to approach the solution, but only one in
twenty was able to fully analyze the problem and
calculate the temperature. The level of analysis re-
quired to solve the problem is similar to that used by
a plant engineer. We expect our students to under-
stand and be able to employ such an analysis.
Figure 2 gives the temperature-duty profile for
the design case (all this information was available to
the student). The temperature of the cooling water is
known and is plotted on the same figure. For the
same recycle ratio, the condenser duty is constant.


TABLE 2
Possible Explanations for Off-Spec Product


Possible Cause
PRE-PROCESS
Feed is not pure





PROCESS
Reactor system


Separation systems
1. High temperature in
purification columns
2. Absorber malfunction

POST PROCESS
Product storage

Product transportation

Customer error
OTHER
Product not discolored


Evidence

1. Increased feed rate to maintain output
2. Increased steam production in waste
heat boiler
3. New supplier
4. Temperature and pressure of storage
vessel has changed

1. Increase in feed rate and lower
conversion
2. Lower selectivity

Change from refrigerated to cooling water
in condenser E-7
Increase in feed rate and steam production
in waste heat boiler.

Inhibitor not added to product in storage

Dirty tank cars or product exposed to air
and sunlight
Mishandling of material by customer


Check analytical equipment, testing
procedures, and records


Solution

Check feed for impurities
and return to the supplier





Replace spent catalyst


Return to refrigerated water

Increase the flow of water
to absorbers.

Check inhibitor and
injection system
Check transportation
company

Rerun samples in our lab
Rerun samples in our lab


Chemical Engineering Education


Chemical Engineering Education










To achieve a constant duty, Q = U*A*T Where A
constant and U is assumed constant, the value of'
must be constant. If the curves for the refrigerate
water coolant are shifted upward until the low tel
perature line corresponds to cooling water, the pr(
ess temperature at the top becomes 460C. This is t]
equilibrium temperature for the exiting acetic aci
Figure 3 provides the vapor pressure curve f
the distillate and the bottoms that were obtain
from the vapor pressures of the pure component
(and Raoult's law). From Figure 3, for a temperatu
of 460C the condenser pressure is 6.2 kPa. The pre
sure drop in the column is largely the result of t
liquid level on each plate and the number of platE
Since this remains constant (at 15.9 kPa), the b(
tom pressure is 22.1 kPa. From Figure 3 this corr
spends to a temperature of 950C. This is well ov
the 90'C limit for polymerization.
Some students elected to look for informati
outside of the material packet that was provided ai
came across the information that iron enhances p
lymerization. Iron may have come in with the fe
and been the cause of the off-spec material.


DUTY (Q)

Figure 2. Temperature vs duty diagram for product
overhead condenser


2 \ Acetic Acid Product
-( 95% pure)
1016.2 kPa


SAcrylic Acid Product
1.0 (99% pure)
0.
o 1.0I

Top Product Bottom Product
(at 46"C) (at 95C)
0.1 \
20 40 60 80 100
Temperature (OC)

Figure 3. Vapor pressure curves for acrylic and acetic
acid product streams.
Winter, 1992


is
r10
T In
ed
m-
)c-
he
d.
or
ed
its
re
8s-
he
es.
ft-


Student performance on this assignment was
poor. Unfortunately, once a single explanation was
identified, most students ended their search. No con-
sideration was given to the economic loss of $500,000/
day when they made their recommendations. Rec-
ommended changes were sequential, and no thought
was given to making simultaneous changes to save
time and money.
Assignment 2
The plant was returned to the original feed mate-
rial, refrigerated water was returned to condenser
E-7, and the catalyst was replaced. The high product
quality returned and the immediate problem was
solved.


The next assignment was to improve the plant
le-
profitability. There are recent changes in the value
of steam and credit for excess steam will no longer
be given.
on The first step was to identify the major costs
under the current conditions or base case. Figure
,o-
d- shows the costs of feed materials and utilities tha
ed
were calculated from information provided. It iden
tifies the major costs and shows that the overwhelm-
ing cost is propylene. There was no information on
S the catalyst that had been replaced. A recommenda-
tion that serious consideration be given to the devel-
S opment of a new catalyst with higher conversion and
selectivity is appropriate.
The next largest cost is steam. If large savings
are to be realized, the purchase of steam must be
curtailed. Two methods of reducing the steam costs
could be reducing the use of steam or generating
steam in the process. The acrylic acid reaction is
exothermic, and the heat generated is removed with
cooling water in E-1 (another example of the carica-
ture nature of the problem). Replacing E-1 with a
steam boiler eliminates the need to purchase steam.


Cost of Utilities Cost of Raw
Materials

Figure 4. Operating costs for the design case.


Cooling Water (New Conditions)



Refrigerated Water (Design Conditions)
Refrigerated Water (Design Conditions)









Table 3 lists some of the options for using the steam
produced to reduce the costs of utilities shown in
Figure 4.
Student performance on this assignment im-
proved dramatically from the first assignment. While
the assignment was more difficult and the changes
were subtle, the students uncovered and corrected
many process changes that improved profitability.
The changes were ranked by incremental return on
investment.
Assignment 3
This assignment requires the design of a new
plant based on new technology. The students are
free to be as innovative as possible under the given
structure. There are new catalysts that provide very
high selectivity at low conversion; they are tempera-
ture sensitive and ideal for a fluid bed. The single
reactor system is replaced by a two-reactor system
that can operate at different temperatures and pres-
sures. The students are provided with detailed ki-
netics along with a computer program that considers
both the kinetics and a fluid-bed model. This pro-
gram allows computation of the reactor size (given
the conversion and operating conditions). It is clear
that the new catalysts provide increased selectivity
and, with a recycle stream, the possibility of high
overall conversion.
Even though the students are provided with the
computer program, they will likely fail if they pro-
ceed to obtain the optimum design by a random
approach. Using the kinetic expressions or perform-
ing a few well-chosen simulations will provide a feel
for the effects of operating variables. Solely from the
reaction kinetics point-of-view, astounding savings
are possible by increasing the recycle. However, once
the cost of compressing the recycle gas is considered,
the advantages rapidly diminish and a sharp opti-
mum is reached. In the old plant without recycle,
large amounts of steam were introduced into the
reactor. With the recycle stream this is not neces-
sary.
The students did well on this assignment. Figure
5 is a typical flowsheet for section -AA- of Figure 1,
using new technology. It is noticeably different from
the old flow diagram. The plant costs are estimated
along with the predicted return on the investment.
CONCLUSIONS
Teaching the student about acrylic acid is not the
objective of this course. It is only a vehicle to demon-
strate how to approach a complex open-ended prob-
lem. While it is true that every process is different, it
is equally true that the methodology of visualizing,


analyzing, and synthesizing are universal and trans-
ferrable to other processes.
The goal of these problems is modest. The stu-
dent is not expected to master the strategies needed
to be effective in working problems of the type out-
lined. We only hope to introduce the student to these
types of problems and to move them to the break-
point on the learning curve where the rise becomes
rapid with a little more experience. We feel that if
we have accomplished this, the process has been
successful. The final conclusion deals with student
motivation. Motivation provides both a high driving
force and a low resistance to learning, enabling the
student to learn more rapidly. One student's view of
the sequence of problems is as follows:
On September 15th, 1988, we were presented with our
first in a series of design projects. It was apparent that our
experience in the classroom had to be challenged by the
real-world situation. Instead of performing a "cookbook"


TABLE 3
Possible Uses for Excess Steam in the Acrylic Acid Plant

1. Place a trim cooler in series with the waste heat boiler and
produce just enough steam for the process.
2. Use any excess steam in a turbine to power the feed air
compressor and reduce the overall power consumption in
the plant.
3. Place a preheater on the feed or air streams into the
incinerator and use excess steam to reduce the consump-
tion of natural gas.
4. Use incinerator effluent to preheat incinerator feed streams
and a trim waste-heat boiler to make steam balance within
the plant.


C-2


Steam


Figure 5. Modified section (detail -AA-) from Figure 1.
Chemical Engineering Education









type problem, we were supplied with a minimal amount of
information. Assumptions, answers and suggestions for
rather vague situations had to be made. The time to test our
knowledge in a real world situation was suddenly upon us.
As the semester progressed, many hours of sleep were
lost, many times tempers became short, and often the most
docile student exhibited impatient behavior. The effort put
forward was not wasted though; the increase in knowledge
greatly outweighed the amount of time put forth towards
the projects.
It is for the above reasons that I am pleased to present
this final in a series of three design projects. The knowledge
which has accumulated over the past two semesters will be
of great importance in the years to come.
An expanded version of this paper and copies of
the original problems are available from the au-
thors.
ACKNOWLEDGEMENTS
The authors would like to acknowledge that the
concept of these individual projects or "majors" has
been in existence at WVU for almost twenty years
and that many faculty have contributed to its suc-
cess. Most recently, we would like to recognize the
contributions ofA.F. Galli and W.B. Whiting.
REFERENCES
1. Bailie, R.C., "The Evolution of a Design-Based Curriculum in
Ch.E.," Eng. Educ., July/Aug (1990)
2. Bailie, R.C., "Mining the AIChE Contest Problem for All It's
Worth," Proceedings of the 1990 ASEE Annual Conference,
Toronto, Canada; p. 506
3. Turton, R., and R.C. Bailie, "The Senior Design Problem in
Chemical Engineering: A Programmed Approach," presented
at the 1989 AIChE Annual Meeting, San Francisco, November
(1989) 0


REVIEW: Chemical Kinetics
Continued from page 33.
Chapter 5 with catalysis. The former is conventional
in its coverage, while the latter introduces autocat-
alysis and oscillating reactions (a subject of much
current interest in chemical engineering) in addition
to homogeneous catalysis, heterogeneous catalysis,
and enzyme reactions. These first five chapters, two
hundred pages in length, comprise nearly half the
book.
Chapter 6, a brief discussion of the transition
from the macroscopic to the microscopic level, sets
the stage for the remainder of the text. It introduces
the description of reactive collisions in terms of colli-
sion cross sections and relates the cross section to
the rate coefficient. Also introduced at this point are
the quantum state description of reactions (state-to-
state kinetics), microscopic reversibility and detailed
balance, and a short discussion of the relationship
between macroscopic and microscopic kinetics.
Winter, 1992


Chapter 7 discusses the role of the potential en-
ergy surface in governing the outcome of reactive
events. Succeeding chapters take the microscopic
approach to kinetics, covering the kinetics of ele-
mentary reactions, and chemical dynamics. A treat-
ment of bimolecular collision dynamics and molecu-
lar beam scattering is followed by another chapter
on experimental methods, this time for state-to-state
kinetics. Chapters on transition-state theory (includ-
ing variational transition-state theory) and unimol-
ecular reactions are followed by one on dynamics in
solution and at interfaces. Chapters 13 and 14 cover
the advanced topics of the information-theoretic ap-
proach to dynamics, including surprisal analysis and
the master equation. The book concludes with a chap-
ter on applications: atmospheric chemistry, and hy-
drogen and methane combustion.
This book has been used for the past two years in
a one-quarter course given to chemical engineering
graduate students at the University of Minnesota.
In the ten weeks of lectures it has not been possible
to cover the entire book. The chapters on reactions
in solution and dynamics in solution, catalysis (which
is covered in another course), information theory,
and the master equation have been omitted, and
supplementary material on bimolecular collision dy-
namics, chain reactions, and the kinetics of combus-
tion reactions has been incorporated. It has proved
to be a satisfactory text, treating the subject at a
level suitable for graduate students and giving more
comprehensive coverage (particularly of modern de-
velopments) than many other texts.
For undergraduates, portions of the book could
be employed as the basis for a series of lectures on
kinetics that would introduce students to material of
interest to engineers in a more modern vein than
most chemical engineering kinetics texts. For ex-
ample, at Minnesota juniors are given approximately
six weeks of lectures on kinetics during the spring
quarter. The first five chapters of this book, plus
some selected material on rate theory from the chap-
ters on collision dynamics, transition state theory,
and unimolecular reactions (and perhaps atmospheric
chemistry and combustion reactions), would provide
good reading material to accompany the lectures.
Students mastering this material would have an
excellent foundation for future work in kinetics and
reaction engineering.
Chemical kinetics in an enormously broad area.
It is difficult to find a text that gives good treatment
of the fundamentals of the subject, as well as cover-
age of subjects of interest to chemical engineers.
This book is one that succeeds and can be recom-
mended. O










I O classroom


EDUCATION IN PROCESS SYNTHESIS

Application to Inorganic Processes


J. M. GUTIIRREZ, J. GIMtNEZ,
M.A. AGUADO


Universidad de Barcelona
08028 Barcelona, Spain

T he first task in process design is selecting
the raw materials and the reactions and opera-
tions needed to manufacture the desired product.
Process synthesis accomplishes this task by gen-
erating different process alternatives and selecting
the best ones.
Taking into account the void existing in the lit-
erature about inorganic process synthesis, this pa-
per is an attempt to show, in a simple and summa-
rized manner, how the principles of process synthe-
sis can be applied to inorganic processes. These prin-
ciples can also be applied in other fields of chemical
engineering.
Process synthesis uses heuristics, or "rules-of-
thumb" based on experience, for the generation and
selection of process alternatives. Education in proc-
ess synthesis implies that students will acquire, in a
systematic way, the necessary experience to enable
them to apply heuristics, or even to "invent" them
when the situation demands it. Douglastl'21 pro-
pounded a systematic application of heuristics to
process synthesis of organic processes and also indi-
cated the possibility of using the procedure in chemi-
cal engineering instruction.
The application of process synthesis methods to
inorganic processes requires specific considerations
due to the different chemistry and properties of inor-
ganic compounds. However, it seems possible to give
heuristics which will permit the student to acquire
systematic knowledge about the processes.
Raw materials are very important in inorganic
processes. Natural compounds are the raw material
for one (or a very reduced number) of the derived
compounds of every element, and from this (or these)
all the other compounds of the element are obtained.
The natural raw material for an element and its
derived compounds is selected according to the ele-
Copyright ChE Division ofASEE 1992


ment concentration, ease of mineral extraction, ease
of separation of the desired compounds, associated
impurities, etc. The most frequent natural raw ma-
terials are sulfides, oxides and hydroxides, chlorides,
carbonates, and silicates.
The next step in process synthesis of inorganic
processes is the selection of the reaction path. Three
types of reactions can be considered in inorganic
processes: redox, displacement (exchange of ions be-
tween compounds), and change of crystal structure.
For redox reactions, the general rules are: oxida-
tions must be made at the beginning of the process;
reductions will, if possible, be the latest steps of the
process; and electrochemical reactions will be the
last step. The oxidant or reductor selection can be
made using Ellingham's (or a similar) diagram and
taking into account the price of the reactants and
the ease of reaction-products separation. Air for oxi-
dations, and H,, C, or CO for reductions, will be
considered first. Displacement reactions are closely


Josd M. Gutierrez is a professor of chemical
engineering at the University of Barcelona
(Spain), where he received his PhD in 1984. His
major research focus has been in the areas of
colloid chemistry, and in chemical processes,
including synthesis and safety.


Jaime Glmdnez is a professor of chemical engi-
neering at the University of Barcelona (Spain)
where he received his PhD in 1985. His major
research focus has been in the areas of kinetics
and catalysis, and in chemical processes, in-
cluding synthesis and safety. -



Miguel A. Aguado is an assistant professor of
Chemical engineering at the University of
SBarcelona (Spain). He received his PhD in 1990
from the University of Barcelona. His major re-
search focus has been in the areas of kinetics
and catalysis, and in chemical processes, includ-
ing synthesis and safety.

Chemical Engineering Education









related with separation steps and are generally used
when a precipitate or an easily separable gas is
formed. When a change of crystal structure is re-
quired, the corresponding reaction will be the latest
step in the process.
Following the scheme given for organic processes,
the next step in process synthesis is to decide the
species allocation, i.e., the route of each species
through the process. Thus, quantity and reactive-
ness of impurities permit one to decide if they are
processed or not. If there are ions in the solution,
every valuable product or by-product and every group
of species with waste destination and similar solu-
bility will constitute an outlet. Recirculation of the
mother liquor, with a purge to avoid the build-up of
impurities, can be convenient.
The general structure of the separation system
depends upon the phases leaving the reactor. In
every case, first separation is a phase split, and after
that the stream of each phase is driven to the corre-
sponding separation system (solid, liquid, or gas).
From each separation system, streams of different
phases are normally produced and driven again to
the corresponding separation system.
The same heuristics as in organic processes can
be applied when choosing the separation task. Crys-
tallization and precipitation operations, followed by
solid-liquid separations (sedimentation or filtration,
by example) are used in many cases. In inorganic
processes there are normally not a great number of
separations and the sequencing problem is not diffi-
cult. In all cases, heuristics such as the removal of
corrosive and dangerous species first, removal of the
most abundant species, making difficult separations
last, etc., can be applied.
The last step in process synthesis is the inte-
gration of operations. In the inorganic processes,
the same general criteria are valid as in organic
processes.
It can thus be concluded that the student's
education in process synthesis can consist of trans-
mission of the necessary experience together with
heuristics or rules-of-thumb. This instruction in pro-
cess synthesis can be an advantageous substitute for
the exhaustive descriptions of chemical processes
which is given to students in subjects such as indus-
trial chemistry.

REFERENCES
1. Douglas, J.M., Hierarchical Decision Procedure for Proc-
ess Synthesis," AIChE J., 31,253 (1985)
2. Douglas, J.M., Conceptual Design of Chemical Processes,
McGraw-Hill, New York (1988) 0
Winter, 1992


LETTER: Langmuir's Isotherm
Continued from page 23

And from Eqs. (4) to (6)


ej =Kjcj9free=Kcj 1- ei
V i=1 )


Kjcj K (Kici jcj
=Kc 1Kici) =K^


m
0 ici


Solved for 6
K.c.
0.- JJ
3 m m
1+ i (Kici)
i=1
The concentration ofj on the adsorbent is proportional to
the surface fraction occupied by j, so that

k.K.c.
j = k.jj J (7)
1+ (Kc,)
i=1
where k is another constant. This is of the form of Eq. (1),
with k. corresponding to a and the K to the b,.
This derivation also gives the same result as Langmuir's
for adsorption of a dissociating molecule, for instance, of
H2 as 2 H occupying two sites.
Langmuir-type adsorption is seen to result from "ideal"
thermodynamic behavior, that is, from the absence of any
specific interactions that would result in activity-coeffi-
cient corrections or call for the definition of additional
species. This is a more germane reason than are the ad-
sorption and desorption rates. That adsorption equilib-
rium has a basis in thermodynamics will be easier for the
student to accept and will convey better insight into why
many adsorbents behave in this manner, and why some do
not.
When I went to school, more years ago than I care to
count, textbooks still derived the mass-action law for reac-
tions with the kinetic argument of equal forward and
reverse rates at equilibrium-one of them even with the
reaction

Cu2+ +6NH3 < [Cu(NH3)6 2+

postulating a single-step mechanism with hepta-molecu-
lar formation and decay into seven fragments! Happily,
we have long outgrown such nonsense and have put the
mass-action law for reactions on a sounder basis. Is it not
time that we accord the Langmuir isotherm the same
courtesy?
Friedrich G. Helfferich
The Pennsylvania State University
1. Langmuir, I., J. Am. Chem. Soc., 38,2221 (1916)
2. For instance, see Fogler, H.S., Elements of Chemical
Reaction Engineering, Prentice-Hall, Englewood Cliffs,
NJ, p. 241 (1986) 0
51










Award Lecture


COMPUTING IN ENGINEERING EDUCATION

From There, To Here, To Where?
Part 2: Education and the Future*


BRICE CARNAHAN
University of Michigan
Ann Arbor, MI 48109

What have engineering schools/colleges been do-
ing about the remarkable computing develop-
ments chronicled in Part 1 of this lecture?*
With the arrival of the first campus computers in
the mid-1950s, many faculty researchers gravitated
to computing centers to solve their equation sys-
tems. The computer has been central to creation of
new research disciplines (e.g., numerical fluid me-
chanics) and reorientation of old ones (such as pro-
cess analysis and design). The availability of com-
puting resources has caused a major shift in the way
research is done.
On the other hand, it would be inaccurate to
claim that undergraduate education has been radi-
cally transformed by the computer. The impact of
computing on engineering education has been much
less profound than research. No paradigm shift away
from the traditional lecture-recitation-laboratory-tu-
torial format, with the textbook as principal instruc-
tional resource, has yet occurred. Nevertheless, there
has been a gradual infusion of computing work into
the curriculum since the early 1960s.
THE ACADEMIC COMPUTING ENVIRONMENT
The computing experience of our undergraduates
has been controlled to a considerable extent by their
access to computing facilities: mainframes for re-
search only in the 1950s; mainframes accessible for
instruction (initially with batch processing and later
with timesharing from hard-wired terminals) in the
1960s; mainframe time-sharing from remote termi-
nals, wide-area networks, and decentralized mini-
computer facilities in the 1970s; personal computer
clusters, local area networks, student-owned com-
puters, and network-accessible supercomputers in
the 1980s; workstations, heterogeneous internets,
distributed services (client-server model), intensely
graphical interfaces and applications, and parallel
Part 1 of this lecture appeared in the last issue ofCEE: Vol.
25(4), (1991), page 218.
Copyright ChE Division ofASEE 1992


processing in the 1990s. Every engineering school
now has computing facilities for all its undergradu-
ates, though the quantity and quality of the hard-
ware, software, and services vary widely from school
to school. A pertinent statistic"1: in 1989 there were
more than 200,000 university-owned microcomput-
ers in the US, and a significant fraction of them were
in engineering school clusters.
EARLY DEVELOPMENTS
In 1959, Don Katz (then chairman of the Chemi-
cal Engineering Department at Michigan) foresaw
the tremendous impact that computing would have
on engineering practice. He convinced the Ford Foun-
dation to support a feasibility study of broad-scale
integration of computer use into undergraduate en-
gineering curricula that would make recommenda-
tions, prepare teaching materials, and train faculty.
In a three-year period, over two hundred faculty
from nine engineering disciplines and sixty-five en-
gineering schools participated in the various activi-
ties of the Michigan project. They jointly produced
more than 120 completely documented computer
problems/programs for classroom use that were
widely distributed to other faculty.
My first contact with this project occurred in the
summer of 1959 when Don offered me a full-time job
with his project. My acceptance put my thesis on
hold and delayed my PhD by an unconscionable num-
ber of years. But I have never regretted that deci-
sion. It provided opportunities that I would not have
otherwise had and steered me toward an academic
career which has brought me much pleasure.
Principal recommendations of the project:
1. Train faculty to use computers
2. Provide ("free") time-shared computing services to all
students
3. Require a computer-programming course
4. Teach numerical and optimization methods
5. Integrate computing assignments into all engineering,
science, and design courses
6. Stress design-like (now called "open-ended") problems
throughout the curriculum
Chemical Engineering Education









Some of these recommendations are a bit dated, but
most are still on the mark.

WHAT SHOULD STUDENTS KNOW?
According to ABET guidelines, our current engi-
neering students must'2' 1) have access to computing
facilities, 2) use computers for technical calculations,
problem solving, data acquisition and processing,
process control, and design, and 3) be able to apply
digital computing techniques to specific problems in
a disciplinary program. In addition, chemical engi-
neering students must be able to 1) program in a
high-level language, 2) use software packages for
analysis and design, 3) use computing facilities such
as editors, and 4) simulate engineering problems.
My own view is that students should be able to
formulate algorithms, be familiar with a procedural
computer language, and be able to understand and
write simple programs in that language.
In addition, I believe that graduating engineers
should have competence in workstation-centered ac-
tivities having little to do with traditional program
writing-in particular
Electronic communication
Preparation of most written materials and reports with
editing, word-processing, page layout, or desktop docu-
ment preparation and publishing software
Drawing, drafting, plotting, and CAD tools
Spreadsheeting for technical and financial analysis
Direct access to engineering and other information
resources (e.g., property and equipment data bases,
engineering abstracts, library card catalogues, on-line
display of journal articles and technical archives)
Signal processing and statistical analysis of machine-
acquired experimental data
Symbolic mathematical software
Numerical methods and mathematical software
Powerful analysis simulation and synthesis packages
for solution of complex discipline-oriented engineer-
ing problems
Interpretation of computer-generated visual images as
important aids to understanding computed results,
physical phenomena, and engineering systems
Group project work in which the computer is a major
factor in effective collaboration
Computer-assisted instruction/training (multi-media)

Are faculty implementing such recommendations
so that our students graduate with computing skills
essential for successful careers? A recent (unpub-
lished) CACHE survey of three large petroleum/
chemical companies suggests that industry views
recent graduates as having acceptable-to-good com-
puting knowledge and abilities. My own view is that
while our students do graduate with significant com-
puting skills, we are closer to implementing the let-
ter than the spirit of even the ABET requirements,
Winter, 1992


particularly when it comes to integrating computing
work into our core engineering science courses.
BUILDING A NETWORKED INFRASTRUCTURE
I believe that a college- or university-wide net-
worked computing infrastructure is an essential in-
gredient if we are to be successful. Departmental
resources alone will be inadequate for the job that
needs to be done. Since both our college and depart-
mental administrations at Michigan are committed
to "computing" as a vehicle for significantly improv-
ing the quality of instruction, I would like to relate
some of what we have been, and are now, doing to
make this happen. The impetus came first in 1983
when our dean (now President of the University)
brought a group of faculty (myself included) together
to plan the "complete" integration of "computing"
into the life of the college (target date, 1993). After
lengthy argument and discussion, our principal con-
clusion was:
The personal computer/workstation, connected to the rest of
the world through a hierarchical, heterogeneous multi-
vendor network, will be central to the engineering comput-
ing paradigm well before the turn of the century.
We decided that a networked personal computer/
workstation infrastructure was the first prerequi-
site for effective implementation of the broad goal of
"computing in the curriculum."
In 1984, the college began to implement the plan
by
Starting to build a college network connected to, but
otherwise independent of, the university's central main-
frame computing facilities.
Creating a college network management structure that
would foster personal, academic, and research use of
the new facilities. This included a full-time network
staff and an Executive Committee representing the fac-
ulty. An "Applications Sector" concept for supporting
general-purpose commercial productivity and disci-
pline-oriented professional software (e.g., CAD, design,
control) was also put into place.
Putting a networked personal computer on the desk of
every faculty member in the college. The dean sent out
the word that electronic mail should be the principal
means of communication for the faculty and staff.
Putting in place a long-range financial plan. The col-
lege made a firm commitment to maintain state-of-the-
art facilities, replacing outdated equipment and soft-
ware with the current "best" affordable.
Since 1984 our network has grown to a size and
complexity that even we did not envision at the
outset. Here are some facts and figures on its cur-
rent status:
1.100 mbps FDDI optical-fiber token ring backbone
2. Two IBM Token Ring Lans interfaced to the backbone
3. Over thirty Ethernets (at least one in each engineering build-
ing) interfaced to the backbone









4. Three bridged Apollo workstation 15 mbps optical-fiber token
rings gatewayed through Ethernets to the FDDI backbone
5. Subsidiary LANS (e.g., office Apple Talk networks) gatewayed
to building Ethernets
6. About 2000 attached machines distributed as follows:
650 in 18 open (24 hours a day) computing clusters, princi-
pally for undergraduates
700 in departmental teaching laboratories and in research
laboratories (principally for graduate student and faculty re-
search)
650 on faculty and professional staff desks, and for college
and departmental administrative and clerical staffs.
The public workstations directly available to our
undergraduate students include the following mix:
125 IBM PS/2 (386DX and 386SX)
250 Macintosh II (various models)
140 Apollo workstations
50 DEC workstations
50 Sun workstations
20 Hewlett Packard workstations
20 IBM (RS6000) workstations

Because all of the high-performance workstations
use variants of UNIX for their operating systems,
the actual number of computing nodes available to
our students from these open machines is much larger
than 650. Computing tasks can be distributed over a
large number of network workstation nodes (e.g., 75
RS6000s, several hundred Apollos, many high-per-
formance computation servers) operating in client-
server mode. The network also supports a state-of-
the-art visualization laboratory.

THE FIRST COMPUTING COURSE
With the infrastructure in place, we began the
transformation of our traditional programming course
into a "computing" course to prepare our undergrads
for immersion in our networked computing environ-
ment. Jim Wilkes and myself are responsible for the
required computing course for all of our undergradu-
ates, including those in the Computing Science Divi-
sion of our Electrical and Computer Engineering
Departments. Our premises for switching from the
traditional programming course format were:
1. Engineering computing involves much more than tradi-
tional programming
2. The networked workstation will be central to engineering
computing in the 1990s; every student and practicing
engineer will have personal access to one
3. Central mainframes will function principally as informa-
tion servers to networked clients
4. The predominant operating system characteristics of the
future will: support multi-tasking; be multi-user and dis-
tributed; have windowed graphical user interfaces; sup-
port pointing and other non-keyboard devices.
5. Application programs will have a uniform interface across
hardware platforms with: common menu appearance;
common text and graphical element selection; on-line,
context-sensitive help.


6. The network will support workstation access to any net-
work file, information, computation server; there will be
a common file storage format for all servers.
7. Future workstations will support rapid and natural ma-
nipulation of text, graphics, sound, video, and other me-
dia.
8. Cooperative activity will be supported for both communi-
cation and computation (e.g., group report, data-base,
spreadsheet access/generation, design project work-even
homework).
We decided (after substantial and continuing con-
sultation with college faculty) to include the follow-
ing topics in the first computing course:
General computer "literacy"
Hardware and network facilities that will be used through-
out the student's academic career
*Basic productivity software that will be used later on,
regardless of discipline (word processing; drawing, draft-
ing, plotting; spreadsheeting; symbolic math; data-base
manipulation; data acquisition/analysis)
Programming in a procedure-oriented language
Solution of problems designed to expose students to topics
from different engineering disciplines
Participation in a group activity (e.g., cooperative design
and implementation of a problem solution on the com-
puter)
To make the balance of our activities a bit clearer,
the course syllabus is shown in Table 1. In truth, we
have not yet been able to include the last two topics
without consuming more than a fair (3 credit-hour)
share of the student's time during the term. It is
likely that we will eliminate some of the introduc-
tory material and include one or both of these topics
in future terms, since our students are arriving with
more knowledge of computers and more experience
in computer programming with each passing year.

IMPACT ON ENGINEERING COURSES
Almost all of our faculty have taken the produc-
tivity enhancing services of the networked personal
computer/workstation (mail, wordprocessing, desk-
top publishing, and even spreadsheeting and draw-
ing applications) in stride, much more easily than I
would have thought ten years ago. Our students
have taken to network services like fish to water-
especially the major productivity tools like word pro-
cessing, plotting, and spreadsheeting.
But what about the impact of computing and the
networked environment on engineering course con-
tent and teaching methodology? In upper-level pro-
fessional design courses the computer already has
had a pervasive influence. In effect, most of the
college's design and control courses have become
computer-aided courses. The computer has also had
a significant impact in our undergraduate labora-
tory courses, which now emphasize experimental


Chemical Engineering Education









design and data analysis.
Unfortunately there is still a reluctance by many
faculty to truly integrate "computing" into the middle-
level engineering science courses. Historically, our
computing activities in these courses have been simi-
lar to those at many other colleges-hit-and-miss,
with little course-to-course coordination. A few
courses, such as our senior/graduate numerical meth-
ods offering depended heavily on computing assign-
ments, while many courses ignored the computer
altogether or had only an occasional computer as-
signment (typically the writing of a program).
Around 1986 our department adopted a policy
that every core course would involve at least two
"computing" problems and at least one "open-ended"
problem. The computer problems started out as pro-
gramming assignments for solving assigned analy-
sis problems, but over the years, the trend has been
toward use of application packages for solving more
realistic (nonlinear) technical problems.
THE COMPUTER AS TEACHER
There is still a vast potential for enhancing our
engineering science courses with computer simula-
tions and computer-assisted instruction. Bob Seader121
points the way with the following comment:
The microcomputer is the most powerful learning device
since the printed textbook. If the computer is used prop-
erly, individualized interactive learning is economically
possible for the first time.
Substituting "networked workstation" for "mi-
crocomputer," I agree with Bob. The networked work-
station with high-quality color graphics (and in the
future, CD-ROM, video and sound) and access to a
variety of information servers has the potential of
being the near-perfect vehicle for bringing unique
"open-ended" problems to individual students who

TABLE 1
Freshman Computing Course Syllabus
2-hr
classes TOPIC
1 General introduction to computers, the PS/2
2 OS/2, PM interface, files
2 Word processing/Text Editing
1 Networks
1 Internet hardware/software, services
1 Electronic mail, on-line services of the university library
9 Programming language (FORTRAN-77 or PASCAL)
1 The Macintosh
3 Graphics (drawing, drafting, plotting, CAD) Claris CAD, Delta Gr
1 Non-numerical methods (sorting, searching, symbol manipulation
1 Symbolic mathematics Maple
2 Spreadsheeting, simple financial analysis tools EXCEL
1 Simple database tools
1 Data acquisition and analysis
27 TOTAL
Winter, 1992


make the technical decisions while the computer
generates alternative solutions and displays them in
easily understandable (principally graphical) forms.
Unfortunately, the creation of good instructional
software is very demanding of faculty time, student
manpower, and money. There is unlikely to be any
significant effort outside the university to generate
such software because there is little prospect of fi-
nancial return. And since the reward structure, at
least at research universities, discourages such ac-
tivity, there is little prospect of major faculty efforts
to develop instructional software except in associa-
tion with popular textbooks or when supported by
outside funding.
Two of our faculty (Scott Fogler and myself) be-
gan to work on instructional modules for personal
computers several years ago. Scott developed a set of
several stand-alone modules for his reaction engi-
neering course. They were well received by students
and have been acquired by many schools that adopted
his reaction engineering textbook. At about the same
time, several students and I developed a CAI
authoring package and several chemical engineer-
ing and numerical and optimization modules for
stand-alone PCs.
More recently, Scott received a NSF curriculum-
improvement grant and has undertaken develop-
ment of several (four to six each) instructional mod-
ules for introduction to chemical engineering ther-
modynamics, reaction engineering, separations, and
fluid mechanics. Many of these will be released to
the chemical engineering community through
CACHE in the next year or so.
In the meantime, I supervised development of a
system-management tool called MicroMENTOR that
supports networked delivery of all IBM-oriented soft-
ware (including Scott's modules) used in our
department. MicroMENTOR allows individual
faculty to install software and other instruc-
tional materials directly on network servers
without going through network administra-
tors, to control access to the materials by their
students, to automatically collect student com-
ments/criticisms, and to gather statistics on
user activity and software use. I believe that
the concept of departmentally-controlled in-
structional software embedded in a more glo-
aph bal environment of network services will prove
to be an effective way to deliver educational
software to our students.
Two other NSF-sponsored curriculum im-
provement grants promise to deliver very high










quality products that will be widely distributed
(through CACHE) at low cost. One project (co-di-
rected by Rex Reklaitis and Bob Squires at Purdue)
is developing comprehensive simulation models of
industrial processes.The simulations are very flex-
ible, involve professionally produced videos, and could
be used effectively in a simulation lab, design course,
or engineering-science courses.
The second project (directed by Bruce Finlayson
at the University of Washington) focuses on numeri-
cally intensive solution of general fluid mechanics,
reaction, and heat and mass transfer models: its
major features are extensive graphical input and
high-quality color display for the visualization of
computed results.
In addition, a few faculty scattered across the
country are producing very useful instructionally-
oriented software. I hope that the recent showcasing
of some of this software at the national AIChE meet-
ing in Los Angeles will be an annual event, giving
faculty an opportunity to demonstrate their pro-
grams/lessons to a nationwide faculty audience.

THE FUTURE
What about the future? The track record for long-
range predictions about computing is only fair, at
best. Nevertheless, I'd like to end with some predic-
tions about the future of computing and the impact
of computing on engineering education in the short
(five to ten years) term.

By 1996
1. Microprocessors will be the dominant computing technology;
differences in personal computer and workstation character-
istics (and probably technology) will virtually disappear. Most
functions of the simpler microcomputers will be incorporated
into a single chip.
2. Typical workstation memories will be 16 MBytes; typical
operating speeds will be 50-100 MIPS.
3. The dominant mode of professional and academic engineer-
ing computing will be networked and distributed with main-
frames functioning principally as large-scale networked infor-
mation servers.
4. Character-based applications will disappear; essentially all
applications will operate in graphical mode.
5. "Three-dimensional" graphical interfaces will be available (no
glasses). Flat panel displays will be widely available.
6. The visualization capabilities of the newest workstations will
be widely used for research and education.
7. Multimedia workstations will be both available and afford-
able for instruction, but only in limited numbers; high-quality
multimedia instructional software will begin to appear on
engineering campuses.
8. With the advent of digital high-definition television, there
will be a significant merging of video and computing technol-
ogy. Some campus workstations will have a camera and mi-
crophone as peripherals.


9. New authoring tools will make it much easier to create visu-
ally attractive and interesting interactive instructional mod-
ules (e.g., with animation and video). Standards will be in
place for capturing and encoding audio and visual informa-
tion. New image compression algorithms will reduce hard-
ware demands for visual-image storage. Hypertext (non-se-
quential information representation) tools will play an impor-
tant role in authoring software.
10. It will still be time-consuming and expensive for faculty to
design, create, and test discipline-oriented instructional soft-
ware. Some colleges will begin to recognize faculty-authoring
efforts as a valued contribution at merit salary time (but not as
criteria for promotion or tenure).
By 2001
1. Procedure-language programming will decrease in importance
and will probably not be required of students. Object-oriented
and visual-programming tools will fill the void left by the
decline of algorithmic languages.
2. Interactive instructional modules will have a significant im-
pact on the education and continuing education of engineer-
ing students and professionals.
3. Workstations operating in the 1000 MIP/100 MFLOP range
will be available on the desktop (and in a briefcase) for about
the price equivalent to a current high-end personal computer.
4. Students will take notes on pen-based notebook computers;
in-class transponders will allow immediate storage of figures,
assignments, etc., in these computers.
5. Network transmission rates will increase several fold to many
gigabit/sec. Much of the paperwork (class handouts, home-
work assignments, collections of assignments) associated with
current engineering classes will be handled electronically
using network services. (So far the computer has hardly pro-
duced the paperless society-but I think this will happen
anyway.)
6. A digital cellular wireless network will allow access to the
international Internet from virtually anywhere in the world.
7. Many technical publications will be electronically accessible,
with new mechanisms for copyright, royalties, etc., in place.
8. Spoken-language interfaces will begin to appear.
9. Parallel processors on a single chip will be common in per-
sonal workstations.
10. Microcomputers will compete effectively with supercomputers
in the solution of many computational-intensive problems.
Supercomputers will be able to maintain superior speeds, but
only by increased parallelism. They will remain, relatively
speaking, quite expensive.
11. Massively parallel machines will be available and widely
used for engineering research. Many will operate at Teraflop
speeds. Programming tools and application software will be
slow to develop, but in place and very effective.
12. Some faculty will still not be using computers, except grudg-
ingly (E-mail only?).
And finally, no one will remember these predic-
tions. I will not be asked to comment on them-and
a good thing too!

REFERENCES
1. Seader, J.D., "Education and Training in Chemical Engi-
neering Related to the Use of Computers," Comp. Chem.
Eng., 13, 377 (1989)
2. Seader, J.D., "A Brief History of Computing in Chemical
Engineering," first Katz lecture, Chemical Engineering De-
partment, University of Michigan (April 1990) 0

Chemical Engineering Education













AUTHOR GUIDELINES


This guide is offered to aid authors in preparing manuscripts for Chemical Engineering
Education (CEE), a quarterly journal published by the Chemical Engineering Division of the
American Society for Engineering Education (ASEE).
CEE publishes papers in the broad field of chemical engineering education. Papers generally
describe a course, a laboratory, a ChE department, a ChE educator, a ChE curriculum, research
program, machine computation, special instructional programs, or give views and opinions on
various topics of interest to the profession.
Specific suggestions on preparing papers.
TITLE Use specific and informative titles. They should be as brief as possible, consistent
with the need for defining the subject area covered by the paper.
AUTHORSHIP Be consistent in authorship designation. Use first name, second initial, and
surname. Give complete mailing address of place where work was conducted. If current address is
different, include it in a footnote on title page.
TEXT Manuscripts of less than twelve double-spaced typewritten pages in length will be
given priority over longer ones. Consult recent issues for general style. Assume your reader is not a
novice in the field. Include only as much history as is needed to provide background for the
particular material covered in your paper. Sectionalize the article and insert brief appropriate
headings.
TABLES Avoid tables and graphs which involve duplication or superfluous data. If you can
use a graph, do not include a table. If the reader needs the table, omit the graph. Substitute a few
typical results for lengthy tables when practical. Avoid computer printouts.
NOMENCLATURE Follow nomenclature style of Chemical Abstracts; avoid trivial names.
If trade names are used, define at point of first use. Trade names should carry an initial capital
only, with no accompanying footnote. Use consistent units of measurement and give dimensions
for all terms. Write all equations and formulas clearly, and number important equations consecu-
tively.
ACKNOWLEDGMENT Include in acknowledgment only such credits as are essential
LITERATURE CITED References should be numbered and listed on a separate sheet in the
order occurring in the text.
COPY REQUIREMENTS Send two legible copies of the typed (double-spaced) manuscript
on standard letter-size paper. Clear duplicated copies are acceptable. Submit original drawings (or
clear prints) of graphs and diagrams, and clear glossy prints of photographs. Prepare original
drawings on tracing paper or high quality paper; use black india ink and a lettering set. Choose
graph papers with blue cross-sectional lines; other colors interfere with good reproduction. Label
ordinates and abscissas of graphs along the axes and outside the graph proper. Figure captions
and legends may be set in type and need not be lettered on the drawings. Number all illustrations
consecutively. Supply all captions and legends typed on a separate page.


























if j 4,




















WNEEODS WORK.E

Long before most people even recognized the problems, CH2M HILL
was developing innovative solutions for the world's changing waste, water
management, energy, transportation and laboratory needs. It's a race
agaJnst rime now. But our over 40 years experience in environmental
consulting engineering makes it a race we're well positioned to win.
Challenging projects ... early significant responsibility the
opportunity to work with top professionals in a creative, yet stable,
environment ... there are many excellent reasons to begin your career
with CH2M HILL. Perhaps the most important is the role you'll have in
shaping the engineering for a new world.
With 60 offices heading more than 4,000 projects. CH2M HILL has
U.S. openings m the following disciplines: Civil Engineering (including
specialization in environmental, structures, water resources,
transportation, geotechnical, ports, harbors and hydraulics)
Mechanical, Electrical, Agricultural and Chemical Engineering
Geology and Geological Engineering Planning Sciences
Economics Computer Engineering
Requirements include a BS degree in engineering from an ABET
Engineering program. A Master's degree is preferred for most specialties.
As a member of our employee-owned corporation, you'll enjoy a
competitive starling salary, attractive bonuses and flexible benefits. We
invite you to learn more about CH2M HILL's current staffing needs by
sending your resume and salary history in confidence to. Staffing Manager.
CEE; 291, CH2M HILL, PO. Box 221111, Denver, CO 80222-9998
If you have a PC and a modem, find out more
about CH2M HILL and other opportunities we have
available Just dial (603) 432-2742. press "return"
.' twice, and enter "water" when prompted for a
password. An Equal Opportunity Employer.

SePure Challenge




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ET45PHACZ_NB969G INGEST_TIME 2012-02-17T16:31:52Z PACKAGE AA00000383_00113
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

CEE cHe mi caiJ! eng in eer. ~ eclucation z 0 j::: ,,c( u ::, a w C) z a2 w w z 6 z w 0::: 2 u. 0 z 0 in > c C) z ii w w z 6 z w _, ,,c( u J: u VOLUME26 NUMBER1 Thomas F. Edgar of the University of Texas, Austin AWARD LECTURE: Part 2 ... Computing in Engineering Education: From There, To Here, To Where? and .. CARNAHAN Ideas About Curriculum WOODS Moments With Mathematica BINOUS, McCOY ChE Design: Problem-Solving Strategy TURTON, BAILIE Computer Control of a Distillation Experiment LIRA More Applied Math Problems on Vessel Draining SOMMERFELD WINTER1992 Introducing High School Students and Science Teachers to ChE BAYLES, AGUIRRE Education in Process ~ynthesis: Application to Inorganic Processes GUTIERREZ, GIMENEZ, AGUADO Introducing Water-Treatment Subjects into ChE Education CACERES, BRUTTNER MONARDES, CONTRERAS G6MEZ-SILVA .... and ChEat INSTITUTO TECNOLOGICO DE CELAYA

PAGE 2

CHOOCAL F..NGJHW I HC EDUC,t,TIOH I ,lo I I, lo lo I I "' QUAITD.LY 1 SEE ATIACHED RATE "'"' ~- -c ---~---,.-CHDIICAL ElfCIHEERl!tG EDUCATIOH, I317. 0.e.iul En1111eerill1 llep,,rtaent Unlurl t o r rlortd C.inenllle Alchu flo r l -'l?#.1 ~ ..... .... _.. .. ,._ .. N .,.,,U O "'-O.e.icl En1lneerin1 Dhillon, Aaericn Soc:ietJ for En1ineerln1 Eduutlon, II DuPont Ci r cle, VHhin1ton, DC 200:,0 ASE 0.eaiul En1ineerl111 Dlvhlon, II DuPont Ci r cle, \l uhlngton, DC 20030 ,,_., _, bJ \I. F,hien, Olea En1 Dl!putaent, 319, UnheultJ of Florid. C.ine.vllle, fl. 3261 Cuoh C. Yocu, Che Ena De11rtaent, R OOII 317, U nh o f Fl or i d, Galnenllle, f'L 326ll 1 -.. "' --.. -~-... ---".. -. ,_., "-", __ ., ... ___ ",.. ,.. ___ .. -:....~--::::::::,~;-;:..-_-:;-..:.::.:: t ____ .. ... .....,_.,__.,.~-. c 1111u,c 1ono a teaoove --An NI ,oonHed to o wn er should JstH a1>ove I ~::.~..:_-:-,::"..=:;:-::;-.,. :=:-,!:.: : :...._-..: ~:.:,-:::.5::':'.:!~l.:::.. .,_..Ow<_ ~'""" ... _,,_ t .. ..,..,.~ .... "'c ... ,_ ($,,,..,.,._ .. ......,_, ,, ... ...--.. .. "" c,....,_ ... _.., ............ .. ,_ ., ......... .. .. ... .. .. .._..,_ ----"I c, .. .. .. .......... c-c..o .. ,.. .. o,-o.o_ .... ..._c ... .. ... .... ~""'""""'''' cC --h- 'cw,.., _..., ___ .. __ ..,.,_ lc'"'11fy1""1 the -ll'MIMby .,...t,onc--=1-c...,,pi.1 ~::::"::.::" -" 2]32 2080 --0--0"" 1982 "" 1 982 "' 2194 ,.,, '" -oI --02332 2080

PAGE 3

EDITORIAL AND BUSINESS ADDRESS: Chemical Engineering Education Department of Chemical Engineering U niversity of Florida Gainesvill e, FL 32611 FAX 9 0 4-392-0861 EDITOR Ra y W. Fahien (904) 392-0857 ASSOCIATE EDITOR T. J Anderson (904) 392-2591 CONSUL TING EDITOR Mack Tyner MANAGING EDITOR Carole Yocum (904) 392-0861 PROBLEM EDITORS James 0 Wilkes and Mark A. Burns Univers it y of Michigan PUBLICATIONS BOARD CHAIRMAN E. D e nd y Sloan, Jr Colorado School of Mines PAST CHAIRMEN Gary Poehlein Georgia Institut e of Technology Klau s Timmerhaus University of Colorado MEMBERS George Burnet Iowa State University Anthony T. DiBenedetto University of Connecticut Thomas F. Edgar University of Texas at Austin Richard M. Felder North Caro lin a State Univers i ty Bruce A. Finlayson University of Washington H Scott Fogler Unive rsity of Michigan J David Hellums Ri ce University Carol M. McConica Colorado State Un i versity Angelo J. Perna New J e rs ey Institut e of Technology Stanley I Sandler University of De l aware Richard C. Seagrav e Iowa State Un i versity M. Sarni Selim Colorado School of Mines James E. Stice University of Texas at Austin Phillip C. Wankat Purdue University Donald R. Woods McMaster University Wint e r 1992 Chemical Engineering Education Volume 2 6 Numbe r 1 Winter 1992 EDUCAT O R 2 Thoma s F. Ed gar, of the University of Texa s, Au s tin D EPARTMENT 8 Instituto Tecnologico de Celaya, Arturo Jim e ne z AWARD LECTURE 52 Comp ut ing in Engineering Education: From There To Here To Where? Part 2. Educa tion and the Futur e Br ice Carnahan CLASSR OO M 12 Mom ents with Mathematica H. Binous B J. McCoy 20 In tro ducing Water-Treatment Subjects into Chemical Engineering Education L. Caceres, E. Gruttner, V. Monardes R. Contreras B. Gomez-Silva 44 Chemical En gineeri ng D esign: Problem-Solving Strategy, Richard Turton, Ri c har d C. Bailie 50 Education in Proc ess Synthesis: Application to Inorganic Proc esses, J.M. Gutierrez, J. Gimenez, M.A. Aguado OUTREACH 24 Introducing H igh School Students and Science Teachers to C h emical Engineering, Taryn Melkus Ba y l es, F ernando J. Aguirre CLASS AND HOME PROBLEMS 30 More Applied Math Problem s on Vessel Dr ai ning Jude T Sommerfeld CURRICULUM 34 Idea s About Curriculum, Donald R. Woods LABORATORY 38 Computer Control of a Distillation Experim ent, Carl T. Lira RAND O M TH O UGHTS 18 How About a Quick One? Ri chard M F e ld e r 7, 23 Letters t o the Edi t o r 33 Book Review C H EM I CAL ENG I NEE RI NG EDUCATION {ISSN 0009-2479) i s published quarterl y b y th e C h e m ic al E n g in ee rin g Di visio n Amer i can Society for E n g in ee ring Education and is e dit e d al th e U ni vers ity of F l or ida. Co rr es pond e nc e regardin g e dit oria l matter, ci r c ulaUon and c hang es of addre ss s hould b e se nt to CEE, C h e mi c al E ngin ee rin g D e partm e nt U niv ers i ty of Florida Gainesville, FL 32 611. Co p yrig ht 1992 b y th e C hemical Engineering Division American Society for Engineering Education. The s tat e m e nt s and opinions ex pres se d in thi s periodical ar e th ose of the writ ers and not n ecess aril y tho se of th e C h E Di v ision ASEE which body a ss umes no re s pon s ibility for them Defective co pi es re pla ce d if notifi e d within 120 da ys of publi c ation Write fo r informal.ion on s ub sc ription cos t s and for back c op y c o s t s and availability. POSTMASTER : Send address c hang es ta GEE, C h e mi c al E ngin ee rin g D e partm e nt., U ni vers it y of Florida Gainesville, FL 32611. 1

PAGE 4

1461 educato r ) THOMAS F EDGAR of t he University o f Texas at Austin W hen Tom Edgar first joined the faculty at UT-Austin in 1971, he was given an excerpt from a 1966 speech by John McKetta, then D ean of Engineering, which summarized the expectations of a faculty member: He/she shou l d expect to write proposals for research, equipment, an d special projects ; lo publish articles reports papers, and books ; to keep up to-date in his/her profession field; to serve on councils, boards, and com mittees ; to maintain the best possible relations with alumni legislators and the business and industry of the region-in short lo be a responsible member of the community and to participate in many of its activities. But we all know that these many activities must never overshadow our great est concern-the student. If our responsibilities lo and concern for, the student ever become secondary, we will be violating the trust we accepted when we joined the faculty. In his twenty years as a faculty member Tom Edgar has pursued this creed excelling in teaching research administration and pro fessional service. The influences that shaped Tom's career began when he was a young boy growing up in Bartlesville Oklahoma ( which at that time probably had the highest number of chemical engineers per capita in B ecky, J eff, Do nna, and Tom on their favorite type of family v a cation. Cop y ri g ht ChE Di v i s i o n of ASE E 1 99 2 2 the U.S.). His academic instincts were honed in the eighth grade when he won the Oklahoma Spelling Bee and received an all-expense paid trip to the National Spelling Bee in Washington D C. ( this was his first national meeting! ) Then, when Sputnik was launched in 1959, thrusting America into the space race, the resulting heightened consciousness of science led Tom to be gin dabbling in various technical fields. Encouraged by his father ( a metallur gical engineer ) his mother ( a teacher and a housewife), and an older brother (an electrical engineer ) he undertook several science fair projects ( incl u ding one on fuel cells ) which whette d his appetite for science and engineering. As a high-school senior he won the Bartlesville Science Fair and a trip to the National Science Fair Ch e mi c al En g in ee rin g Education

PAGE 5

During his senior year, Tom's parents became concerned about his choice of a career, so he took a battery of special aptitude tests. He scored high on memorization and vocabulary (the spelling-bee in fluence!), but low on spatial reasoning-which indi cated he should not become an engineer. Perhaps organic chemistry would be a good field .. But Tom ignored the aptitude-test results, and in 1963 he enrolled in chemical engineering at the University of Kansas. He was influenced to con sider a research career in his second semester when his audition for a part-time job to become a radio announcer for a classical music station was un successful. Tom mentioned the audition to J.O. Maloney, who p,romptly arranged for a lab-assistant job with Russ Mesler making high-speed movies of nucleate boiling. At KU, Tom's only "B" was in transport phenom ena (darn that Bob Bird!); it is ironic that transport was the first course he was assigned to teach at Texas. He became heavily involved in campus activi ties while at KU, which presaged the future level of his professional service. Also of benefit were a num ber of summer jobs in industry and an NSF under graduate research project in process control. Tom has fond memories of his first experience with computer programming. The students in his material and energy balance course were told to write a computer program for flash vaporization (it erative calculation) within a two-week period of time, and it was suggested that they could learn how to write the computer program on their own. This effectively forced Tom to learn how to use the com puter! The result was that computer decks took up more than their share of his desk space during the next ten years. With this background, it was only natural for Tom to study for the PhD under Leon Lapidus at Princeton, who was probably the leading authority on numerical analysis and optimal control at that time. Since Tom had a NSF fellowship, Leon left him alone to "do his thing," so in his free time Tom became involved in several athletic clubs and even tually started as prop for the Princeton rugby team for two years. Tom's PhD dissertation dealt with the "minimum time" control problem, which is consistent with his mother's comment that he always wanted to be first to finish any activity (unusual for a second child). He managed to finish his PhD in the "minimum time" (less than three years) even though he was balancing child-care responsibilities (his wife Donna Wint e r 1992 was pursuing a Master's degree at Rutgers), athletic pursuits, and research activities. STARTING A CAREER One of the worst times to graduate was in 1971there were probably only three or four academic jobs available that year, and industry was not hir ing. As a college freshman, Tom had decided he Tom's research in coal gasification and combustion focused on developing fundamentally based mathematical models in a field noted for empirical approaches wanted to be a professor, so he was delighted when the University of Texas offered him a faculty posi tion with a princely start-up package of $10,000. At that time there was limited government research funding (es pecially in control), few PhD students, and a teaching load of four courses per year. Tom set about to identify fundable projects where new mod eling and control techniques were needed to solve the problems. During the early 1970s Tom decided to broaden his background by exploring energy technology. He was influenced by having an office next to John McKetta who was at the time gaining great atten tion for his views on the energy crisis. When Tom came across an obscure reference to underground coal gasification (UCG), he became interested in pur suing research on the problem. As he recalls, the brief article stated that UCG was an economically attractive synfuel alternative, especially for deeper coal seams-but that a major difficulty existed in that the process could not be controlled and was not well understood. "A perfect application for modeling and control," he thought. Since the State of Texas was blessed with large reserves of deep lignite and was accustomed to drilling for energy, Tom put to gether an interdisciplinary research project involv ing chemical, petroleum, and environmental engi neering, in addition to geology, which was funded by NSF, DOE, and a ten-company consortium. Tom's research in coal gasification and combus tion focused on developing fundamentally based mathematical models in a field noted for empirical approaches. The general thrust of the UCG research was to determine those physical and chemical condi tions conducive to application ofUCG. Between 197 4 and 1984, the interdisciplinary group that he di rected developed computer-based methods for scale up and experimental methods to characterize a given 3

PAGE 6

coal deposit for UCG. Mathematical models were verified in small-scale reactors, combustion tubes, and at the field scale. These techniques were used by several oil companies and by government laborato ries and included computer models and experimen tal correlations for channel growth/sweep efficiency product gas composition, gas-solid reaction charac teristics, drying and mechanical properties, sulfur reactions, combustion tube tests, block gasification, flow characteristics, environmental impact and tech nical and economic evaluation. Many of these results, presented in some forty papers, have also been applied to conventional gas ification and combustion processes Of his papers on UCG, a 1978 AIChE Journal review article and a book chapter in Chemistry of Coal Utilization ( 1981 ), stand as key chemical engineering references in the field. Tom's research efforts were recognized by the AIChE Colburn Award in 1980; it was probably the only time this award has been given for "coal burning. While the UCG work focused heavily on model ing, Tom did not leave the control field during this period. His interests in multivariable control broad ened to include adaptive control in the 1970s. In 1977 he received one of two best-paper awards at the Joint Automatic Control Conference. His 1981 work with a PhD student, Ernie Vogel, on an adap tive dead-time compensator solved a long-standing problem inherent in many chemical processes and has been cited or used by a large number of subse quent investigators PROFESSIONAL INVOLVEMENT Tom's leadership in professional activities began when he and another engineering faculty member founded the minority engineering program at UT Austin in 1974. At that time there were few minori ties in engineering and no engineering scholarships for economically disadvantaged students at Texas. Tom helped get the recruitment program started established a tutoring program, raised $60,000 for scholarships from industry (a large sum in 1975! ) and served as director for two years. After a success ful start, the Equal Opportunity in Engineering Pro gram now has the fifth largest number of minority students in engineering in the U.S. Having learned how to balance a variety of out side activities with a heavy research and teaching load, Tom became active in a number of groups including the CAChE Corporation, the AIChE CAST Division, and the American Automatic Control Coun cil. He was selected by CAChE to edit the AIChEMI 4 Modular Instruction Series in Process Control and an issue of Computers and Chemical Engineering on the "Application of Computer Graphics in Chemical Engineering." These modules are still being distrib uted by AIChE and have been important suppleA bearded Tom with Dale Seborg at an authors' meeting in Kuwait in 1983. ments to the standard textbooks for process control. Tom also worked with Duncan Mellichamp to de velop the eight-volume CAChE Monograph Series in Real-Time Computing His success in leading a vari ety of CAChE projects led the other CAChE trustees to elect him as Vice-President ofCAChE in 1980 and as President from 1981-84 solidifying CAChE's fi nancial base so the non-profit educational group could grow in influence during the 1980s. Since 1981, Tom has served in a variety of posi tions in the AIChE CAST Division, and he thus provided important leadership as the division grew to the second largest division in AIChE These of fice s included Director Vice-Chairman and Chair man (in 1986 ) For the past three years he has been the AIChE Council Liaison to the CAST Division, having been elected a Director of AIChE in 1988. Since 1974, Tom has attended the interdiscipli Ch e m ic al Engin ee r i n g Edu c ation

PAGE 7

Tom's leadership in pr o fessional activities began when he and another engineering f a c ul ty member fo u nded the minority engineering program at UT-Austin in 1974 At th at time there were few minorities in enginee r i n g a nd no engineeri n g sc holar s hip s for eco nom ic all y d is ad v a nt a ged stu d e n ts at T ex a s nary American Control Conference every year and has seen it become the premier automatic control conference in the world during that time. His service activities in that organization include Arrangements Chairman (1974), Finance Chairman (1982), Pro gram Chairman (1979), and General Chairman (1987). He was selected as Director (AIChE repre sentative) of the American Automatic Control Coun cil (AACC) in 1982 and in 1988 was elected by repre sentatives of the six other sponsoring societies as Vice-President. He is just now finishing up a two year term as President and is leaving AACC in ex cellent shape for the future. Tom has also been a driving force behind the CPC (Chemical Process Control) series of confer ences and co-chaired (with Dale Seborg) the second conference in 1981. He served on the organizing committee and as session chair in 1986 and 1991 and was also the 1991 conference manager for CAChE (held at South Padre Island). Perhaps Tom is best remembered for his role as awards co-chairman in 1986 and 1991; CPC awards are like Harvard Lam poon awards and rely on a highly developed sense of humor. During the 1980s, Tom also took on a variety of editorial board activities. He and other colleagues founded a new journal on subsurface processing in 1978 (In Situ, published by Marcel Dekker), and he served as General Editor until 1988. In addition, he has served on a variety of journal editorial boards including AJChE Journal, Computers and Chemical Engineering, Chemical Engineering Reviews, Jour nal of Process Control, and most recently as process control editor of Chemical Engineering Education. He serves on the advisory board to the chemical engineering editor of John Wiley and Sons, and is also highly sought after as a program evaluator (Utah and Arizona State) and on departmental advisory committees (Kansas, MIT, and Maryland). BOOK-WRITING The decade of the 1980s also resulted in a flurry of books by Tom and his colleagues. In 1982 he had three book projects in progress. His first book, a professionally-oriented book on Coal Processing and Pollution Control, (Gulf Publishing, 1984), brought Winter, 1992 together a wide array of information on all aspects of coal utilization, covering extraction, conversion, and pollution control. Tom taught an undergraduate elective course on optimization since 1972, and as a result joined with David Himmelblau to write Optimization of Chemi cal Processes (McGraw-Hill, 1988). This book contin ued the important role that UT-Austin has played in this area of chemical engineering, beginning with Beveridge and Schechter's successful book (Optimi zation: Theory and Practice) in 1966. The recent textbook shows how the optimization field has ma tured in the past twenty years; its focus is on prob lem formulation (modeling) and it emphasizes only those optimization methods proven to be the best ones, while extensive coverage on various applica tions of optimization is also provided. Optimization of Chemical Processes is highly student-oriented in its presentation and sold nearly two thousand copies in its first year of publication. Perhaps Tom's best known book is Process Dy namics and Control, written with Dale Seborg and Duncan Mellichamp. This book had its origins in a five-day short course that Tom, Dale, and Duncan first gave at the University of California, Santa Bar bara, in 1978. It had a very long gestation period due to extensive class testing at ten universities (at vari ous times) changes in process control technology, geographical separation, extensive revisions by each of the three authors, UCSB faculty governance as signments, building a house, etc (the excuses and accusations are endless). In the California "spirit," Dale's philosophy was to "sell no wine before its time." Tom wrote most of the first drafts for the twenty-eight chapters, and Dale and Duncan then tore them to shreds (Tom's version). In any event, the book (pub l ished by Wiley in 1989) has receive d excellent reviews and is now the number-one text book on process control. It won the AS EE Meriam Wiley Distinguished Author Award as the top new engineering textbook in 1990. RECENT RESEARCH ACTIVITIES In spite of his many outside activities an d re sponsibilities as Chairman at Texas since 1985, Tom has managed to continue his strong research efforts 5

PAGE 8

and is currently investigating modeling and control applied to such diverse areas as chemical reactors, distillation columns, pressure swing adsorption, and most recently, microelectronics manufacturing. While he was a "lone wolf' in his previous research efforts, alliances with his UT colleagues Jim Rawlings, Ike Trachtenberg, and David Himmelblau have permit ted him to continue supervising a large number of graduate students. Tom still does much of the pro posal and publication writing and his students have learned to expect many corrections on their writing efforts. During his twenty-year career at Texas he has supervised forty-one MS and thirty-seven PhD students, which is due in part to his many good ideas and effective salesmanship techniques. Tom's laboratories contain a wide array of highly instrumented equipment, always connected to some computer (PCs, workstations, commercial distributed control systems). Asked about recent technical achievements, Tom lists several examples : demonstrating high sensitivity of some nonlinear con t rol schemes (with Chuck Alsop) developing theoretically-based criteria for measure ment selection in distillation columns (with Wayne Bequette) developing a new computationally efficient scheme for using nonlinear programming to compute model predictive controllers and demonstrating it on a packed dis t illation column and packed-bed chemical reactor (with Jim Rawlings Ashu Patwardhan, John Eaton, and Glenn Wright) validating a fundamental reaction-transport model for a commercial multiwafer low pressure chemical vapor deposition reactor, the first highly documented modeling study on polysilicon in the industry (with Tom Badgwell and Ike Trachtenberg) Over half of his current research group is focus ing on the modeling and control of microelectronics processes, with support from NSF, the State of Texas Sematech, Semiconductor Research Corporation, and Texas Instruments. Tom says the state of control technology in the microelectronics industry today is the same as it was for the chemical industry in 1970 and thus presents many opportunities to have an impact. His goal is to see model-based feedback con trol developed and implemented for a wide variety of unit operations, mainly in etching and deposition reactors. Tom believes strongly in teaching excellence; all of the books he has written have arisen out of supple mental materials developed for his courses ( he still can't believe how he had the time to teach four different courses for seven years ) Students appreci ate his sense of humor, his gregarious manner, and 6 his genuine interest in them as individuals. Because of his excellent memory, he is able to learn all of his student's names in one week, aided by his semi Socratic classroom pedagogy. Students like the fact that the door to his office is always open, even though he is busy as department chairman and as an advi s or to many graduate students. Tom has been extremely effective in influencing a rise in national visibility for the Department of Chemical Engineering at UT-Austin. He served as Department Graduate Advisor from 1979-1985 a time when PhD enrollment tripled. UT-Austin now ranks third in the U.S. in PhDs produced and is also third in research funding. The process control group is the largest of its type among chemical engineering departments. While the UT research focus has re cently become more fundamental and science-ori ented, it is still held in high esteem b y industrial practitioners, as is evidenced by the latest poll by U.S. News and World Report. Consistent with his personal juggling act Tom recognizes that the Department must have an atmo sphere where good teaching and good research are valued and where the students will receive a qual ity education. Evidence of this at Texas are the engi neering teaching awards won by Presidential Young Investigators and the Outstanding AIChE Student Chapter Award in 1990 ( the last time they won was when Tom was student chapter counselor in 197 4 ). Tom also places a high priority on alumni relations and interactions with industrial supporters. As Chairman Tom spends a good part of his time in setting new directions for the department and in young faculty development Adam Heller ( who came to Austin in 1988 from Bell Labs ) states "Tom Edgar s leadership has allowed the Department to expand beyond its previous national strengths in polymers, separations, and process control into two new ar eas-microelectronics and biotechnology." Jeff Hubbell comments, Tom Edgar spends considerable time so that young faculty have the opportunity to become successful." His friend Bob Seader ( at Utah ) sums up Tom s impact on chemical engineering as follows: Tom is one of the best known and most well-respected young engineering educators in the United States. During a period of twenty years, he seems to have had one major goal: to do everything he can do to advance his profession and to help others to best utilize these advancements He is a tireless worker of almost unlimited energy, an inspiration to his many col league s and a servant to his profession." 0 Ch e mi c al En gi n ee ring Education

PAGE 9

I iillldliia..:.:le::.:t~te::.:r...:t:.::o:..:t~h:!:e:..:e:;,:d~,:.:.:to~r:..,_ _______ ) E Q U ILIBRIU M T HERMODYNAMICS-REVISITED Dear Editor: I am pleased that Williams and Glasser 1 2 1 have out lined the course in thermodynamics that I introduced at the University of the Witwatersrand in 1982 and lectured there until the end of 1984. There are, however, a few points that I feel need clarification. One reason for the difficulties students often encoun ter in thermodynamics is the confusion over variables and functions. One encounters references in the same context to H, H ( T,P ) and H ( T,V ), for example, in which the single symbol H means at least three different things: e.g., a variable and two distinct functions. An important objective ofmy approach is to attempt to overcome the confusion to which this symbolism gives rise. The symbol for any function is always constructed so that it explicitly displays the independent variables (as superscripts) and the dependent variable. Thus the func tion from T and P to H is written W P. On the other hand, the value of the function at T and Pis written H TP( T,P ) One can write but not H=HTP ( at least for the present ). (1) (2) Williams and Glasser explain the symbolism and write statements like Eq. ( 1 ). However they then apparently break the rules in a number of ways: a variable is equated to a function and not its value (e .g ., S = STV in their Eq 19 ); the sa me symbol is used for a variable and a function (e .g. V = V(S)); they talk of "a new function A = U TS "; etc. Having warned the reader of the importance of not confus ing functions and values of functions they do just that throughout both papers. From the outset it is necessary to confine the use of the term function to things that are functions. A function always takes the form W P : HTP : T,P H (3) Then A may be a variable, but it is certainly not a func tion. The term state function is not used-param ete r or property is better. If one has particular values for the independent vari ables, then one can write (4) for exa mple Very often one does not in which case the presence ofT and P twice in Eq. (1) may appear unneces sary. When the students have become familiar with the notation, and confident in its use I draw attention to this apparent redundancy ( if th e students have not already done so). I then say that if th,e context mak es clear that the value of the function is meant, we can agree from now on, Winter, 1992 to an abbreviated notation : we may write H TP for HTP(T,P). The convention is that W P always means the function unless the context shows that it must mean the value of the function. In the latter case the ( T,P ) is to be under stood even though it is not explicitly displayed. Wherever there is any chance of confusion or doubt (T ,P) is not to be omitted. Now, and only now does it become legal to write such equations as Eq. ( 2 ) above and Eq (1 9 ) of the paper by Williams and Glasser. I found it necessary to resort to the sort of fuss and circumspection resorted to here. If we are taking care to distinguish functions and their values, we need to do so with derived functions or deriva tives as well. Thus ausv or us'v as is a derived function and ausv s v ~(S,V) or U (S,V) its value (t he above convention still applying). It now becomes possible to agree with Eq. ( 18 ) of Williams and Glasser T=us 'v = Tsv The context implies an abbreviation for T = us v (S, V ) = T 8 v (S, V) Williams and Glasser write ( their Eq. 25 ) .l_ (ausv J as av This violates the rules of notation just as much as aP as does. One needs to write _Q_ ( ausv Jsv as av (5) (6) (7) Otherwise one has no means of distinguishing the expres sion from _Q_ ( ausv JsP as av (8) for example. While the order may normally be reversed with impunity in Eq. ( 7 ) (t hat is, when the superscripts repeat ), it cannot in Eq (8) ( that is, when they do not). On the other hand a2usv a sav is unambiguous. There are hidden traps in Eq. (6) out of which I have resc u ed more than one student. St u dents easily obtain cross differentiation identities that are sim ply not true. My equation dHI _a HTP dT ~ dP = O (9) ( Eq 9 of Williams and Glasser ) I have preferred since Continued on page 29.

PAGE 10

i4di department Main entrance to the Institute. INSTITUTO TECNOLOGICO ) OCEAN PACIFIC DE CELAYA ARTURO J !MENEZ Instituto Tecnol6gico de Celaya 38010 Celaya, Gto. Mexico M exico has been trying to improve her scientific programs and her industrial productivity ever since the heavy economic prob lems that were experienced in the 1980s. Industry presently faces an important transitional period from simple technology users to one in which indigenous resources must be used more intelligently and local processes must be more fully developed. To meet this challenge, Mexico needs qualified engineers. The Instituto Tecnol6gico de Celaya, founded in 1958, is one of sixty technological institutes in Mexico, all part of a National Sys tem of Technological Institutes. The size of each institute has been kept under 3000 stu dent s. Celaya is a province of the state of Guanajuato and is located in the center of Mexico, about 150 miles northwest of the capital, Mexico City The chemical engineering department in Celaya has been recognized as one of Mexico 's most dynamic departments in recent years. Traditionally, few universities in Mexico have had faculties with a complete or nearly complete staff of professors holding PhD de grees, due in part to a lack of available professionals at that level. In 1980 only one faculty member in Celaya held a PhD degree. Today however, our faculty has expan ded to include seven members with PhD degrees ( see Table 1 ) The average age of the faculty is only thirty-seven, and their backgrounds include visiting appointments Copyright ChE D ivis i on ASEE 1992 8 THE UNITED STATES OF AMERICA C N IMl,l l Ml,1 1 MEXICO a uuncu G ................. .. GULF OF MEXICO Students controlling one of the reactors. Partial view of the ChE laboratory Chemical Engin ee ring Education

PAGE 11

at places such as the Universities of Wisconsin, Texas at Austin, California at Davis Texas A&M Univer sity, and CSIRO in Australia We hope to establish a solid faculty by sending our professors abroad for further graduate studies in the future, and possibly by adding new members to the staff. Eight faculty members are currently involved in PhD studies (see Table 2), and our eventual goal is to have a staff of fifteen professors with PhD degrees by the mid-point of this decade. THE UNDERGRADUATE PROGRAM The undergraduate program began in 1961 and has grown t~ a current figure of 300 enrolled stu dents (see Figure 1 ). Female students have been enrolling at higher rates in recent years; from six TABLE 1 1991 ChE Faculty, Instituto Techn6logico de Celaya Rafael Chavez (PhD, University of Utah) Alejandro G6mez (PhD, University of Utah) Arturo Jimenez (PhD, University of Wisconsin) Gustavo Iglesias (PhD, Texas A&M University) Alberto Ochoa (PhD, University of California, Davis) Pedro Quintana (PhD, University of Texas, Austin) Antonio Rocha (PhD University of Texas, Austin) Maria-Guadalupe Almanza (MS Tecnol6gico de Cel aya ) Carlos Cardenas (MS, Tecnol6gico de Celaya) Eleazar Escamilla (MS, Universidad de Mexico) Teresa del Carmen Flores (MS, University of California) Irma-Concepcion Galindo (MS, University of North Texas) Salvador Leon (MS Portland State University) David Trigueros (MS, Tecnol6gico de Celaya) Ma. de las Angeles Vazquez (MS Tecnol6gico de Monterrey) Angel Vazquez (MS, Tecnol6gico de Monterrey) Juan Ledesma (MS, Universidad de Guanajuato) Emma Torres (MS, Universidad de Guanajuato) TABLE 2 ChE Faculty Currently Pursuing Ph.D. Degrees Professor University Francisco Javier Alvarado Texas A & M University Cristina Coronado Universidad de Barcelona Guillermo Gonzalez University de Salamanca We try to give our students good exposure to laboratory activities with the result that the chemical engineering laboratory is used in three courses on unit operations in addition to one in kinetics and reactor design percent in 1980 to twenty-nine percent in 1989. About forty chemical engineers graduate each year from our program. The undergraduate curricula comprise typical chemical engineering courses together with addi tional courses emphasizing industrial engineering. We try to give our students good exposure to labora tory activities, with the result that chemical engi neering laboratory is used in three courses on unit operations in addition to one in kinetics and reactor design. The laboratory includes several units on a pilot-plant scale for operations such as distillation, absorption extraction, evaporation, and drying. There are also sections for chemical reaction, solids treatment, and transport phenomena, and students must take two courses on analytical chemi cal analysis. Recently we have also incorporated computer-aided techniques more significantly into our teaching. THE GRADUATE PROGRAM Our graduate program ( one of the youngest in Mexico) began in 1980 and offered an MS degree in chemical engineering. Initially the MS program con sisted of nine courses and a thesis project. There was a set of core courses on thermodynamics, applied mathematics, transport phenomena, reactor design and separation processes. Elective courses included process design, process control, design of experiments, Y earo f Termination 1992 1992 1993 400 300 i ,. ti! 200 '; 0 100 Gloria Maria Martinez Universidad Aut6noma Metropolitana 1992 Ramiro Rico Princeton University Julio Rocha Imperial College Fernando Tiscareno University of Wisconsin Rodolfo Trejo Universidad Aut6noma Metropolitana Winter 1992 1993 1993 1992 1992 0 80 81 82 83 84 85 86 87 88 89 90 Year FIGURE 1. Students enrolled in B.S. program. 9

PAGE 12

An interesting and vital challenge for us is to offer a solid PhD program that emphasizes research activities of interest to Mexico while at the same time keeping the solid technical basis that characterizes many of the graduate programs in the first-world universities. Operation of a double-effect evaporation system and selected topics on thermodynamics and trans port phenomena. We revised our program in 1985, giving more emphasis to research by reducing the academic load by two elective courses. This change yielded positive results. It shows the history of admissions and gradu ating students for the Master s program and demon strates that our graduating efficiency has improved significantly in the past few years. The low gradua tion rate noted in earlier years was not unique to our graduate program but in fact has been a problem of major concern to most Mexican universities offering graduate programs in chemical engineering. One system we have implemented for supervis ing our students research progress is to assign a committee ( in addition to the major advisor ) from the beginning of the thesis project. At the end of each semester, the student must present a seminar before the committee, and based on the progress reported, a grade for the research work is assigned by the committee. The student also benefits from committee feedback at each step. We have also made special efforts to recruit top students for our graduate program and this fact accounts in part for our recent good results Some of our MS graduates have gone for further PhD work at universities such as the University of Wisconsin University of Texas at Austin, Texas A&M Univer sity and Princeton University In Mexico graduate programs offering PhD de grees are new and at present there are only three 10 universities where one can be obtained. The Autono mous Metropolitan University ( since 1984 ) the Na tional Autonomous University of Mexico ( since 1988 ) both located in Mexico City and just recently the Technological Institute of Celaya. It was not until 1991 that the first PhD degree was granted by the Autonomous Metropolitan University. An interesting and vital challenge for us i s to offer a solid PhD program that emphasizes research activities of interest to Mexico while at the same time keeping the solid technical basis that charac terizes many of the graduate programs in the first world universities. For this challenge to be success fully met, we must build a proper infrastructure for research activities. RESEARCH ACTIVITIES As a natural consequence of our pioneering ef forts to establish a solid graduate program in Mex ico part of the job has been devoted to obtaining financial support for research activities a task that has been complicated by the economic situation in our country. We have, however made significant progress. Until 1982, federal support for res e arch activi ties was weak, but in spite of our economic problems the situation has improved steadily s i nce then. For instance, CONACYT ( the National Council for Sci ence and Technology ) initiated a special program ( Programa de Apoyo al Posgrado ) that offers support for improving the infrastructure at selected univer sities. Celaya has received support through this pro gram since 1984 and as a result our computing and laboratory facilities have expanded significantl y Support for individual research projects has also been received from COSNET ( the Council of the National System of Technological Education ) The support from these two agencies has particularl y influenced our growth. Our research programs have centered on com puter-aided design, experimental work on separa tion processes process control, and production of specialty chemicals The computing facilities in the department are based on a Micro Vax II s y stem which is also used as a network server for twenty PCs Some process simulators and software for process C h e m ic al En g in ee r i n g Edu ca t io n

PAGE 13

control studies are available. To support experimen tal work, we have some of the newer analytical equip ment such as GCs, atomic absorption, TGA, etc., along with the more conventional equipment. Some major results of experimental work con cerned two projects for the production of ferrous fumarate and oxalic acid. The work was undertaken due to an interest expressed by industry, and as a result there are now industrial plants for the pro duction of these two specialty chemicals. The eco nomic benefit for our institution and for the research team, however has been marginal (a fact that has produced strong criticism from our politicians!). The experience has been part of a learning process for us. These two projects are indicative of our efforts to bridge the traditional gap between academia and industry in our country. We have also provided serv ices to industry in the form of instrumental analysis and other specific research projects. In addition (an d taking into account our previous experience), we are exploring routes for the manufacture of s pecialty chemicals that are currently being imported by Mexico, and a multi-purpose pilot plant is nearing completion. Celanese Mexicana has recently given additional scholarships for four graduate students who will develop their research projects on topics of general interest to the company. OTHER ASPECTS AND OUTLOOK Professor J.M. Smith (University of California, Davis) was associated with our graduate program between 1984 and 1987. He twice taught a graduate course on reactor design and helped conduct a re search project on the thermodynamics of three-phase reactor systems. In our effort to stim ulate excellence in the department, in 1984 we instituted the J.M. Smith award, given each year to the graduate stu dent with the highest GPA. Seven students have received the award and four of them are presently pursuing the PhD degree. A simi lar award, given in honor of Professor Jose Martinez-Avella, one of the first lecturers in our institute, was instituted in 1989. The recipient of that award is the stu dent graduating from our under graduate program with the highest GPA. In order to stimulate research activities, in 1984 the Mexican government established a National Sys tem of Inve stigators. Eleven professors from the Celaya department have been inducted into the sys tem, the highest representation of any of the techniWint e r, 1992 cal institutes in Mexico. Each week a lecturer is invited to address the faculty and graduate students on some topic of inter est. On occasion the speaker will be one of our own graduate st udents talking about his or her research project. These seminars have been held since 1980. Each year for the past eleven years, we have organized a week of chemical engineering, the Semi nario Anual de lngenieria Quimica, an event that Students undertaking an experiment on chemical reaction analysis. has now become a tradition in Mexico. The main part of the event is a series of courses given by recognized professionals from both Mexico and abroad to an audience of national professors, graduate stu dents, and practitioner engineers. Among the past participants in this Seminario have been J. M. Smith, Octave Levenspiel, Charles Hill, Brice Carnahan, J. D. Seader, James Fair, Warren Stewart, Edwin Light foot, Carlos Smith, Richard Felder, Charles Holland and George Stephanopoulos. We also attempt to share nonacademic activities with our students. Thus, activities such as our "Thank God It's Friday basketball games have become tra ditional events enjoyed by everyone. Recent developments in Mexico seem to indicate that those who are in government positions are ac quiring a better awareness of the importance of aca demic and research activities. Industry, too, is show ing more interest in academia. Scholarship programs, such as CONACYT, are essential for providing greater numbers of qualified professionals who will be capable of meeting the challenges of the future. Consolidating a modern chemical engineering program in Mexico will not be an easy task, but those of us at Celaya hope to take a leadership role in effecting the needed changes. 0 11

PAGE 14

Mfii classroom ) MOMENTS WITH MATHEMATICA H. BINous, B.J. McCoy University of California Davis, CA 95616 M athematica, an interactive software system that does computer algebra and other mathemati cal computations, is a valuable tool for solving large problems involving symbol manipulation Mathe matica symbolically performs algebraic and calculus operations such as factorization substitution, linear algebra, solution of polynomial equations, limits, and evaluation of integrals and derivatives. Numerical computations and visualization of advanced func tions by twoor three-dimensional contour plots are also attractive features of the system. Although more efficient packages (e.g., Matlab) are available for large-scale numerical tasks such as matrix inver sion, Mathematica can algebraically perform ma nipulations of this type. Many potential applications of computer algebra systems in chemical engineer ing research and education are possible to imagine and in this article we discuss some examples. The developer of Mathematica, Stephen Wolf ram, is the author of a helpful manual called Mathe matica: A System for Doing Mathematics by Com puter ( Addison-Wesley Publishing Company, 1988 ). The book is readable and readily understood by those with a general familiarity of mathematics and com puters. Our experience is that studying the manual with the software at hand to work the numerous illustrations is helpful in learning the system. Op erations with Mathematica can also be programmed (e.g., do-loops for iterative computations ); see, for example, Programming in Mathematica .f1 1 Several computer algebra systems are available on the market and promise to automate mathemati cal computations Mathematica is particularly at tractive because it can be installed in a desktop computer, although speed is compromised. In its workstation configuration, however, Mathematica is fast and user-friendly. Computer algebra systems probably will radically change problem solving in applied mathematics; tedious mathematical compuMathematica is a Trademark name owned by Wolfram R esearch Inc. PO Box 6059, Champaign, IL 61821-9902 12 tations will be much less important than the funda mental problem formulation. Opportunities are plentiful in chemical engineer ing research and teaching for the application of com puterized symbolic manipulation. As an example of its use, we have applied Mathematica to the compu tation of moment expressions for several problems in separations and chemical reaction engineering. MOMENTS IN CHEMICAL ENGINEERING The computation of moments is useful for ( 1 ) interpreting experimental concentration profiles, (2) determining mass transport parameters from ex perimental data ( 3) predicting concentration histo ries, and ( 4 ) designing and scaling -up separation processes The temporal moments are given by mn(z)=(-l)n lim dnc / dsn = Jtnc(t z)dt (1) s O 0 where the Laplace transform of the concentration is defined by c(s z)= f c(t z)e s t dt (2) 0 Numerical or analytical integration provides values or expressions for the moments when c (t,z) is known or when c (s, z ) can be derived from a model of the Hausam Binous born in Tunis in 1964 re ceived a general engineering degree (with bio technology electives) from the Ecole Nationale Superieure des Mines de Par is in 1988 He re cently received his MS degree in chemical engi neering from UC Davis His research interests are in separation science and app lication o f com puters in engineering Ben McCoy received his BS in chemical engi neering from the Illinois Institute of Technology (1963) and his MS and PhD from the University of Minnesota (1964 1967) He has been at UC Davis since 1967 serving as Department Chair from 1980 -8 8 He is currently Associate Dean for Research in the College of Engineering Hi s re search interests are in chemical reaction engi neering and separation processes Copyrigh t ChE D ivision ASEE 1 99 1 Chemical En g in ee r ing Educat ion

PAGE 15

process. Thus, we avoid inverting c, a task that often is not feasible. The algebraic manipulations to obtain the moments from a mathematical model are: 1. solution of differential equations to determine c 2. derivatives with respect to the Laplace transform parameter s 3. limits as s 0 The development of c in a series provides an easy way to find the lower-order moments. Expanding e 1 in Eq. (2) leads to c(s,z)= :I sn(-ltmn(z)/n! (3) n=O The reduced moments are defined as (z)=m / m 0 n n (4) and the central moments are given by n(z)=(llm 0 )J (t-~r c(t,z)dt (5) 0 The binomial expansion of Eq. (5), written in terms of the binomial coefficients, yields n(z) = (( 1 / m 0 ) t (7)mn -i (-:r) (6) The first four moments ha ve an important physi cal significance. The zeroth moment, m 0 is the area under the concentration plot c(t) and is proportional to the mass of the species. The mass balance ensures that m/z) = m/z=O) for species that do not react. The first reduced moment 1 is the position of the peak and gives the average position of the species. The second reduced central moment, 2 measures the spread of the peak. The third reduced central TABLE 1 Hermite Polynomials and Coefficients of Gram-Charlier Expansion, Eq. (12) Ho(x) = 1 H1(x) = 2x H2(x) = 4x 2 2 H3 (x) = 8x 3 12x H4 (x) = 16x 4 48x 2 + 12 ao = Winter, 1992 a 1 = 0 a 2 = 0 a3 = a /(3!2 2 .,/;~) a4 = / ~ 3)/( ( 11) (12) (13) (1-4) (15) (16) (17) (18) (19) (110) For many cases, mathematical models provide the moment expressions in terms of mass transport parameters and geometric properties. From experimental methods we obtain the concentration c(t). Then the moments can be obtained by numerical integration ... moment, 3 is a measure of the skewness of the peak. For many cases, mathematical models provide the moment expressions in terms of mass transport parameters and geometric properties. From experi mental methods, we obtain the concentration c(t). Then the moments can be obtained by numerical integration using Eq. (1). Equating the experimen tally determined moments to the moment expres sions based on a model of the process represents a method for obtaining equations where the mass trans port parameters are the unknowns. Schneider and Smithr 21 obtained estimates for the intraparticle dif fusion, axial dispersion, and adsorption coefficients using this method. Mitchell f3 1 used probabilistic ar guments to obtain the spatial distribution, mean position, and variance about the mean of macro molecules moving in an external field and undergo ing reversible isomerization. By comparing these re sults to experiments, the author was able to provide values for the forward and backward switching rates. The same problem was treated by Killalea and McCoy,f 41 who derived expressions of spatial moments. The concentration profiles can be constructed us ing the moments in a Gram-Charlier expansionC 51 i.e., where c(t,z)=m 0 e-x 2 L anHn (x) n = O (7) Expressions of a 0 al' a 2 a 3 and a 4 are depicted in Table 1. The coefficients of the series, a, depend on J the reduced central moments and are obtained using the orthogonality condition satisfied by the Hermite polynomials, H", +~ J H (x)H (x)e-x 2 dx=2n nh /n O n m run (8) where e x2 is the weighting function The first five Hermite polynomials are given in Table 1. This pro cedure is useful for predicting concentration pro files when values of the model parameters are known. Breakthrough curves for fixed-bed ad sorbers and reactors can be represented using mo ments of the impulse response.C6 1 As an alternative to Hermite polynomials, the Laguerre polynomials 13

PAGE 16

are fre q uently used C 7 J Mehta, et al., 1 81 were able to obtain agreement between experimental elution curves and Hermite polynomial series representation. In addition, fitting the two results provided a criterion for the trimming of tails and leading edges. Thus, it is possible to account for phenomena such as end effects. Elution curves in chromatographic columns are nearly Gauss ian when dispersion is small relative to convection. For such cases, the first term in the series, the Gaussian approximation, will provide an accurate representation of the pulse. Chromatography, a widely used separation proc ess, utilizes differences in species behavior, e.g., solid fluid adsorption, to effect the separation. Reliable ways that make use of the moment theory to analyze and optimize chromatographic separations have been developed. For two solutes A and B, separation is usually satisfactory when (9) tion, the adsorbed molecules undergo a reaction that is assumed first-order in the adsorbate concentra tion. The Laplace transform of c(t), c(s), is readily found from the governing equations listed below. Mass balance in the fluid Eck/at=0(c 0 -c)-APkP(c-ci(R)) Mass balance on an individual particle (11) ~aci I at= (D; / r 2 )a( r 2 aci / ar )/arkad sci + kdca (1 2) Reversible adsorption with first-order surface reaction ac 1at= k d c. -kdc -k c a as1 a ra Initial conditions c=c. =c =0 I a for t=0 Boundary conditions (13) (14) ( ac. /a r) =0 and D (ac /ar) =kr(c-c.) (15) 1 r=O 1 1 r = R I r=R The inlet concentration is considered an impulse TABLE 2 The resolution, R provides a criterion for separation of two species when narrow pulses are injected. We can define the Height Equivalent to a Theo retical Plate concept by The Laplace Transform and Temporal Moments for Fluid-Solid Adsorption and Reaction in a CSTR HETP=L 2 / 2 (10) Contrary to R ., HETP depends only on one input sample size and is an important charac teristic for evaluating chroma tographic processes. EXAMPLE PROBLEMS To illustrate how Mathe matica can be used to assist in finding moment expressions for problems of interest to chemical engineers, we discuss three examples. A list of es sential commands to solve these problems is available from the authors. Fluid-Solid Adsorption and Reaction in a CSTR We consider a continuous flow stirred tank reactor (CST R ) fed with an input pulse of adsorbable species. As a rep resentation of a catalytic reac14 0c c(s)= o i::s+0+APkP -RAPk: / (oi[bRcoth(bR)+kpR / Di -1]) h= ~s +kads -kad s kd / (s+kr +kd) Di (2-1) (2-2) (2-3) (2-4) (2-5) (2-6) (2-7) Chemical Engineering Education

PAGE 17

(16) The complex nature of the expression for c(s), determined from the Laplace transformed solution of Eqs. ( 11) to (16), makes using Mathematica to compute m 0 'I' 2 very attractive. We provide ex pressions for c(s), and for m 0 ml' and m 2 in Table 2. Chromatographic Separation Based on Fluid-Solid Adsorption Let us consider separation processes based on fluid-solid adsorption such as the adsorption of hy drocarbons on silica gel in a chromatographic col umn.121 The concentration of adsorbing gas, c(z t), is the solution of the following equations: Mass balance of adsorbable component in the gas phase (EA /a)a 2 c /a z 2 -vac/a z -a c /a t -3 Dp -a)/( Ra {~ ; tR =0 (17) Mass balance of adsorbable component in the par ticle (D; / ~)(a\ /a r 2 +2/rac; la r) -a c ; /a t -( p/~)ac ads /a t=O (18) Linear rate of adsorption ac ads/ at=kads( Ci -cad s / KA) Boundary conditions TABLE3 (19) D (ac /a r) =kr(c-c ) 1 1 r = R 1 (ac; 1a rt 0 =0 for t>O Initial conditions c=O C =0 I at z>O at r;:,:o Bed-inlet condition for t=O for t=O c=c 0 ( t) at z=O for 0<:;t<:;t 0 A (20) (21) c=O at z=O for t>t 0 A (22) KubinL 9 J and Kucera L 10 1 solved this system of equa tions and obtained c ( s,z), the Laplace transform of c ( t,z ), given by Eq. (3-1) in Table 3. Expressions for m 0 1 and 2 were also presented by Schneider and SmithL 21 ( see Table 3). The computations done by hand require many hours of tedious labor. Simple and fast computations with Mathematica provide an identical result. The procedure is straightfor ward when a few special techniques are applied. Instead of A defined by Eq. (3 -4 ) in Table 3, we use its development in series around the point s = 0 to order s 7 Around the point 'A= 0, we apply the series -.fA tanh(R -.fi. )=RA-(R 3 11.2) 3+2(R 5 1? )115+0(A 3 ) (23) To avoid taking the derivatives of 'A 112 around 'A = 0, which presents some problems, we substitute the left-hand side of Eq. (23) into the definition of h(s), Eq. (3-3) in Table 3. This allows one to obtain a development in series of h(s) around the point s = 0 to order s 4, which is used in Eq. (3-2) to develop the series for y. The Laplace Transform and Temporal Moments for Chromatographic Separation Based on Fluid-Solid Adsorption Finally, we are able to compute the derivatives of c(s,z) and obtain 'l' and 2 Unfortunately, rearranging the result into a neat form requires human judgement and is not convenient with Mathematica. However, this task is readily performed by hand ~(s ,z) = (c 0 /s)( l -est0 A )e-yz y = -( va. / 2E v-a.-/ 2_E_A_) 2 -+-(-sa._/_ E _A _)( l _+_ h _( s-)) h(s)= (3kr / R)((l-a.) /a.)[1/s] ll((sD; / kr ).Ji coth(R.Ji)+s(l-(D; / Rkr ))) A= (sPIDJ(l+ppKAkads (P(KAs+kads))) m 0 = m 0 (z = 0) ~ = (z/v)(1+8 0 )+~(z=0) 2 = (2z/v)[8 1 +(EA /a.)( 1+8o)2(vv 2 )]+ 2 (z=O) 80 = ((l-a.)P/a.)(1+(KApp I P)) 8 1 = ((1-a.)Pt a.) [ K!pp l (Pkads)+ l (R 2 p/ 15)(1+(ppKA t p)}2(IID; -5/(Rkr)) When c (t = 0) = c for O :5 t :5 t A we have the following results : 0 0 0 (3-1) (3-2) (3-3) (3-4) (3-5) (3-6) (3-7) (3-8) (3-9) m 0 =c 0 t 0 A ~(z=O)=toA /2 2 (z=O)=t~A / 12 (3-10) Winter, 1992 Spatial Moments of Moving and Interchanging Isomers We are interested in finding the spa tial moments of two isomers A 1 and ~' which are moving, and switching back and forth between the two isometric states as a first-order reaction. This system describes electrophoresis, gel filtration, or sedimen tation if the species are moving respec tively in an electrostatic, velocity, or cen trifugal field. The system is a type of chro matographic reactor. The governing equations of the concen tration are 15

PAGE 18

TABLE4 Spatial Moments for Moving and Interchanging Isomers m 01 =( k 2 / k +){f 1 -f 2 / k Je k.t {41 ) m 11 = L'>vk 2 ( k 1 k 2 +f 1 kJ(l-e k )/k where D is the diffusion of axial dis J persion coefficient, v is the species J t[(k 2 f 1 k + )(v 1 k 1 +v 2 k 2 )e k + t -k 2 (v 1 k 2 +v 2 k 1 ) ]/k : {4-2) velocity, and k is the rate constant J m 21 = (2k 2 / k )( D 1 -D 2 )[k 2 (1-f 1 )-k 1 (1 + f 1 )]( e k +t 1 ) for isomerization The boundary conditions are c(t,x ==O) for j=land2 ( 25 ) J The initial conditions are infini tesimally narrow distributions (2k 2 {L'>v) 2 I k! )[k~(l-f 1 )( e k t 1 )+ k~ ( 1 + 2f 1 )( e k t 1 )+ k 1 k 2 ( 4-f 1 )( 1 -e k t ) ] +(2t/ k : )[D1k~ + D 2 k1k 2( 1 +f le k.t )+ Dlk ~f le k t k ~ (1-f 1)e k \ n !kl+ D 2 k 2) ] -(2t{L'>v) 2 /k!)k 1 k 2 [k/ 1 e k + t -k 2 (1+(1-f 1 )e k t) ] -( 2t {L'>v)/k! )(v 1 k 1 +v 2 k 2 ) [2k ~f 1 +k 2 ( 1-f 1 )(k 2 k 1 )]e k t c (t= 0 ,x)= f8(x ) for j = 1 and 2 (2 6 ) J J -( t 2 e k t /k :)( r 1 k 2 / k Jh k 1 +v 2 k 2 ) 2 -(t 2 k 2 / k )hk 2 +v 2 k i}2 {4-3) where with the total initial concentration given by f 1 + f 2 = 1. Mathematica provides ( a ) the Fou rier transform of c (x,t) defined by L'>v=v 1 -v 2 k + =k 1 + k 2 (4-4) (4-5) To find m 02 m 12 and m 22 interchange subscripts land 2 J += c i( k,t )= f c/x,t)e ikxdx (2 7) and ( b ) the limits of the successive derivatives of < \. In terms of c (x, t ) and C ( k ,t), the n th spatial moment f A 1 b J o species 1s given y J += m ( t) = f c ( x,t )xn dx=(i t limd n C j( k ,t)/ dk n ( 28 ) nJ J k -; 0 Thus we are able to compute expressions for m 0 i' m li' and m 2 ( see Table 4 ). Similar results were presented by Kil!alea and McCoy L 41 and can be used to con struct the concentration profiles as explained in sec tion 1.3. Mathematica can handle readil y such alge braic calculations /somerization in a Countercurrent Chromatographic Reactor Based on the movement in opposite directions of a sorbent and a fluid, a countercurrent chromatogra phic reactor can carry out separation and reaction simultaneously. This can push to completion a reac tion limited by equilibrium. Thus such a device can enhance the conversion of the product in reversible reactions of type A 1 H A 2 Experimental investiga tion of such a system was provided by Takeuchi, et al. ,r 111 who st udied xylene isomerization. The governing equations are given by a special form of Eq. ( 24 ) (2 9) 16 TABLE 5 Concentration Profiles for Isomerization in a Countercurrent Chromatographic Reactor c 1 = Ae rx +B (51 ) c2 = -( u 1 /u 2 )Ae rx +(k 1 / k 2 )B where (5-2) where A= (ci 0 k 1 u 2 erL )/(k 2 u 1 + k 1 u 2 erL) (5-3) B = (c 10 k 1 u 1 +c 20 + u 2 k 1 ) / ( k 1 u 2 + k 2 u 1 ) (54 ) c 20 + =(cio k 1 u1(l -erL)) /(k 2u 1 +k 1 u 2 erL) (55 ) r=k1/u 1 +k2u 2 (5-6) We have made the following assumptions: ( 1 ) steady state, ( 2 ) adsorption equilibrium, ( 3 ) no axial disper sion, ( 4 ) linear adsorption isotherm, ( 5 ) isothermal operation, ( 6 ) first-order reversible reactions and ( 7 ) constant linear velocities The solid adsorbent moving countercurrentl y to the fluid at velocity u carries the adsorbates A 1 and ~' which have adsorption equilibrium and ~ We select the feed, sorbent, and species velocities in such a way that ( 1 ) A 1 is continuousl y fed at the bottom of the reactor, ( 2 ) sorbent is supplied at the top of the reactor, ( 3 ) A 1 is less strongly adsorbed than~ ( 4 ) A 1 is moving upward with the t1uid and A 2 is transported downward by the sorbent, and ( 5 ) A 1 is completely converted in the reactor and the exiting stream is free of ~Thus the boundary conditions are Chemical Engine e ring Education

PAGE 19

(30) The concentrations present discontinuities at the inlet and outlet of the reactor The flux conservation equations provide values for these jumps. Jc 1(L + }=c 1 (L ): 1 / u lc 2 (o )=-c 2 (0 )u 2 /u 5 (31) Expressions for the concentrations c 1 and c 2 are depicted in Table 5 With Mathematica, one can first find these expressions, then perform numerical simu lations, and finally plot the results CONCLUSIONS We have described the computation of expres sions of moments using Mathematica for chemical reaction or separation processes These computa tions are exceedingly tedious to perform and to con firm when done by hand. Mathematica is an inter esting tool for solving problems with algebraic ma nipulations because it is user-friendly, powerful and fast. This is particularly true when the software is run on a Unix machine rather than a smaller, slower personal computer. It should be clear that Mathe matica has limitations some of which can be overcome by skillful organizing of the computa tional steps. The numerical, graphic and program ming capabilities of Mathematica are exciting po tentials that can be applied in multiple areas of chemical engineering. NOMENCLATURE a = interparticle void fraction in the adsorbent bed AP = 3 n / Rp P = interparticle void fraction (i nternal porosity ) of the adsorbent c = concentration of the adsorbable fluid in the interparticle space c 0 = input concentration of the adsorbable fluid cl'c 2 = concentration of species Al'~ cad s = concentration of the adsorbed fluid c, = concentration of the adsorbable fluid in the pore space 1\,8 1 defined by Eqs. (3-8) and (3 -9 ) D 1' D 2 = diffusion or axial dispersion coefficient of species A 1 D = effective intraparticle diffusion coefficient Winter 1992 E = 1n / p p E A = effective axial dispersion coefficient f 1 ,f 2 = rate constant of species A l' A 2 y = defined by Eq (3 -2 ) h ( p ) = function given by Eq (3 -3 ) ~,1 = adsorption equilibrium constant k l' k 2 = rate constant of species Al' A 2 K A = adsorption equilibrium constant k ads = adsorption rate constant kd = desorption rate parameter k r = mass transfer coefficient in column kP = mass transfer coefficient in the CSTR k = surface reaction rate r L = length of chromatographic reactor "A = defined by Eq ( 3-4) m = n t h moment n m "J = n th moment of species A where j = 1 or 2 J nP = number of particles ( grams /vo l. ofreactor ) 0 = ratio of volumetric flow rate to CSTR volume R = radius of the spherical particle of adsorbent p P = apparent particle density s = variable in the Laplace transformation t = time t 0 A = time of duration of the injection of adsorbate u = superficial fluid velocity ul'u 2 = velocities of species A,,~ in the count e rcur rent column u = superficial solid velocity v = linear velocity of the carrier gas in the inter particle s pace vl'v 2 = velocities of species Al'~ x = axial coordinate in the countercurrent column z = length coordinate of the bed of adsorbent REFERENCES 1. Maeder R. Programmin g in Math e matica '" Addison-Wesley Publishing Company ( 1990 ) 2. Schneider, P. and J.M. Smith AJChE J. 14, 762 ( 1968 ) 3. Mitchell R.M ., Biopolym e r s, 15 1717 ( 1976 ) 4. Killal ea, M.K., and B.J. McCoy Biopolymers, 19 1875 ( 1980 ) 5. Kendall, M., and A. Stuart Th e Advan ce d Theory of Statis tic s, Griffin, London C h a p 6 ( 1977 ) 6. Lin ek, F ., and M P Dudukovic Th e Chem. Eng. J ., 23, 31 ( 1982 ) 7. McCoy B J ., Ch e m. En g Comm. 52 93 ( 1987 ) 8. Mehta R.V ., R.L Merson, and B.J McCoy J. Chromatog raphy, 88 1 ( 1974 ) 9 Kubin M. Coll ec tion Czechoslov. Chem. Commun., 30 1104 ( 1965 ) 10. Kucera E. J Chromatography, 19 237 ( 1965 ) 11 Tak e uchi K. T. Miyauchi, and Y. Uraguchi, J Chem Eng. Japan II, 216 ( 197 8) 0 17

PAGE 20

Random Thoughts ... HOW ABOUT A QUICK ONE? RICHARD M. FELDER North Carolina State University Raleigh, NC 276957905 0 fall instructional methods, lecturing is the ~ost common, the easiest, and the least effective. Unless the instructor is a real spellbinder, most students cannot stay focused throughout a lecture: after about ten minutes their attention begins to drift, first for brief moments and then for longer intervals; they find it increasingly hard to catch up on what they missed while their minds were wan dering; and eventually they switch the lecture off altogether like a bad TV show. McKeachie l 11 cites a study indicating that immediately after a lecture students recalled 70 % of the information presented in the first ten minutes and only 20 % of that from the last ten minutes. There are better ways. Actively involving stu dents in learning instead of simply lecturing to them leads to improved attendance, deeper questioning, higher grades, and greater lasting interest in the subject.1 1 2 1 A problem with active instructional meth ods, however, is that they sound time-consuming. Whenever I describe in workshops and seminars the proven effectiveness of in-class problem-solving, prob lem-formulation, trouble-shooting, or brainstorming exercises, I can always count on someone in the third row asking-usually sincerely, sometimes bel ligerently-"lf I do all that, how am I supposed to get through the syllabus?" I have a variety of answers I trot out on such occasions, depending on my mood and the tone of my questioner, but they mostly amount to "So what if you don't?" Syllabi are usually made up from the 18 Richard M. Felder is Hoechst Celanese Pro fessor of Chemical Engineering at North Caro lina State University. He received his BChE from City College of CUNY and his PhD from Princeton. He has presented courses on chemi cal engineering principles reactor design, pro cess optimization and effective teaching to vari ous American and foreign industries and insti tutions He is coauthor of the text Elementary Principles of Chemical Processes (Wiley, 1986) standpoint of "What do I want to cover?" rather than the much more pertinent "What do I want the stu dents to be able to do?": when the latter approach is adopted, it often turns out that large chunks of the syllabus serve little educational purpose and can be excised with no great loss to anyone. But never mind: let's accept-for the remainder of this column, at least-the principle that it is critically important to get through the syllabus. Can I (asks my friend in the third row ) use any of those allegedly powerful teaching techniques and still cover it all? Yes (I reply ), you can. Here are two techniques for doing it. In-Class Group Problem-Solving As you lecture on a body of material or go through a problem solution, instead of just posing questions to the class as a whole and enduring the subsequent embarrassing and time-wasting silences, occasion ally assign a task and give the class one or two minutes to work on it in groups of three to five at their seats. For example: Sketch and label a flow chart (schematic, force diagram, differential control volume) for this system. Sketch a plot of what the problem solution should look like. Give several reasons why you might need or want to know the solution. What's the next step? What's wrong with what I just wrote? How could I check this solution? What question do you have about what we just did? Suppose I run some measurements in the labo ratory or plant and the results don't agree with the formula I just derived. Think of as many reasons as you can for the discrepancy. What variations of this problem might I put on the next test? (This and the last one are particu larly instructive.) Copy r ight ChE Division of ABEE 199 2 Chemical En ginee ring Education

PAGE 21

You don't have to spend a great deal of time on such exercises; one or two lasting no more than five min utes in a fifty-minute session can provide enough stimulation to keep the class with you for the entire period The syllabus is safe! Warning, however. The first time you assign group work, the introverts in the class will hang back and try to avoid participating Don't be surprised or dis couraged it's a natural response. Just get their at tention-walk over to them if necessary-and re mind them good-naturedly that the y' re supposed to be working together. When they find out that you can see them they ll do it, and by the time you ve done three or four such exercises most of the class will need no extra prodding. Granted there may be a few who continue to hold out, but look at it this way : in the usual lecture approach 5 % of the stu dents ( if that many ) are actively involved and 95 % are not. If y ou can do something that reverses those percentages or comes close to it you ve got a winner In-Class Reflection and Question Generation The one minute paper is an in-class assignment in which students nominate the most important and/ or the most confusing points in the lecture just con cluded. L 3 4 J Variations of this device can be used to powerful e ffect. About two minutes from the end of a cla s s ask the students-working individually or in small groups-to write down and turn in anonymou s responses to one or two of the following questions : What are the two most important points brought out in class today (this week, in the chapter we just finished covering)? Examination of the responses will let y ou know immediately whether the students are getting the essential points. Also, when the stu dents know beforehand that this question is coming, they will tend to watch for the main points as the class unfolds, with obvious pedagogical benefits. What were the two muddiest points in toda y 's class (this week's classes this section of the course)? Rank the responses in order of their frequency of occurrence and in the next class go over the ones that came up most often. The responses to this question will surprise you. What you would have guessed to be the most diffi cult concepts may not show up on many papers, if they show up at all; what will appear are concepts you take for granted, which you skimmed over in your lecture but which are unfamiliar and baflling to the students. What would make thi s material clearer to you? You also never know what you'll get in response to this one-perhaps requests for worked-out examples Students in a cl a ss o f m o r e t h an fi ft ee n a l ways i m agine t h ey' r e invisi bl e. W i n te r 199 2 The first time you assign group work the introverts in the class will hang back and try to avoid participating. Don't be surprised or discouraged it's a natural response. of solution procedures or concrete applications of abstract material or pleas for you to write more clearl y on the board, speak more slowly or stop some annoying mannerism that you weren't aware you were doing Responses to this question can pro vide valuable clues about what you could do to make your teachin g more effective Make up a question about an everyday phenom enon that could be answered u s ing material pre sented in class toda y (this week) (Optional:) One or two of your questions will show up on the next test. I used the last exercise-including the zinger about the next test at the end of a course segment on convective heat tran s fer and got back a wonderful s eri e s of questions about such things as why you feel much colder in 20 C water than in 20 C air ; why you feel a draft when you stand in front of a closed window on a cold day ; why a fan cools you on a hot day and why a higher fan speed cools you even more ; why a car windshield fogs up during the winter and how a defogger works ; and why y ou don t get burned when y ou ( a ) move your hand right next to ( but not qu i t e touching ) a pot of boiling water; ( b ) touch a very hot object very quickl y; ( c ) walk across hot coals I typed up the questions ( sneaking a few addi tional ones onto the list ) and posted them outside my office-and in the days preceding the test I had a great time watching the students thinking through all the questions and speculating on which one I would put on the test. (I used the one about the fan. ) There are other short, easy and effective instruc tional methods but thes e should do for starters. Check them out and let me know how they work for you. If I collect some good testimonials ( positive or negative ) I'll report them in a future column REFERENCES 1. M c K ea chi e, W J ., T eac h i n g T i p s, 8t h Edn. D C H ea th & Co., L e xington MA ( 19 8 6 ) 2 Bonw e ll C.C., a nd J.A. Ei s on A ctive L e arning: Cr e atin g Exc it e m e nt i n th e Cla ss r oo m ASHE-ERI C High e r Educ a ti on R e port No 1 G e org e W a shin gt on Uni ve r s ity Wa s hing to n D C ( 1991 ) 3. Wil so n R. C Imp rovi n g F ac ul ty T eac hin g: Eff e cti ve Use o f S tud e n t E va luat io n s an d Co n s ul ta n ts," J H ig h e r Ed ., 57 1 96 ( 19 8 6 ) 4. C r oss, KP. a nd T.A. Ange l o, C l ass r oo m Assess m e nt T ec h n iqu es: A Handb oo k fo r F ac ul ty, N a tion al Ce nt e r for R e sea rch t o Impr ove P ostseco nd a r y T eac h i n g a nd L ea rnin g, Ann Arb o r MI ( 19 88) D 19

PAGE 22

Nd classroom ) INTRODUCING WATER-TREATMENT SUBJECTS INTO CHEMICAL ENGINEERING EDUCATION L.CACERES, E. GRUTTNER, V. MONARDES, R. CONTRERAS, B. G6MEZ-8ILVA Universidad de Antofagasta Casilla 170 Antofagasta, Chile T raditionally, water-treatment subjects are cov ered extensively only in civil engineering, even though several aspects can be analyzed without dif ficulty under the light of current chemical engineer ing theoretical tools. Severe damage to the environ ment caused by discarded wastewater, alternatives for treatment, and theoretical considerations can all be efficiently introduced in courses such as the intro duction to chemical engineering, transport phenom ena, reactor design, and unit operations. Including water subjects in engineering educa tion is particularly important in arid regions of the world where water is a limiting factor for industrial and urban development. In Northern Chile, a region of extreme aridity, 11 21 every industrial activity is af fected by a limited supply of water, and chemical engineers are often faced with industrial processes susceptible to be upgraded, e.g., water recycling. Also, the trend of ever-increasing environmental regula tions in industry justifies the need for introducing Luis Caceres V is an assistant professor of chemical engineering at the Universidad de Antofagasta (Chile). a nd he is a researcher at the lnstituto def Desierto He received his Master s Degree in 1984 from the University of Toronto (Canada). Erika Gruttner D is an associate professor of chemical engineering at the Universidad de Antofagasta (C hile) and is a researcher at the lnstituto def Desierto She finished her studies in chemical engineering at the Universidad Tecnica def Estado Santiago Chile in 1978 Vinka Monardes V. received her Bachelor 's Degree in 1989 from the chemical engineering department at the Universidad de Antofagasta (Chile) and since then she has been actively involved in research at the lnstituto def Desierto Rene Contreras N. is an associate professor at the Universidad de An tofagasta (Chile) and is a researcher at the lnstituto def Desierto He received his PhD from the University of Perpignan ( France) in 1984 Benito Gomez-Silva is a biochemist associate professor and director of the lnstituto def Desierto Un iversi dad de Antofagasta He received his PhD degree in photobiology a t Brandeis Un iversi ty in 1985 20 Including water subjects in engineering education is particularly important in arid regions of the world where water is a limiting factor for industrial and urban development. wastewater treatment subjects into the chemical en gineering curriculum. Most educators agree that environmental-related issues need to be incorporated into the chemical engineering curriculum, but there is no agree ment on the procedure for doing so. To achieve this, some U.S. universities have considered the fol lowing options: including some environmental-related issues in the capstone senior design course consideration of elective courses containing environ mental-related issues coordinating environmental-related-issue training through examples and homework problems in the core chemical engineering courses 1 1 The chemical engineering department at the Universidad de Antofagasta has recently incorpo rated environmental-related water issues into the curriculum with good results and without a signifi cant increase in the students' academic load. The aim of this activity is: 1 ) to promote a working knowl edge of both the basic indexes used in the descrip tion of water quality and water treatment technolo gies, 2 ) to advance an understanding of the need to rid the water of microorganisms and chemicals, and 3 ) to enhance the students' perception of water and its relationship to environmental issues. Our ap proach exposes the students to exercises with incom patible or incomplete sets of data in order to encour age discussion and creativity. This non-traditional approach is a well-recognized practice that has the effect of improving the student's insight into chemi cal engineering processes. 141 We also introduce key water-treatment concepts through exercises and disCopyright Ch E Diuision of ABEE I 992 Chemical Engineering Education

PAGE 23

cussions, and we emphasize both water contamina tion problems and water reclamation during routine visits to local mines and factories. Water recycling and wastewater reclamation are two environmental-related issues discussed in classes at the Universidad de Antofagasta and they are described in detail below. EXAMPLE1 Water Recycling in a Copper Concentrating Plant During the course Introduction to Chemical En gineering, a once-a-year visit to the nearby Mantos Blancos mine is planned. Figure 1 shows the recy cling water system used there. [ 5 1 Chemical engineer ing students are asked to determine all the informa tion necessary to complete Table 1 by interviewing supervisors and operators at the plant. Both the table and the figure are given to the student in advance of the plant trip They must also inquire about the details of the process strategies which minimize water content in the band filter cake. The final field report should contain a discussion on Data Compatibility: in order to find out how well connected all the information collected during the field trip is a mass balance and the corresponding degree of freedom analysis must be performed l 0 1 Water recycling: From the analysis above and given the fresh-water cost, the importance of maximum cake dewaterability in the filter band must be deduced. COPPER CONCENIRl!I As a result of these exercises and discussions, the students achieve a straightforward understanding of the basic operating principles of wastewater stabilization ponds. At this point students should "translate" the in formation given by the plant personnel i.e., the use of percentage solid index, over or under 200 mesh, when standards for cake dewaterability are estab lished. Thus, compounding elements of mass bal ance, integral process control, and concern for water are all involved in this example. Some field reports contain interesting suggestions on the processes stud ies, e.g., experimental measurements of cake dewa terability under different particle size distribution Water recycling is also partially reviewed in courses such as unit operations, where bench scale filtration tests are computed EXAMPLE2 Water Treatment in Stabilization Ponds Theoretical aspects of water treatment can be discussed in a single exercise in the reactor design course Wastewater stabilization ponds are open lagoons where aerobic bacterial and microalgae growth and metabolism transform organic matter into stable com pounds and pathogenic microorganisms are gradu ally eliminated.17 1 Water can then be safely reused, either for agriculture or industrial applications, e.g., TABLE 1 Process Flow Values at Mantos Blancos Concentration Plant ORE FRESK W IIER 6 FF DUNPIIG SI IE Figure 1. Diagram flow of the recycling water system used at Mantos Blancas copper mine II Region, Chile. 1) Concentrating plant. 2) Hydrocyclone. 3) Thickener. 4) Mixing tank. 5) Band filter 6) Conveyor b e lt. WW) Discarded tailin gs FF) Fine fraction. CF) Coarse frac ti on. SJ Sludge. M) Sludge. BF) Band filter. F) Filter cake. R) Re cycled water. Wint er, 1992 Equipment Purpose Hydrocyclone Fast slurry class ification into fine and coarse fractions Thickener Mixing tank Filter band Separation of settleable particles Mixing thickener and hydrocyclone underflows Solid separation b y filtration as to obtain cake with high solid content Effluenl(s} lnf/ow(s} and % solids* and % solids Remarks WW= 1 FF= CF= Xww = 15-40 % Xff= Xcf= FF= BF= Xff= Xbf= S= CF= Xs= Xcf= M= Xf=62-64% S= R= Xs= Xr= M= Xm= BF= FC= Xbf= Xfc Particle size distri bution in WW can b e partially modi fied, if necessary Blank spaces must b e filled in b y students. 21

PAGE 24

ore leaching. Therefore, municipal wastewater can be treated by this relatively inexpensive method in places where both low cost land and high solar irra diation are available_rsi Since continuous reactors and stabilization ponds operate similarly, the latter are also known as bioreactors. Thus, the same efficiency concept and equations studied during the reactor design course can be applied to bioreactors. Organic matter in raw wastewater is a complex mixture of different chemicals and microorganisms. Therefore it is necessary to find a single variable or index in order to use a single kinetic expression properly, either for total biodegradable organic mat ter or microorganisms population. Examples of these indexes are Total Coliform Bacteria per 100 ml of water (TCB) and Biochemical Oxygen Demand (BOD). The latter indicates the total amount of oxy gen required for the biological breakdown of carbon and nitrogen-containing material, i.e., the amount of oxygen required to sustain biological activity in an aerobic pond. f 91 To better expose the students to wastewater treat ment through wastewater stabilization ponds, the following topics are emphasized: a. Due to the complexity of the process, a realistic model must consider that organic matter bioconversion and microorganism deaths are both linearl y dependent on sunlight intensity, and also that the kinetics of the process can be expressed as -dA /d t = k* A *I, where A represents BOD or TCB, I is sunlight intensity, and k is the temperature-depend ent kinetic constant. b. Since solar energy is absorbed by waler, a vertical temperature profile is reached in the pond. Therefore, there should be a discussion on how BOD or TCB is affected by such a temperature profile (assuming that k has an Arrhenius-type temperature dependency) but the kinetic expression is not affected by sunlight inten sity. c. An analysis should be made on the assumptions nec essary for obtaining a single k value at any place in the wastewater stabilization pond. d. If Fick's equation, with analytical solution, is to be used to account for the overall efficiency, the implicit assumptions involved in Fick's equation must be ex plained. Also, the student should try to answer the question of what happens with Fick's equation when a completely mixed or plug reactor is considered. e. Finally, the class discusses the beneficial uses of water reclamation in Northern Chile. Temperature, pH, and absorbance measurements at different depths and times of the day demonstrate the complexity of the biological processes occurring in a wastewater stabilization pond, and therefore its modeling. If time and depth are not considered 22 during the measurements, a simpler model can be reached. In topics a and b above, the discussion is cen tered on the need to work out mass and heat trans fer equations simultaneously, including velocity terms and using the assumptions required. Further more, the proposed analytical form for dA/dt can be questioned since organic matter can also be degraded by anaerobic bacterial activity and TCB can also decrease by changes in pH. A discussion on the clas sical dispersion model for a non-ideal reactor is in volved in topics c and d, but using the wastewater stabilization-pond terminology. As a result of these exercises and discussions, the students achieve a straightforward understanding of the basic operating principles of wastewater stabi lization ponds. In addition, they are able to practice key kinetic reactor concepts with a minimum addi tional academic load Example 2 is complemented by a field trip to the Universidad de Antofagasta's waste water stabilization pond pilot plant. Finally, a simi lar discussion can be carried out on activated sludge as an alternative process where oxygen concentra tion replaces sunlight intensity in the analysis. CONCLUSION Water treatment subjects can be successfully in troduced in the core chemical engineering courses as practical examples of standard concepts. It requires a coordinated effort in several courses in order to discuss similar wastewater aspects through various approaches. The stu dents can obtain a critical view of water treatment processes for recycling purposes through field trips, where real processes are seen in action. Introducing wastewater treatment subjects in the chemical engineering curriculum provides direct access to biotechnological and ethical issues. The students are exposed to environmentally related is sues and treatment alternatives which require an interdisciplinary approach in both industrial and academic research. ACKNOWLEDGEMENT Part of this work was funded by the United Na tions Development Program project CHI-87-024. REFERENCES 1. Geografia d e Chil e. Tomo II Region, Edited by Instituto Geografico Militar, 1st Edition Santiago, 221 ( 1990 ) 2. Klohn, W "Hidrografia de las Zonas Deserticas de Chile, J. Burz, Ed ., United Nations Development Program Santiago, 188 ( 1972 ) 3. Lane, A.M., "Incorporating Health Safety Environmental, Chemical Engineering Education

PAGE 25

and Ethical Issues into the Curriculum ," Chem. Eng. Ed., 23 ( 2) 70 ( 1989 ) 4. Felder R.M. The Generic Quiz : A Device to Stimulate Creativity and Higher-Level Thinking Skills ," Chem. Eng Ed. 19:4 ), 176 ( 1985 ) 5 Narvaez, A., and R. Contreras, Reciclaje de Agua en Planta Concentradora de Cobre, Innovaci6n, 2 (1), 2 ( 1989 ) 6 Reklaitis G V ., Introduction to Mat e rial and Energy Bal ances, John Wiley & Sons, New York ( 1983 ) 7 Mara D. and H W Pearson WHO ERO "Waste Stabiliza tion Ponds ," EUR/ICP / CWS 053 7384V (1987 ) 8. Arthur J P., Notes on the Design and Operation of Wast e Stabilization Ponds in Warm Climates of Developing Coun tries, World Bank Technical Paper No 7 ( 1982 ) 9 Wastewat e r Stabilization Pond s: Principl es of Planning and Pra c tic e," WHO EMRO Technical Publication No. 10 ( 1987 ) 0 lfii letter to the editor ) LA N GMUIR'S ISO T HERM : Ki net i c s or Thermodynamics? Dear Sir: The Langmuir isotherm for adsorption equilibrium of m solutes is of the form a c J J m ( j=l, ... ,m) (1) 1+ I,b c i= l I I where the q and c are the concentrations of the solutes i on the adso;bent ;nd in the fluid phase, respectively and the a and b are constant coefficients. Langmuir 11 derived his equatio~ in 1916 with a kinetic argument: that the rates of adsorption and desorption must be equal at equi librium. To this day almost every textbook follows his reason ing, even though it does not stand up too well to close scrutiny. It postulates the adsorption rate to be propor tional to the concentration of the solute in the fluid phase and the unoccupied surface area ; and the desorption rate to the amount adsorbed. While this is plausible, there is no proof Also, in real ity, adsorption and desorption rates often are controlled by mass transfer and so may obey different laws The derivation should therefore retreat to declaring its rates as only those of attachment to and detachment from the surface ,r2 1 and even then the rate law may differ if attach ment and detachment were to involve more than a single step More seriously where Langmuir's equation is not obeyed do we really believe the deviation is caused by a kinetic anomaly? Lastly have we not warned our students on other occasions that intermingling kinetic and equilib rium arguments is fraught with pitfalls? Although its original derivation is shaky, the Langmuir isotherm has proved to be more successful than any other of comparable simplicity. So why not show our students a derivation that is more convincing and free of merely Wint e r 1992 plausible assumptions? The isotherm describes adsorption equilibrium, and the finest tool we have for equilibria is thermodynamics. We used thermodynamics to derive the mass-action law. To recapitulate: The Gibbs free energy of a (closed) system is a function only of temperature, pressure, and conver sion and is at its minimum at equilibrium; with m partici pants, stoichiometric coefficients v and at constant temperature and pressure (2) at equilibrium, where n i is the number of moles of species i. Introducing the chemical potential i and its concentra tion dependence ciG ~ea .= ~ +RT f n c. un I I I I provided the system is ideal ( activities equal concentra tions ) The reaction stoichiometry requires dn oc v. I I so that m m m L v. = L v + RT L ( v e n c J = 0 i=l I I i = l I I i=l I at equilibrium. Since the term involving the standard potential, ;, is constant m rr C v, seK=const. i=l (3) at equilibrium This is the mass-action law with K as the equilibrium constant. Nowhere in this derivation was it necessary to assume chemical bonds to be formed or broken or even to identify the participants as molecules or chemical species. The derivation covers any kind of process in which something is changed into something thermodynamically distinguish able. Its result can therefore be applied to adsorption as well, and it yields the Langmuir isotherm An easy way of showing the long-known mathematical equivalence of the mass-action law and Langmuir iso therm is as follows. For the reversible "reaction" of a solute moleculej with free adsorbent surface to form occu pied surface the mass-action law gives e __ J -=Kj =const. (4) 9 rr ee cj where the free and occupied surface are expressed as fractions 0 of total surface. Also m err = 1I,e. ee i = l ( 5) From Eq. (4) e K.c _J_ = 1 I 0. K c (6) J J J Continued on page 51 23

PAGE 26

ifii outreac h ) INTRODUCING HIGH SCHOOL STUDEN T S AN D SCIE NCE TEACHE RS TO CHEMICAL ENG IN E ERING 1 TARYN MELKUS BAYLES, 2 FERNANDO J. A GUIRRE University of Nevada, Reno Reno NV 89557 E ngineers and scientists will be key professionals in the future development and implementation of new technology programs that will serve to keep our country in its leadership role. Recent enrollment trends in engineering programs L ll show a decrease in the number of high school students who are inter ested in technical careers, and many who are inter ested do not have an adequate background for com pletion of an engineering program because they do not take advanced science and math courses while in high school. Following the national trend, the Mackay School of Mines and the Chemical and Metallurgical Engi neering Department at the University of Nevada Reno have also experienced significant enrollment reductions for the past few years. Figure 1 depicts the number of freshmen and total enrollment for the Numb e r of students 100 80 1 I 60 I 40 Fre s hm e n ._-_::::J T o t a l J J I J l I i] J I I ~ l l ?O I I 0 1 .. J 80 81 82 83 84 85 86 87 88 89 YEAR FIGURE 1. ChE enrollm e nt at the Universit y of N e vada R e n o 90 1 Pap e r presented at AIChE Annu a l Meeting Chicago IL 1990 2 curr e ntl y with Westinghous e El e ctri c Corporation Resourc es Energy Systems Division Pittsburgh PA 24 Taryn Melkus Bayles is currently a Senior En gineer with Westinghouse Electric Corporation She was formerly an assistant professor of chemi cal engineering at the University of Nevada Reno where she was active with the recruiting and scholarship programs She received her BSChE from New Mexico State University and her MSChE MSPetE and PhD in chemical engi neering from the University of Pittsburgh Fernando J. Aguirre is an associate professor of chemical engineering at the University of Ne vada Reno He received his MS and PhD in chemical engineering from the University of Pitts burgh and worked for several y ears as a re search engineer for Bethlehem Steel Corpora tion His major research interests are mathemat cal modeling simulation neural networks and computer applications last ten years. After reaching a low of one freshman in 1986 and a total of twenty-four in 1987 we began an active recruiting program which involved visiting high schools and telling students about careers in chemical engineering. We visited high schools in Reno, Las Vegas, northern Nevada, and neighboring counties in California. These visits proved to be very useful, with the result that students are now more aware of careers in engineering and that teachers are now better informed and prepared to advise stu dents who have technical interests. As a complement to the active recruiting pro gram, a "Summer Institute was developed. Again, the primary purpose was to introduce high school students and teachers to chemical engineering. The approach here, however was through hands on ex perimentation and exercises which were conducted with the help of university students, a more involved step than the recruiting seminars. Active participation was the vehicle through which students could discover the challenges, the usefulness, and even the fun that is associated with this scientific discipline. The students also acquired a better feel for what types of preparatory math and Copy r ight C h E D ivi s ion of AS EE 1 992 Ch e mi c al Engin ee ring Education

PAGE 27

science courses they would need to pursue an educa tion in chemical engineering The program which was developed for this sum mer instituted included an introduction to chemical engineering, a description of typical chemical engi neering jobs, and general instruction in material balances, fluid mechanics, heat transfer, mass trans fer, and process control. Of course, these topics could only be covered superficially in such a short time, but it was enough to provide some insight into the problems that chemical engineers solve and the cor responding scientific background and preparation that is required The main goal of the program was to increase enrollment in engineering and to encourage women and minority groups to increase their representation in the engineering workforce. STUDENT SELECTION The science teachers in several local high schools (within a 75-mile radius of Reno) were invited to attend the summer institute, and each of the teach ers was asked to select three or four of their better students who had an interest in math and science to also attend the institute. They were also asked to encourage females and minorities to participate. Five high schools responded to the invitation, with a total of five teachers and nineteen students participating The student group consisted of nine women and ten men with four of them classified as minorities. Eleven of the students had just com pleted the freshman year, seven had just completed the sophomore year, and one had completed the jun ior year. Therefore, most of the students had two remaining years of high school in which to prepare themselves in math and science in the event they wished to pursue an engineering degree. PROGRAM ORGANIZATION AND SCHEDULE The Summer Institute in Chemical Engineering was one week in duration, with sessions lasting from 9:00 a.m. until 4:00 p.m. each day The overall sched ule of activities for the week is presented in Table 1. The program began on Monday morning with a gen eral introduction and slide presentation about chemi cal engineering, attended by all students. For the remainder of the Monday session the group of twenty-four students was divided into three groups of eight students each. For the remaining days ( Tuesday through Fri day ) the students were divided into two groups, A" and "B," each of which was divided into two sub Wint e r 1992 The program ... included an introduction to chemical engineering, a description of typical chemical engineering jobs, and general instruction in material balances, fluid mechanics, heat transfer, mass transfer, and process control. TABLE 1 General Schedule for Summer Institute DAY TOPIC Monday Introduction Careers in chemical engineering (videotapes and personal experiences) Material balances (lecture, problems and computer exercises) Tuesday Fluid mechanics session: Group A Mass transfer session: Group B Wednesday Mass transfer session: Group A Fluid mechanics session: Group B Thursday Heat transfer session: Group A Process control session: Group B Friday Process control session: Group A Heat transfer session: Group B groups of five to seven students each Dividing the students into smaller groups provided for a more personal level of participation and understanding by all students involved, which fulfilled the course cri teria of active participation in experiments rather than simply listening to a lecture or watching a demonstration. Each group of approximately twelve students ( groups A and B) had a common one-hour lecture in their first morning session. This was necessary in order to provide background information and to in troduce the students to the subject Some of the activities and exper iments scheduled for the da y were also discussed in this session and explanations about the phenomena they would be observing or measuring were provided. After the morning lecture the groups separated and participated in activities such as laboratory ex periments, calculations to analyze experimental data and computer exercises. Each of the laboratory ses sions was directed by a teaching assistant who pro vided guidance to the students. A rotation procedure was developed so that each of the smaller groups had an opportunity to participate in all laboratory sessions. A list of activities for each daily session (fluid mechanics heat transfer mass transfer, and 25

PAGE 28

process control) is presented in Table 2. Lunch and refreshments were provided for all participants. These periods provided a more infor mal atmosphere in which the students could inter act with the instructors and the teaching assistants, as well as make friends with each other A comprehensive notebook was prepared and dis tributed to each of the student participants. It con tained all of the lecture notes, a de scription and explanation of all the experiments and demonstra tions, and worksheets for performing calculations and other exercises. This material enabled the stu dents to follow along more easily, and hopefully it will serve as a reference for them in the future. It should also be a u seful guide for science teachers who might later integrate some of the material into their teaching. The Summer Institute was taught by two profes sors of chemical engineering (the authors ) with the help of four teaching assistants who were seniors or graduate students in the Chemical and Metallurgi cal Engineering department. DESCRIPTION OF TOPICS A list of the topics covered in the Summer Insti tute, with a brief description of their content, fol lows. Introductory Topics Material Balances Thi s topic introduced the students to chemical engineering. They learned how to formulate and solve material balance problem s which are the basis in the design of process plants. The topic was covered in three different sessions: introductory lecture, solution of problems, and dem onstration of applications using a microcomputer In this last session the students used the CAAPS (Com puter Aided Analysis for Process Systems) softwarel 21 which is a linked system of menu-driven, compiled BASIC program s for the elementary steady-state analysis of chemical processes The software was demonstrated for balancing chemical reactions, con verting units, and performing some simple steady state material balances. Biotechnology A ten-minute videotape which explores the variety of careers open to chemical en gineers in the field of biotechnology was shown. Advanced Materials A fifteen-minute video tape which explores the variety of careers open to chemical engineers in the field of advanced materi als was shown. 26 TABLE2 List of Activities Fluid Mechanics Introductory lecture Reynolds number experiment Reynolds number calculations Pressure drop experiment Pressure drop calculations Computer design of slurry pipelines Heat Transfer Introductory lecture Temperature profiles in solid rods (experiment) Temperature profiles in solid rods (calculations) Double-pipe heat exchanger experiment Double-pipe heat exchanger calcula tions NASA space shuttle tile demonstration Mass Transfer Introductory lecture Diffusion experiment Diffusion calculations Distillation column experiment Distillation column calculations Flotation demonstration Process Control Introductory lecture Valve characterization experiment Valve characterization calculations Liquid level control experiment Liquid level control calculations Computer simulation of level control Environmental Protection A twenty-five min ute videotape exploring the variety of careers open to chemical engineers in this field was shown. It described many of the current environmental prob lems and the role chemical engineers have in their solution. Chemical Engineering Job Experience A de scription of typical job assignments given to chemi cal engineers was presented The two instructors have had prior work experience in the petroleum refining and steel industries and therefore a pres entation of each of those industries was made, in cluding a description of the petroleum refining proc ess and the steelmaking and cokemaking processes. In addition, each instructor presented examples of his/her own job assignments while employed in those industries. Chemical Engineering Education

PAGE 29

Fluid Mechanics Topics Reynold's Number Experiment Visual obser vations of various types of flow (laminar, turbulent, and transition ) were made by allowing water to flow through various sized and shaped conduits, and in jecting a small stream of dye. These visual observa tions were compared to predicted values via calcula tion of the Reynold's number. Pressure Drop in Pipes and Fittings Labora tory experiments were performed to measure the frictional losses of water flowing through various sized pipes, valves, and fitting. These frictional losses were compared to theoretically-predicted pressure drops. Computer Design of Slurry Pipelines Per sonal computers were used to demonstrate a pro gram that is used for designing slurry pipelines. l 31 Use of the computer enabled students to investigate problems associated with slurry pipeline design through some specific design problems Heat Transfer Topics Double-Pipe Heat Exchanger Three steam jacketed, copper circular tubes of different diame ters were used to demonstrate the operation of double-pipe heat exchangers. Water was passed through the inner tube and steam was passed through the annular space The heat transfer coeffi cients, both as a function of flow rate and tube di ameter, were determined for comparison with litera ture values. Temperature Profiles in Solid Rods Three cylindrical rods of varying diameters and construc tion materials were heated at one end The tempera tures along the rods were measured to determine the temperature distribution of each rod These re sults were then compared to theoretical predictions The use of a thermocouple as a temperature-meas uring device was also demonstrated. NASA Space Shuttle Tile Demonstration An actual NASA space-shuttle tile was used to demon strate the thermal insulating power of modern ce ramic materials. The tile was heated to 900 C in a furnace, and then, due to the thermal conductivity of the material, the students found they were able to hold it without burning their hand. Mass Transfer Topics Diffusion Experiment The diffusion coefficient of acetone in air was determined at various tempera tures Acetone was placed in a graduated cylinder Wint e r 1992 and maintained at a constant temperature while air was blown through the top. The rate of diffusion was determined by measuring the change in acetone liq uid level as a function of time. The experimental results were compared to theoretical predictions. Distillation Column Operation A pilot-plant size distillation column was operated to demonstrate the principles underlying separation of two liquid components in solution. Liquid samples and tem perature measurements were taken during steady state operation while separating an acetone-water solution The students were taught how to plot equi librium data and to calculate the number of stages required for a given separation by using the McCabe Thiele method They used this information to com pare with their experimental measurements and to calculate the overall efficiency of the distillation col umn Flotation Demonstration A Denver labora tor y cell was used to demonstrate the underlying principles of flotation This process uses differences in the surface properties of particles in an aqueous pulp to affect a separation. Hydrophobic particles are floated to the surface by finely dispersed air bubbles and are collected as froth concentrate. Hydro philic particles do not adhere to the air bubbles, but remain in suspension in the pulp and are carried off as underflow. Process Control Topics Valve Characterization The valve-sizing coef ficient of a miniature control valve was determined as a function of valve-diaphragm pressure The stu dents also determined whether the control valve had linear or equal percentage inherent characteristics. The procedures used for specifying control valves in practical applications were also discussed Automatic Control of Liquid Level An analog controller was used to maintain the liquid level in a tank at a desired value. The principles of automatic process control were described and the student was exposed to the actual instrumentation used in the field. Performance of the level controller was evalu ated for various operating conditions. The problems which can arise by improper design or through selec tion of the wrong settings was also demonstrated. Computer Simulation of Level Control A computer program that simulates the operation of three non-interacting tanks in series was used to demonstrate some of the principles of process con trol. The program runs on a microcomputer and graphically displays the level of one of the tanks. 27

PAGE 30

This level can be monitored on-line as the user changes various controller parameters The program runs interactively and the student can become the plant operator who has the capability of making changes to the process. The student can open or close the control valve select manual or automatic operation change the setpoint, or change controller settings As a result he or she can observe the re sults of any given change immediately on the screen. COURSE EV ALU A Tl ON At the end of the week the students were asked to complete a course evaluation to help us improve the course in future years. The questions included in the form are shown in Table 3. The response to the Summer Institute was very favorable; many of the students and teachers indi cated that they would be interested in attending the next session also. The students selected the com puter sesions and the laboratory experiments ( which involved more participation ) as their favorites. The videotapes and some of the calculation sessions seemed to be less attractive. However in response to the question about the session they liked the least the answer that students repeated the most was none. Several students indicated that they would include all sessions as those they liked the most and one student wrote, "I liked almost all of them the same. They put into perspective the thoughts installed into our brains in the morning." The diffi culty level of the material covered was considered appropriate by 74 % of the students while the re maining 26 % thought it was somewhat high. One student commented "A few early-morning sessions were hard to understand at first but were cleared up by the afternoon. Having daily periods build on each other was very effective. The majority of the students ( 62 %) indicated that the time spent on lectures should remain the same and 33 % of them would like this time reduced The opposite occurred for the time spent in the labora tory doing experiments or computer work: 62 % sug gested an increase and 33 % no change More mixed results were obtained for the time spent doing calcu lations: 52 % suggested no change, 38 % indicated that the time should be decreased, and 10 % sug gested an increase. The notebook that was prepared received very favorable comments, as did the in structors and teaching assistants One student said about the notebook It was nice to have notes com pleted so we could look back and read what we missed. It will be nice to keep and look back at." Another student wrote, "I plan to study this book 28 TABLE3 Questions on the Course-Evaluation Form 1. Has this course been of help in understanding something about chemical engineering? 2 What session(s) did you like the most? Comments. 3 What session(s) did you like the least? Comments. 4. Was the overall material covered and discussed too difficult? too easy? about right? Circle your answer and provide any comments. ff any specific session was too difficult or too easy, please indicate below. 5. How do you feel about the time spent on each activity? Mark your suggestions and provide any comments. Time spent on lectures Time spent in laboratory Time spent doing calculations In c rease D ec r ease No C han ge 6. Did you find the notes useful and clear? Comments 7. What would you suggest changing to make the course better? and use it for future reference The students made several suggestions on how to improve the course. Their suggestion s ranged from making the course and/or days shorter to making it a four-week course. Even though the course empha sized hands-on laboratory experimentation by the students, they would like to see more experiments in a variety of areas One general suggestion was that students should have had a course in chemistry be fore attending the institute since some of those who had not been exposed to chemistry had trouble under standing the material. In summary the Summer Institute in Chemical Engineering proved to be a very enjoyable and use ful experience for everyone involved ACKNOWLEDGEMENT The financial support of the University of Ne vada System Chancellor s Office and Nevada Mining Association is gratefully acknowledged. We would also like to thank Thomas Lugaski and the teaching assistants Timothy Burchett, David Castillo Bain ian Liu and Carl Nesbitt, for their help REFERENCES 1. Don a ld so n T ., "C h e mi ca l En g in ee rin g Enrollm e n ts," AJ C hE xtra p g 2 Sep te mb e r ( 1990 ) 2 Ca dm a n T W. C APPS : Co mpu te r A ided Ana l ysis fo r Pr oc ess Sys t e m s, U s e r 's G uide, ENS C I In c ( 19 88) 3 Provin e, W ., B. Fr ee m a n G Dow a nd M. D e nn, "D e sign of a Slur ry Pip e lin e," C A C HE IBM PC L esso n s for C h e m ica l Engin eer in g C ours es Oth e r Than D es ign a nd Control 0 C h emica l En g i nee r i ng Ed uc a tio n

PAGE 31

LETTER: Equilibr i um T he r mod y nam i c s Continued from page 7. 1984 to write dH aHTP dT=~ if dP=0 (10) (dP = 0 does not have any special attachment to the ratio that precedes it.) It is extremely useful but I have seen it nowhere else. dH and dT are differentials, smallness not being implied. Williams and Glasser give the following definition of an integral: (11) (A missing comma has been added on the right. ) I believe the definition is faulty. The main point about the proposed notation is that the superscript variables are independent. In ausv;as, the variable V has nothing to do with the differentiation as such. One can of course, define a new function U 8 by (12) for some function (or path" ) VS and then obtain the de rivative aus;as. Because sv ausv T =~ (13) one logically writes the corresponding anti-derivative or integral as (14) and S2 JTSV dS= uSV (Sz,V)uSV (S1,V) (15) S1 The latter is a function of the single variable V, say JV. It is clear that the definition should in fact be (16) The definition Williams and Glasser give is the definition for the constant (17) where (18) again for a particular function (or path ") VS. Of course, the integral then depends on the function VS and is, there fore, function (or "path") dependent ( It is commonly termed a functional. ) The integral ofEq. ( 16 ), on the other hand, is variable d ependent, not function dependent. The descrip tion "path -d epen d ent function" is clearly ill-advised. Both of the integrals given by Eqs (16) and ( 17 ) are useful. Incidentally, ex p ressions of the form Winter 1992 Q= f TdS (19) are justifiable in this context only as abbreviations for integrals like those above. In their treatment of my approach to conservation of energy, Williams and Glasser employ what I think is a little unnecessary circularity. They use the concept well insulated enclosure" to discover the concept commonly called "heat." But it is not clear what a "well-insulated enclosure" means when one has yet to meet "heat." My approach is to argue that experiments are found to fall into two classes : those for which 6U = w (20) holds, and those for which it does not. The former are called anything we like-say, adiabati c ( but it could as well be well-insulated for that matter )-or are said to be surrounded by an adiabatic wall; the latter, non-adiabatic or surrounded by a diathermal wall. At this stage of the argument these terms mean nothing except that Eq. ( 20 ) is or is not obeyed. (It is easy to add terms on the left where necessary to account for kinetic and potential en ergy.) Now one is ready to invent a new quantity which can be given the symbol q; it is merely the quantity that allows Eq. ( 20 ) to be modified so that it always holds The modified equation ( and definition of q ) is 6U=q+w (21) The confused thinking one encounters elsewhere over "work" and "heat" is quite remarkable. Suggestions that the gerunds "working" and "heating" be used instead 121 do not seem helpful. I use "mechanical" and "t hermal trans fer of energy ." I also quote Callen's 131 analogy of the pond and the modes of transfer to it of water by rain and by stream which, when I encountered it gave me just the sort of aha!-insight into thermodynamics that I had badly needed. I borrowed (a nd modified ) the postulates from Callen that now appear in the latter half of Williams and Glasser 121 Much of what follows there is also based on Callen. But there are other sources and influences; for example, the early ideas ofresearchers such as Georgian 14 51 with whom I had corresponded about the units of temperature and the "universal gas constant" R. Williams and Glasser describe what Callen calls the basic problem of thermodynamics and use his method of solving it when the internal adiabatic constraint is re moved: one maximizes entropy over all constrained states. At the end of Part 2 they suggest that the student try three ostensibly similar problems in which other internal constraints are removed Their first problem requires re moval of an internal rigidity constraint and no other. It turns out however that the method does not work in this case ( as it is not difficult to show). Solution by maximiza tion of entropy always assumes removal of the adiabatic constraint at least Earlier in Part 2, Williams and Glasser raise the practical difficulty of how to relax the permeabil ity constraint without relaxing the adiabatic constraint as well. The answer is that one does not Continued on pag e 37. 29

PAGE 32

liiiillllfili._ : -.:c::/a:.:s:..:s=-=a.:...:n-=d~h_:_:o:.:m:..:..:..:e~p:.:r:...::o:..:b:..:.l-=-e:..:.m.:..::s :..__~ ) The object of this column is to enhance our readers' collection of interesting and novel problems in chemical engineering. Problems of the type that can be used to motivate the student by presenting a particular principle in class, or in a new light, or that can be assigned as a novel home problem, are requested, as well as those that are more traditional in nature and which elucidate difficult concepts. Please submit them to Professors James 0. Wilkes and Mark A. Burns, Chemical Engineering Department, Univer sity of Michigan, Ann Arbor, Ml 48109-2136. MORE APPLIED MATH PROBLEMS ON VESSEL DRAINING JUDE T. SOMMERFELD Georgia Institute of Technology Atlanta, GA 30332 P rocessing and storage vessels in the chemical and allied industries come in a large variety of shapes. There are almost as many reasons for this variability as there are shapes; these reasons can include convenience insulation requirements, land and material costs, safety considerations tradition, etc. Drainage of such vessels through an orifice-type hole at the bottom of the vessel represents a class of non-linear, ordinary, first-order differential equa tions, amenable to analytical solution. Thus, from an academic standpoint, this category of practical applications provides engineering educators with a wide variety of useful problems in the a:rea of ap plied mathematics. Solutions to these drainage probl ems have ap peared for many of the geometrical configurations that typically occur in practice. These solutions nor mally appear in trade journals or similar outlets. For example, in one of the earlier such articlesD 1 on this subject, formulas were summarized to compute the time requirements for emptying vessels of four different shapes: vertical cylinder, cone horizontal cylinder (with flat ends), and sphere. Later articles gave similar formulas for draining elliptical vessel heads at the bottom of vertical cylinders ,l2 1 elliptical saturator troughs (horizontal elliptical cylinders with flat en ds ),13 1 and horizontal cylinders with elliptical dished heads or end .r 4 J One can conceive of a number of other geometri cal shapes for vessels or tanks. Admittedly, they might not occur often in the real world, but such configurations may be of some use for academic pur poses, e g., examination or homework problems. Thus (an d also in the interest of completeness) this brief article presents tank-drainage formulas for five new configurations: parallelepiped (or box), vertical ellip tical cylinder, regular tetrahedron, pyramid (in verted), and paraboloid. GENERAL CONSIDERATIONS There are two fundamental engineering equa tions which must be invoked in the solution to any of these tank drainage problems. The first of these is a dynamic material balance for the liquid in the tank which in this rather simple case merely reduces to the rate of accumulation being equal to the negative of the output rate or, more specifically dV =-q dt A~~= -A 0 v2 (1) (2) For the simpler geometric configurations (e.g., vertical cylinders [circular or elliptical] and box) the Jude T. Sommerfeld is a professor in the School of Chemical Engineering at Georgia Tech. He received his BChE from the University of Detroit and his MSE and PhD degrees a lso in che mi cal engineering from the University of Michigan His twenty-five years of industria l and academic ex per ie n ce have been primarily in the area of com puter -ai ded design and he has published over one hundred articles in this and other areas. Copyright ChE Di vision, ASEE 199 2 30 Chemical Engineering Education

PAGE 33

cross-sectional area (A) of the liquid surface in the tank is a constant quantity and not a function of the variable liquid level (z). This results in a very trac table, non-linear differential equation. The effluent liquid velocity (v 2 ) through the drain hole or orifice is determined from a mechanical en ergy balance, specifically between elevation points 1 and 2 in Figure 1. This results in the classical Ber noulli equation 2 2 P1 gc V1 P2 gc V2 p g + 2 g + Z1 = p g + 2 g + Z 2 + h e (3) Two assumptions are conventionally made at this point The first of these assumes that the vessel is 1 ) vented to, and 2) drains to the atmosphere and thus P 1 = P 2 The second assumption asserts that the rate of change of the liquid level in the vessel ( v ) is negligible at all times in comparison with the liquid velocity through the drain hole ( v 2 ) After replacing the variable elevation difference, Z 1 Z 2 with the liquid level ( z ) in the tank we have v~ = 2 g ( z h e ) (4) When the orifice discharge equation is used the fluid head loss ( h e ) due to friction is not explicitly calculated Rather, an orifice discharge coefficient ( C generally less than unit y) is introduced to at te;;_uate the fluid head and compensate for this head loss (5) As shown in most unit operations L 5 J and fluid flow textbooks this quantity C 0 is a function of the fluid velocity ( as incorporated in the Reynolds number ) and the downstream ( orifice )/ upstream diameter raV en t A Q V 2 FIGURE 1. S k e t c h of a vesse l of a rbit rary s hap e wi th a drain h o l e or or i f i ce lo c at e d at th e b o tt o m. W i n ter, 199 2 Sol u tions to these drainage problems have app eared for many of the ge o metrical configur a tions that typically occur in practice. [T h ey] norma lly app e a r in trade j ou rnals or similar outlets. tio although a constant value is generally assumed for a given application. T y pical values of C 0 are be tween 0 60 and 1.0 Indeed a value of 0.75 for this coefficient is reported in a recent article 161 describing an undergraduate experiment on efflux times through a drain hole at the bottom of a horizontal cylinder with flat ends. By way of information civil engi neers L7 1 know Eq. ( 5 ) ( with C = 1 ) as the Torricelli theorem. Insertion of this latter expression into Eq. ( 2 ) then yields (6 ) as the non-linear differential equation to be integrated. SPECIFIC CASES The general integrated form of Eq. ( 6 ) can be written as h t = 1 f Adz C O A D-fig 0 ( 7 ) It should be emphasized at this point that the inte gral in the above expression cannot be replaced by the total volume ( V ) to be drained divided by the square root of the average fluid head over the total height ( h ) to be drained That i s h h f Adz f }z dz* 0 = V ( 8) h 2.-Jh 0 f dz 3 0 h where the denominator represents the average fluid head. When A = const the integration of Eq. ( 7 ) yields (9) as the general equation for the efflux time to com pletel y drain a tank w ho s e cros s sec tional ar ea i s not a function of height 3 1

PAGE 34

Parallelepipeds This first simple case will employ Eq. (9). We consider a rectangular parallelepiped with a length and width (both in the horizontal plane) of a and b units, respectively. In this case, A = ab, and the efflux time from Eq. (9) is t = 2ab,vh COA0.,/2g (10) Clearly, in the case of a square (a = b) parallel epiped one merely replaces the product ab in Eq. (10) with a 2 Vertical Elli p ti c al Cy linder This is another case wherein the surface area (A) formed by the liquid level is a constant, namely nab, where a and b are the lengths of the major and minor axes (again both in the horizontal plane), re spectively, of the elliptical cross-section. Equation (9) in this case then becomes t = 21tab-vh COAO .,j2g (11) and in the special case of a vertical circular cylinder (a= b = D/2), Eq. (11) reduces to the equation pre sented earlier by Foster. 11 J Regul ar Tet r ahedron We consider here only the case of a regular tetra hedron with four equilateral triangular surfaces and with the drain hole located at a bottom vertex opposite the top triangle in the horizontal plane. The length of any edge of this figure is denoted by a. By application of the Pythagorean theorem, the height of any one of these triangles is equal to a -f3 I 2 The total height of this figure is determined to be a {21:3 from a second application of this theorem, and then, from similar triangles, the cross-sectional area (A) of the liquid surface at any level z is given as ( 3-V3 / 8 )z 2 Lastly, insertion of this result into Eq. (7), followed by integration, then yields for the efflux time 3-v'S h5 ;2 t=---~= 20 COAO .,/2g (12) Note the interesting result here that the edge length (a) does not appear in the above formula. It is clear, however, that the initial liquid level (h) cannot be selected in a manner inconsistent with a given such length; that is, h cannot exceed a{2i:3. There is also the not-surprising result of h appearing to the 32 power of 5/2 in the integrated expression of E q. (12), which is consistent with Foster's resuW 11 for a coni cal tank. Inverted Pyramid This case of an inverted pyramid, like the tetra hedron above, is also similar to the case of an in verted cone. The drain hole is at the bottom vertex of the inverted pyramid, the total height of which is equal to c. A rectangular cross-section is assumed for generality. Thus, let a and b (both in the horizon tal plane) represent the length and depth, respec tively, of the pyramid at its top. Two successive applications of similar triangles yield (ab/c 2 )z 2 as the expression for the area (A) of the liquid level at any elevation z. Integration of Eq. (7) with this expres sion for the area then gives t = 2abh 5 12 5c 2 C 0 A 0 .,/2g (13) as the expression for the time for complete drainage We note again the appearance of h to the power of 5/2 in Eq. (13), as in the preceding case. The product ab in this equation is merely replaced by a 2 in the case of a regular pyrami d with a square side length ofa Paraboloid This last case examines an elliptical (again for generality) paraboloid of total height c. As with the vertical elliptical cylinder, a and b here represent the lengths of the major and minor axes, respec tively, of the ellipse in the horizontal cross-section at the top of the paraboloid. The equation for this fig ure then becomes 2 2 ~+L=~ (14) a2 b2 c It should be noted that in many mathematical handbooks and textbooks the right-hand side of the above equation is written as cz or 2cz, in which case c would have the units of reciprocal length. In any event, any horizontal cross-section of this figure is elliptical, and from Eq. (14) it follows that the lengths of the major and minor axes of any such intermedi ate ellipse at an elevation of z are equal to a-rz:Tc and respectively. Thus, the area of this ellipse becomes 1tabz/c, and the resulting efflux time formula is 21tabh 3 12 t 3cC 0 A 0 {2i (15) Chemical Engineering Education

PAGE 35

Finally, Eq. (15) becomes D 2h 3/2 t = 7t 6cC 0 A 0 -j2g (16) as the expression for the efflux time in the special case of a circular (a = b = D/2 ) paraboloid. NOMENCLATURE A cross-sectional area of the liquid level in a tank at any time, L 2 A 0 cross-sectional area of the drain hole or orifice, L 2 a length of a rectangle edge of a regular tetrahedron, or major axis of an ellipse, L b width of a rectangle or minor axis of an ellipse, L C 0 orifice discharge coefficient c height of a pyramid or paraboloid L D diameter of a circle, L F forceunit g acceleration due to gravity ur 2 g c conversion factor, MIJFT 2 h initial height ofliquid in a tank, L h e fluid head loss due to friction, L L length unit M massunit P pressure F/L 2 q liquid volumetric flow rate out of a tank L3 / T T time unit t time T In book review ) CHEMICAL KINETICS AND DYNAMICS by Jeffrey I. Steinfeld Joseph S. Francisco and William L. Hase Prentice Hall New York; 548 pages $48. 75 ( 1989 ) R evie wed b y Robert W. C a rr Univ e r sit y o f M i nn es o ta This book is a text intended for use in courses on chemical kinetics at the advanced undergraduate and graduate level. It covers a broad range of sub jects in empirical (macroscopic) chemical kinetics, the kinetics of elementary reactions, the quantum state (microscopic) approach known as chemical dy namics, and the connections between them. Some background in thermodynamics, quantum and sta tistical mechanics, and kinetics at the level of an introductory course in physical chemistry is assumed, making it a suitable text for chemical engineering students. The book consists of fifteen chapters and three appen d ices. After each chapter there is a list of ref erences, a bibliography, and a number of problems. Wint er, 1992 V liquid velocity, ur x length coordinate in the horizontal plane, L y width or depth coordinate in the horizontal plane, L Z vertical elevation, L z variable elevation of the liquid l eve l in a tank, L Greek Letters n numberpi (3 .14159 ... ) p liquid density M/L 3 Subscripts 1 liquid surface in the tank at any time 2 bottom of tank (at drain hole ) REFE R ENCES 1. Foster T C ., "Time Required to Empty a Vessel," Chem. Eng. 88 105 May 4 ( 1981 ) 2 Koehler F H ., Draining Elliptical Vessel Heads, Chem Eng., 91 90, May 14 ( 1984 ) 3. Sommerfeld, J T., "Inventory and Drainage of Saturator Troughs," Amer. Dyestuff Reporter 79 44, April (1990) 4 Sommerfeld, J T ., Tank Draining Revisited," Ch e m. En g., 97 171,May(1990 ) 5. Brown G.G ., et al., Unit Op e rations John Wiley & Sons, New York, NY p 158 ( 1953 ) 6 DeLozier KR. KA. Platner and J.T. Sommerfeld, An Inexpensive Laboratory Experiment on Fluid Dynamics ," Intl J. Appl. Eng. Ed., 6 377 ( 1990 ) 7. Vennard J .K, and R.L. Street El e mentary Fluid Mechan ics, 5th Ed. John Wiley & Sons, New York, NY, p 131 (1976) 0 The references and bibliography will be useful for those seeking entry into the literature of various topics in chemical kinetics. The book is unusually broad in its coverage, deal ing with a number of subjects that are usually in cluded only in more specialized texts, but which have now become commonplace in kinetics practice. Chapter 1 is conventional in its treatment of ele mentary concepts and definitions but Chapter 2, dealing with complex reactions goes beyond the usual presentation of analytical solutions to coupled sets of ordinary differential equations to discuss applica tions of Laplace transforms, matrix methods, nu merical methods ( Euler Runge-Kutta, predictor-cor rector) and stochastic methods. Computer programs for Runge-Kutta integration and Monte Carlo simu lation are included. Chapter 3, on kinetic measure ments, emphasizes modern instrumental methods for direct detection ofreactive intermediates and the treatment of kinetic data. Sensitivity analysis, an other subject not normally covered in introductory texts but which is of enormous help in understand ing a mechanism is introduced in this chapter. Chapter 4 deals with reactions in solution and Continued on pag e 49. 33

PAGE 36

Cini curriculum ) IDEAS ABOUT CURRICULUM DONALD R. WOODS McMaster University Hamilton, Ontario, Canada LBS 4L7 W hat a challenge it is to develop a curriculum that will prepare our students for the years 2020 to 2040! Should we focus on breadth or depth? Produce a generalist, or a specialist? Besides subject knowledge, how much explicit training should be included to develop problem-solving skills commu nication skills, interpersonal and team skills and the ability to learn-how-to-learn? What is the role of projects? Our department at McMaster is relatively small and does not have expertise in some of the "emerg ing technologies." 1 1 1 We feel that the design project is the capstone course, that synthesis activities should be spread over the whole curriculum and that com puter skills should be well developed. In this con text, what should our curriculum be and how should we go about developing it? GOALS AND ANALYSIS OF OPTIONS To address the issues of curriculum design sub ject to our unique constraints we examined and defined our goals, and consulted with many others during our two years of deliberations. As a depart ment we set goals so that our graduates 3 4 would be technically sophisticated, with an emphasis on the fundamentals would not only be computer literate, but also be able to program effectively and efficiently and would be familiar with a wide range of computer executive programs such as FLOWTRAN, PROCESS, HYSIM, and various spreadsheets Don Woods is a professor in the department of chemical engineering at McMaster University He is a graduate of Queen s University and the Uni versity of Wisconsin His teaching and research interests are in surface phenomena, plant de sign cost estimation and developing problem solving skills Cop y r ight ChE D ivision of ASE E 1 992 could learn on their own would have good communication, problem-solving, interpersonal, and group skills would know how to create hypotheses perform experiments, and draw valid conclusions would have enriched experience through senior electives drawn from our research expertise Acceptance of these goals immediately h i ghlighted some directions : depth of the core fundamentals, generalist with electives available in our areas of expertise. Because the students cannot learn in class all that they will need to know in their career, we will have explicit activities to develop the student s confidence in their own how-to-learn ability. We held a number of faculty retreats for identify ing and streamlining the core concepts in chemical engineering and removing the overlaps and the in consistencies Concurrently a team of students per formed the same tasks from their perspective Spe cial workshops were held on such topics as the labo ratory (led by O.M. Fuller r 21 ) and on the results of our four-year research project to discover methods of improving the development of problem-solving and interpersonal skills. 13 1 We discovered, among other things that using open-ended problems and modeling how we solved problems did not develop the student's problem-solv ing skills, just as giving them good trouble-shooting problems did not improve their trouble-shooting abil ity If they entered the activity with the ability to think of several causes ( whereas we hoped they would be able to think of, say ten ), they left the activity still able to think of only several. To develop the student's ability to learn-how-to learn, we chose as a model the problem-based, small group, self-assessed, self-directed learning model used by McMaster Medical School.l 41 However we did not have the advantage of its admission policies (that eliminated students who did not have well-devel oped problem-solving and group skills ) the time table flexibility, or the resources to provide a faculty tutor for small groups of five students each. So we had to rephrase the problem to "how can we adapt their approach to our constrained system?" and chose Ch e m ic al Engin ee ri ng Edu c at io n

PAGE 37

We discovered, among other things, that using open-ended problems and modeling how we solved problems did not develop the student's problem-solving skills, just as giving them good trouble-shooting problems did not improve their trouble-shooting ability. to use a thread of courses throughout the curriculum which would develop problem-solving and interper sonal skills and small-group learning skills. c 5 61 The development of problem-solving and interpersonal skills was an important goal in its own right as well as an important component needed for the design project. Another issue was the integration of previous courses with the capstone design course so that the team-approach project course could focus on synthe sis. Our approach was that the principles of engi neering economics, ethics, professionalism process improvement, and process design would be handled in a separate "course"L 7 J and to have industrial cli ents with real projects .cai Teams of about ten stu dents and two faculty would tackle a problem. For example, a recent project was with Essofl'exaco on assessing an industrial wastewater-treatment facil ity Difficulties arose since the students entered the project without sufficient experience with computer executive programs and without having had some of the core courses (such as reaction kinetics and reac tor design, mass transfer and process control). In rethinking the approach for the new curricu lum we were not trying to replace the design experi ence-we wanted to enrich it. Four important changes pertaining to the design course were: 1. Introducing computer programming (especially the executive programs PROCESS and HYSIM) in the sop homore and junior years so that the students would enter the project with this expertise already in place. 2. Providing all of the basic fundamentals before the end of the junior year. 3. Shifting the course load in the design course so that it was two tutorials per week in both semesters. 4. Providing explicit training ahead of time on how to work effectively in groups and how to be a team leader. Thus, in addition to a "stream" of core fundamen tals that would be completed by the end of the junior year, we envisioned a series oflearning activities on computer programming numerical analysis, and executive programs on "how to experiment" to develop communication skills to develop problem-solving and interpersonal skills Th ese activities would culminate in and be inteWinter, 1992 grated through the capstone design course. The skill oflearning-how-to-learn would be acquired through problem-based, small-group, self-directed learning in the course "engineering economics. CURRICULUM For the core fundamentals the former senior level courses were altered as follows: the topi cs of reaction kinetics and reactor design, stagewise operations diffusional transport phenom ena, and separations were split into two phases, with the junior level offerin g the basics and a senior level offering enrichment. This actually increased the amount of coverage of this material process control was shifted to the junior year. Thus, by the end of the junior year the students had the core fundamentals in all areas. To provide expertise in the computer executive programs that would be needed for the design project course, a string of three courses in the sophomore junior years focused on computer programming, nu merical methods and modeling, and the use of execu tive packages. For example PROCESS or HYSIM was introduced in the sophomore course, was used to enrich other courses in the sophomore-junior years, and was used explicitly in a computer-simulation course in the junior year that considered a relatively well-structured problem on a small section of a pro cess. Since the content in this string of courses is not ambitious we decided to combine the problem-solv ing/interpersonal skills development activities with each of these three courses. In principle this would provide a ready-made "prob lem and "group activ ity" where the problem-solving skills could be ap plied. Thus the problem-solving activities would not be developed in a vacuum. Rather, they would be embedded and bridged into the computer program ming, model building, and process simulation activi ties. An example of the latter is given by Woods, Wood, and Gallinger. L 91 The problem solving/interpersonal skills/self-di rected learning components were dovetailed into the three team-taught computer programming courses described above and into a senior-level course on process economics and synthesis. In addition, the first sophomore course has as a co-requisite the mass and energy balances course. In this way we hoped to ensure that we had sufficient variety and opportu nity to apply the problem-solving skills in solving 35

PAGE 38

homework problems. Details of this approach are available 1101 A revised series of activities for the laboratories and on experimentation were added. In the sopho more year the emphasis was on familiarity with the hardwarel11l and on measurement techniques. In the junior year the emphasis was on doing a series of medium-size, medium-duration experiments to il lustrate principles. In the senior year, the students do several in-depth experimental projects. A lecture component is included with the laboratories to con sider activities on data treatment, hypothesis test ing, and providing individual feedback about the written reports. Another stream of experience focused on commu nications skills. The Faculty of Engineering has a policy that before students are admitted to the sopho more year, they must pass a Test of Writing Compe tence. So the students in our sophomore course on communication (see Woods and FeuersteinD 21 ) had the necessary basic fluency in English. We then for mally reinforced this skill with private reviews and feedback on the laboratory reports in the junior and senior years and in the presentation of the design project. Many of our introductory, graduate-level courses were modified so that although the lecture compo nent was common, a graduate and undergraduate offering of the course would be presented in areas of our research expertise: biomedical, polymer engi neering, process control, statistics, environmental engineering, particle processing, and surface phe nomena. Senior students could elect to take three of these courses. Figure 1 shows the resulting new curriculum. The freshman year is common to all branches of engineering, so we had no direct control over the content in that year. Accreditation requirements dic tate the eighteen credits of English and humanities (in the freshman, sophomore, and senior years). Fac ulty requirements dictate the engineering courses in mechanics/structures and electricity/magnetism. 36 design project: senior year, 4 credits enrichment chemical engineering electives: senior year (compared with general technical electives), 9 credits computer/simulation stream: introduction, freshman year, 4 credits: sophomore year, 1 credit; junior year, 1.5 and 2 credits communication stream: sophomore year, 1.5 credits; part of 2-credit lab in both junior and senior years and part of the design project problem solving/interpersonal skills/self-directed learning: sophomore year, 2 credits; junior year, 1.5 and 1 credit; Senior year (with special co-requisite requirements to embed the skill development 101 ), 2 credits experimentation and laboratory: sophomore year, about 1/2 credit; junior year, 2 credits; senior year, 2 credits mass and energy balances: sophomore year, 8 credits thermodynamics: junior year, 3 credits reaction kinetics/reactor design: junior year, 3 credits; senior year, 3 credits fluid mechanics: sophomore year, 4 credits heat transfer: junior year, 4 credits stagewise operations and mass transfer: junior year, 4 credits; senior year 3 credits process control: junior year, 3 credits engineering economics and process engineering: senior year, 2 credits HOW WELL HAS IT WORKED? In two words-extremely well. Employers com ment very positively on the quality of our entry-level graduate. Alumni are enthusiastic about the cur riculum and the preparation they have received. The most unique innovation has been in the problem solving/interpersonal skills/self-directed learning components. For this reason, we have conducted a variety of preand post-tests as well as control group assessments. They have shown a statistically significant increase in performance in the main stream chemical engineering courses and a 1 to 2 standard deviation shift in confidence that they now possess skill in problem solving. The self-directed ... """' IV .. 4U ... .... ... .. ... ,_ _,, .. Ill .... _, IZl!l ..... .. II Figure 1 Chemical Engineering Education

PAGE 39

learning has been more difficult to quantify. Anec dotal evidence abounds as to the student's prefer ence for this mode oflearning and to instructors who feel that the students learned more and performed as well, if not better, on written exams than did students who previously "learned" the material via traditional lectures.1 5 5 1 We are changing some components since they did not work out as we had hoped. In particular, the blend of problem solving with computer program ming in the sophomore year did not provide the synergy we anticipated. The students viewed them as two separate parts. So we have integrated com puter programming into the mass and energy bal ances course and maintained a separate 2-credit course on problem solving that applies the skills in the required co-requisite mass and energy balances course. This latter bridge has worked extremely well. The extra course in reactor design in the senior year, likewise, has not lived up to its expectations. So we combined the non-ideal behavior with the junior course (now 4 credits ) and created a richer elective course in reactor design. Also, the students found that the sophomore fluid mechanics course was very challenging, so we have now switched this to the junior year. The design project is much more effective in the new format. We have shifted from PROCESS to HYSIM as the computer executive program of choice. These changes have been minor. All in all we feel that this new format offers a very viable curricu lum model for the year 2000. ACKNOWLEDGEMENT I am pleased to acknowledge the efforts and in put ofmy colleagues. To Bob Moore, Leslie Eubanks Cam Crowe, Terry Hoffman, Joe Wright Andy Hrymak, Phil Wood, Paul Taylor Bob Marshall and Ian Doig for nurturing the problem-solving courses; to Derek Ryder, Steve Kelly, Joe Laricchia, Craig McDougall Bret Cousins, and Sandra Allen ; to Bob Anderson, Les Shemilt, John MacGregor, Irwin Feuerstein, Malcolm Baird, Marios Tsezos Andy Benedek, Keith Murphy, Archie Hamielec, John Brash, John Vlachopoulos, and Jim Dickson. The Ontario Universities Program for Instructional De velopment and McMaster University funded the four year program to understand and develop the prob lem solving/interpersonal skills components. In par ticular I thank Alvin Lee, Vice-President Academic, and Les King and Alan ~lizzard, Instructional De velopment Center, for ongoing support. Winter, 1992 REFERENCES 1. Baird, M.H.I., Things are Humming at McMaster," Chem. Eng. Ed. 4 112 ( 1970 ) 2. Fuller O.M., On Experimentation: A Workshop, McMaster University, Hamilton May 6 ( 1981 ) 3. Woods, D.R., J.D. Wright, T.W. Hoffman, R.K. Swartman, and I.D. Doig "Teac hin g Problem Solving Skills," Annals of Eng Ed., 1 238 (1975) 4. Barrows, H.S., a nd R.M. Tamblyn Problem Bas ed L earn ing : An Approach to Medical Education, Springer, New York ( 1980 ) 5. Woods D.R., "Implementing Self-Directed, Problem Based Learning in a Traditional Curriculum," Chapter 4 in The Challenge of Problem Based Learning by D. Boud and G. Feletti Kogan Page ( 1991 ) 6. Woods, D R., Serf-Directed Learning ," MPS Unit, McMaster University, Hamilton Canada ( 1990 ) 7. Woods, D.R.,"Innovations in a Design and Developm ent Course," Chem. Eng. Ed 2 162 ( 196 8) 8. Woods, D.R., T.W. Hoffman, and A.I. Johnson, "Teaching Experience with Design and Simulation Projects," Chem. Eng. Ed., 7 96 ( 1973 ) 9. Woods, D .R., P.E. Wood, and F.H. Gallinger, Exploiting the On-Campus Boiler House," Chem. Eng. Ed. 20 28 ( 1986 ) 10. Woods D.R., C.M. Crowe, P.A. Taylor, and P.E. Wood "The MPS Program for Explicitly Developing Problem Solving Skills," 1984 ASEE Conference Proceedings, Salt Lak e City, UT, 1021 ( 1984 ) 11. Woods, D.R. R.W. Dunn J J. Newton and D.J. Webster, The Travelling Circus as a Means of Introducing Pr actical Hardwar e," Chem Eng. Ed. 12, 116 ( 1978 ) and Touching Technology," Chem. Tech, Oct., 586 ( 1981 ) 12. Woods, D.R., and I.A. Feuerstein, Eng. Ed., 70, No 7, 745 (1980) 0 L E T T E R: Equ ili b rium Therm o dyn a mics Continued from page 29. The adiabatic constraint is removed in their second and third problems which should, therefore, cause the student no difficulties. ( Many of the points discussed here I raised in the literature a decade ago or more 16 11 1) W.F.Harris Optometric Science Research Group Department of Optometry Rand Afrikaans University P.O. Box 524 Johannesburg 2000 South Africa REF E RENCES 1. Williams, D.F ., a nd D. Glasser, Chem. Eng. Ed 25 74 ( 1991 ) 2. Williams D.F., and D Glasser Chem. Eng. Ed. 25 164 ( 1991 ) 3. Ca llen, H .B., Thermodynamics, John Wiley & Sons, New York, NY ( 1960 ); 2nd e d ( 1985 ) 4. Georgian J.C. Nature, 201, 695 ( 1964 ) 5. Georgian J.C ., J. of Chem Ed ., 43, 414 ( 1966 ) 6. Harris W.F., ChemSA 6 195 ( 1980 ) 7. Harris W.F., Int J of Mech. Eng Ed., 9 181 ( 1981 ) 8 Harris, W F ., Int. J_ of Mech_ Eng. Ed., 9 317 ( 1981 ) 9. Harris, W.F. ChemSA, 7 259 ( 1981) 10. Harris W.F., ChemSA, 8 ( 7), 82 (1982) 11. Harris, W.F., Int J. of Mech. Eng. Ed., 10 287 ( 1982 ) 0 37

PAGE 40

iifi laboratory ) COMPUTER CONTROL OF A DISTILLATION EXPERIMENT CARL T. LIRA Michigan State University East Lansing, MI 48824-1226 U sing the GENESIS environment, software has been developed for computer control of a 10-tray column in the chemical engineering department at Michigan State University. The software permits sequenced start-up of the column, transition to steady-state cascade control, and data acquisition. During runtime, process data are displayed on an equipment schematic or a graphical trend window. The GENESIS software is a powerful authoring sys tem for writing control and data acquisition pro grams and requires minimal background in com puter control. EQUIPMENT AND INSTRUMENTATION Our department developed the control program to help introduce industrial-style computer control to undergraduate students The system has two dif ferent columns which may be used independently and a single reboiler, condenser, and tank system. Flows through the desired column are obtained by moving blind flanges just above the reboiler and just below the reflux splitter. The columns are Pyrex, which permits the students to observe upsets in the columns as well as normal tray operation. The six inch diameter columns have ten trays. One column consists of five valve trays and five bubble cap trays. The other column consists of sieve trays. The trays are constructed from stainless steel and Teflon (Fur ther reference to the system will refer to "the col umn since only one column is used at a given time.) Our department developed the control program to help introduce industrial-style computer control to undergraduate students. The system has two different columns which may be used independently and a single reboiler, condenser, and tank system. Copyright ChE Di vision of ASEE 1992 38 Carl Lira is an assistant professor in the chemical engineering department at Michigan State Univer sity He received his BS degree from Kansas State Univer sity and his MS and PhD from the Univer sity of Illinois Urbana-Champaign He joined the MSU faculty in 1986 His current research inter ests are in the area of processing with super critical fluids. A schematic of our equipment is shown in Figure 1. The input and output signals are summarized in Table 1. All of the wiring from the column passes through conduit to a wall panel box. An identical panel box is installed on the back of the computer cart. BNC and computer cables are used to connect the two panels. This provides flexibility in computer location as well as in use of the computer with other experiments. INTERFACE HARDWARE An IBM AT is used for computer control. We purchased interface hardware from Analog Devices Inc. We use both RTI815 and RTI817 boards in the computer. These boards require signal conditioning which must be performed with additional boards. The combination of boards and configuration are dependent on the large number of signals which were interfaced in our system, and details will be supplied upon request. SELECTION OF SOFTWARE Several different commercial programs are avail able which could equivalently fulfill the objectives of this project. GENESIS was selected because of the following capabilities: 1. A user-friendly interface which requires minimal expertise in hardware/software interfacing and data input 2. Graphic display of the process flowsheet and process variables Chemical Engine e ring Edu ca tion

PAGE 41

T ABL E 1 Summa ry of Computer 1/0 Comp u te r Inputs I Device Signal Type Thermocouple Analog (V) DP cell Analog (mA) Float Level Digital (V) Pressure Analog (mA) Feed F l ow Analog (mA) Valve Position Digital (V) I Computer Outputs I Reflux Splitter Digital (V) Bottoms Overflow Solenoid Valve Digital (V) Steam Control # of Signals 13 2 8 1 1 6 1 1 To other c olumnf-----==::r::C o n de n s er [><)-B V B Y B a ll Va l ve DP -Diffe r entia l Pressure Ce ll F I M ass F l ow M e t e r FCFlo w Contr o l Input LI F l oa t L eve l Indi cato r PR Pr ess ur e R eg ul ato r PS Pr ess ur e S e n so r RS R e flu x Spl i tter SY S o l e n o id V a l ve TI The nn oco upl e Valve Valve Valve Position Analog (mA) Digital (VJ 1 6 Figure 1. Sc h e mati c of th e i n s trum e ntation u t ili ze d fo r th e di s tillati o n c olumn 3. Gra p hic-based control-loop strategy to provide easy editing and visualization of the control loop 4. Sequencing to permit stepwise start up of the column 5 Security to limit the control parameters and setpoints the students can change 6. Capability of "on-the-fly" alteration of control parameters. In addition to the above advantages MSU had previously purchased the interface hardware and th e refore did not need a hardware/software combi nation We also knew that GENESIS software had rec e ived favorable reviews. 11 21 PROGRAMMING WITH GENESIS There are two main components of GENESIS the strategy builder and the runtime system. The str a tegy builder is used to write the control pro gram. Actually, no code needs to be written since the control loops are entered graphically on the screen. A s part of the strategy builder GENESIS provides a powerful graphics program ( the graphics builder ) to construct schematics of the process which are to be viewed during process control. An example of a simple strategy is shown in Figure 2. The strategy is constructed using a mouse to s elect icons from pictorial menus By clicking on an icon, the algorithm blocks are selected an d may be placed on the screen, dragged, and sized with the mouse Each of the algorithm blocks is labeled with an W int e r, 1992 acronym which identifies the type of block. In Figure 2 DEV represents the interface board ( hardware device ) AIN AOUT and DOUT are analog inputs analog outputs and digital output s, respectively. DGAP and PID repre s ent a gap controller and PID controller Many other blocks may be used includ ing an F ( x ) block which may contain up to five lines of calculations using several inputs and registers. Each algorithm block is given a tag name to make the block unique and easil y identifiable. In this example temperature input blocks are indi cated by PLATEl and PLATE7 The DGAP control ler and PID controllers are used to control the re boiler level and steam flow. Specification of details of a block is achieved b y s electing the block with the mouse and entering information in parameter winAIN A I N A I N REBLVL PLATE I PLATE? -I t -I DGAP DEV PID AIN OVRFLW RTI 8 1 5 TM PCN TR L iT M P SET -I r L -I DOUT AO U T PID A I N I+ 1-OVR STMFLOW STMCNTRL STMMEAS Figure 2. Ex ampl e co nt ro l st rat egy as co n s tru c t e d by s trat egy build er. 3 9

PAGE 42

dows which pop up on the screen. The actual control loop is more complex than Figure 2 because it is desirable to use direct manipulation of the steam flow setpoint while keeping the temperature control ler disabled during start-up of the column or for manual override of cascade during operation, as shown in Figure 3 An example of our control screen designed with the graphics builder is shown in Figure 4. As the students step through the start-up procedure differ ent control screens are displayed. For each of our screens the equipment schematic remains the same but the control panel in the upper left contains dif ferent instructions. Figure 4 is the cascade control screen The control screen is displayed using colors which are not reproduced here During runtime, the graphics screens provide instructions and data dis play. Depending on how the graphics are designed, proce s s data may be displayed numerically, graphi cally or pictorially using a change in the size or color of an object. In our schematic the valves are dis played in brown when closed and bright green when opened. On the control screen all process data are nu merically displayed near the sensing device in the schematic and are updated each second. The secu rity of cursor positions and the range of valid entries is controlled. The only valid cursor locations are the data-entry positions in the control panel in the up per left of the screen Alarm messages in the boxes on the right side of Figure 4 are displayed Screen 8 Control Var-rabies Plate 7 Set.po I nt. C CJ Measured (CJ Ref I ux 85 5 85 4 Cyc I e T 1 me(se c) 40 0/L Ratio D 58 Man Valve Control? N Man Steam Set.point. C 0500 I bs/ h) 2 0 0 Valve Posit io ns Feed Tank 1 BV 1 Fee d Tank 2 B V-2 Column F eed B V-3 ing runtime to follow the controller performance and the approach to steady state. Trends may be displayed over time intervals of 1, 6, or 30 minutes only. Up to 20 variables may be trended and are displayed on the screens in groups of up to 5. The screens may be selected using the "Page Up" and Page Down keys. An example of a trend window is shown in Figure 5, illustrating the effect of a tem perature setpoint change on the steam flow rate. The trend windows utilize colors to differentiate be tween variables and labels. For Figure 5, different fonts have been use d to differentiate variables for T emp Meas R eboiler 00-=--~ Temp Erro r Controller Sig n a l Temp Se t poin t Feed Orifice P late Steam Flow Con t rol l e r Ou tp u t ---~Xl--cc--~ Steam Flow Swit ch 1 0 bypass Cascade Control Controller Figure 3. S c h e mati c of th e cas c ad e c ontrol lo o p u s ed for th e MSU distillation c olumn. CNDSR PLATE1 PLATE2 PLATE3 PLATE4 PLATES PLATES PLATE7 PLATES PLATE9 PLATE10 REBO ILER FEED 79 2 80 2 81 1 81 9 82 5 83 0 84 1 85 4 87 2 BB 7 92 1 94 2 28 4 I sHuTOO W N SOUTH COLUMN FEED OPEN FEED TANK FULL FEED TANK 1 LOW only if triggered by a spe cific condition Different alarm levels may be used to identify the se verity of an event. In our application, all of the alarms are warnings ex cept for a low reboiler level which triggers the emergency state and shuts off the steam line. BV-4 FEED TANK 2 LOW BOTTOMS RE CEIVE R FULL PRODUCT RECEIVER FULL EMPTY RECEIVERS RUNTIME Once a strategy and associated graphics dis play have been pro grammed the process may be executed. Trend windows permit graphic display of the data dur4 0 Eme rg en cy S hut.d o wn ? N FT-1 FT2 [J '?_ Flo w ~I BV 1 BV-2 _s.2 P1 HX -2 To enter data ~ 80 use tab key# make ----t:kJ cha nges ~ p ress Al t -0 LOW STEAM PRESSURE LOW REBO ILER. EMERG SHUTDO W N! Pressure 53%1197PSig Meas 1.:17 Open >-R-eb_o_i_l_e_r -++--~ t:kJ-STEAM Steam lb/h Set.p 145 CV1 27 5 I rl Programmer BV-5 C T Lira. Figure 4. Exampl e c ontrol sc r e en for th e MSU c ontrol pro g ram This s c r ee n is the s c r ee n for cascade control Ch e mi c al En g in ee ring Education

PAGE 43

purposes of this publication During runtime, ac quired data may be written to the hard disk at time intervals ranging from 1/lO th -second upwards. STUDENT USE The objectives of the use of the distillation equip ment are: to expose the students to the industrial equipment and sensors to run the column at steady-state and collect and analyze tray samples to calculate tray efficiencies to expose the students to an industrial-style control system. The students are provided with a short problem statement. Typically they are to separate an equimo lar feed of ethanol and propanol to a product of 90 % ethanol and bottoms of 90 % propanol and to deter mine Murphree tray efficiencies. They must obtain literature data to construct the x-y diagram and then perform a McCabe-Thiele analysis to predict the composition on tray 7 at steady-state, which determines the setpoint temperature. Also, the ap propriate reflux ratio must be calculated. During the first 3-hour laboratory period the stu dents familiarize themselves with the experiment by watching a video tape which illustrates the start-up procedure and use of the computer. The teaching assistant demonstrates the use of the gas chromato graph used for sample analysis. Students read through the laboratory manual which contains por500 500 100 100 0 0 15 15 JUN 11/91 14: 34 :43 TREND STMFLW STMSET TEMPSET PLATE? OUT OUT OUT OUT 30 M inutes 120.2 LBS / HR 122.0 LBS / HR 84.12 C 83 93 C Figure 5. Example trend window which may be viewed during experimental runs. Winter 1992 During the first 3-hour laboratory period the students familiarize themselves with the experiment by watching a video tape which illustrates the start-up procedure and use of the computer. tions of the data sheets and documentation which the manufacturers supply with the instrumentation. Upon student request, the column is started to fur ther illustrate correct operation The students submit a written project proposal twenty-four hours before the second laboratory pe riod which summarizes their McCabe-Thiele calcu lations and proposed laboratory procedure Upon ap proval, the students may begin the experimental work. A six-page instruction summary in the labora tory manual provides the necessary guidance for the use of GENESIS with this experiment Although an experienced operator can start the column without the instructions, the printed information provides useful reference regarding the steps performed at each screen during the sequence, hints on steam flow during start-up, explanation on handling alarms, and important keystroke combinations to plot trend data on the control screen. The sequencing capability of GENESIS is uti lized during the start-up to display instruction screens to the students. At each screen the students must take some action such as verifying that the cooling water is flowing or draining condensate from the steam line, or turning on the pump. When they are ready to proceed, a keyboard command takes them to the next sequence step. Feed is prepared by mixing alcohols from the product and bottoms tanks. The composition is de termined by gas chromatography. Usually, students use 30-45 minutes to drain the tanks and prepare the correct feed composition. Start-up of the column is achieved using the steam-flow controller only. The setpoint is entered on the screen by the students. Although the written instructions provide sugges tions on the steam flowrate, the final choice is made by the student. Typically, the students will encoun ter some type of difficulty during start-up, such as flooding due to vapor flow up the downcomer of the bottom tray flooding on the feed tra y reboiler level loo high or too low slugs of vapor in the reboiler thermal siphon causing incorrect readings in the reboiler level. 41

PAGE 44

Usually, we permit the students to encounter these difficulties because they help to illustrate the limitations of control systems. In the absence of com plications, vapors reach tray 7 in about fifteen min utes and the condenser in thirty minutes. After vapors have reached the condenser the stu dents proceed to the cascade control screen shown in Figure 4. Two factors are important in determining the approach to steady-state. First, the temperature controller has very little integral control due to the sluggish behavior of the system. Students are en couraged to use the manual steam flow setpoint until the temperature of plate 7 is within three de grees Celsius of the desired setpoint before proceed ing to cascade. The second factor is that GENESIS provides bumpless transition of the steam flow set point upon transfer to cascade control, which means the steam setpoint will be adjusted gradually from the value existing at transfer to cascade control. Students are encouraged to anticipate the sluggish ness and adjust the steam flow setpoint up or down before transferring to cascade control. This adjust ment requires them to make a judgement as to whether the system will overshoot if operating at the current settings. An overshoot of a couple of degrees will result in stabilization at setpoint taking up to fifty minutes. A well-planned transfer to cas cade will result in a stabilization in about twenty minutes. The trend screen data plots are useful in watching the approach to setpoint and steady-state. After tray 7 has reached the setpoint value, tray temperatures are monitored for at least twenty min utes to assure that temperature drifting (composi tion changes) are not present on any trays. The students then collect tray samples. Shutdown of the system is also performed using GENESIS. Shutdown uses the same procedure as an emergency shutdown, so the student operator enters a "Y" in the cursor location next to the "EMER GENCY SHUTDOWN?" prompt in Figure 4. (This prompt is present on all screens.) The record of col umn history is printed from the hard disk following completion of the experiment and is submitted with the student's laboratory report. Typically, students will repeat the experimental run in a third laboratory period, making slight modi fications to the reflux ratio or setpoint temperature. With our current laboratory objectives, the students do not have adequate time to fully explore the wealth of control experiences feasible with GENESIS. Pos sible control exercises include studying the effects of sudden changes in feed flowrate or reflux ratio. An42 other possibility is to further explore the tuning of the controllers. Typically, the students do not fully appreciate the sluggishness of the system. An ex ample exercise would permit them to increase the integral control to improve the rate of approach to setpoint and then observe that the change results in continuous oscillations about the setpoint value. An other possibility is to permit the students to perform the Ziegler-Nichols tuning of the controllers, or to attempt control without using cascade methodology. We do not utilize these unsteady-state exercises be cause, with the available laboratory time, it is not possible to restabilize the system and collect tray samples for meaningful comparison with McCabe Thiele analysis. CRITIQUE AND SUMMARY Student response to the use of GENESIS is ex tremely positive. We have chosen to install the con trol system on equipment that illustrates the use of a large number of data inputs. GENESIS would also be extremely useful for a smaller scale experiment with a single control loop. The flexibility of software permits the instructor to prepare exercises and con trol screens which require little knowledge of con trollers or which are very advanced. The capability for watching unsteady-state behavior is excellent. It would be easy to plot the integral, derivative and proportional components of a controller on the same trend window as the overall controller and loop response. The use oflogged data from the spread sheet compatible files permits the students to fur ther manipulate their data and prepare plots. The data files are compatible with LOTUS 1-2-3 and may be manipulated further, although we do not use this feature. We have found the application of GENESIS to provide a powerful method for design of control and data acquisition programs, especially for those with out experience in computer control. Because of the graphic-based strategy builder, the programmer does not need to develop or maintain expertise in reading or writing to registers. Once an understanding of the software has been developed, it becomes easy to write new control loops. The control program does not greatly change the instructional time that must be spent with the students, but provides exposure to an important aspect of control technology. ACKNOWLEDGEMENTS I would like to thank Al Paulsen of Dow Chemi cal for his assistance in specification of instrumenta tion, Dow Chemical for funds for the instrumenta tion, and Bruce Wilkinson of the Chemical EngiChemical Engineering Education

PAGE 45

neering Department at MSU for planning and su pervising the installation of the instrumentation. REFERENCES 1. Bottali, P., and M. Kelly, "Software Monitors Electrical Loads at IBM, "" Control, March ( 1989 ) 2 Software Engineering Consultants, "GENESIS: Human Fac tors Approach to SCADA ," Control Engineering November ( 1989 ) APPENDIX 1: Equipment Description The l evel indicators are floats which indicate only whether a tank is nearly full or empty. All of the ball valves are air actuated with a sensor to indicate valve position. The air supply is switched to the valves using solenoid valves in an air-supply line The feed pump is a small bellows pump Feed rate is sensed with a Micromotion mass-flow meter The oscillations from the recipro cating pump cause fluctuations in the flow-meter readout which are dampened by adjusting the electronic filter furnished on the flow meter The flow-meter signal is furnished to the computer as a 4-20 mA signal. The feed enters the column after passing through a s ingl e tube in s h ell preheater constructed from two different sizes of stainless-steel tubing. The steam is regulated from a supply pressure of 90 psig to 20 psig before entering the system. A steam pressure sensor provides a 4-20 mA signal to the computer. Steam flow is sensed with a differential pressure (DP ) cell across an orifice plate. The DP cell provides a 4-20 mA signal on a sing le pair of wires using a 24V supply. Steam flow is regulated by a 4-20 mA signal from the computer. An I/P (current to pressure) converter provides the air pressure to driv e the control valve. The relative plac ement of the orifice plate and control valve have provided satisfactory control of the column. In the current configuration, the upstream pres sure on the orifice plate depends on the control-valve position The reboiler consists of a stainless-steel heat exchanger. The reboiler l evel is sensed using a DP cell. This DP cell operates on the same principles as the steam-flow DP cell. The reboiler l evel overflow to the bottoms receiver is controlled with a solenoid valve The computer controls the solenoid with a bistate gap controller signal based on the reading from the DP cell. A needle valve in the overflow line serves as a restriction to limit the flow rate The gap controller is set to open the solenoid at hydrostatic pressure of 25.1" of water and to close at 24 9". The reflux splitter is driven by a 120 VAC solenoid. The computer may be used to close a relay to provide the AC power or a separate electrical/mechanical timer may also be used. APPENDIX 2: Interface Hardware and Software Analog signals are isolated and converted with an Analog Device 3B signal conditioning system. Analog input signa ls from the thermocouples and current inputs are converted to 0-lOV signals which the interface boards then converted to numerical values of 0-4096 GENESIS reads the registers and converts the register values to engineering units. The analog output for con trol of the steam flow is passed through the conditioning system with a reverse of this procedure Digital signal I/0 is handled through optical isolator modules and the signals are passed through either the RTI815 or the RTI817 GENESIS is available from !CONICS ( phone 508-543-8600) A variety of control versions is available in addition to a wide range of device drivers to communicate with hardwar e from many manufacturers. Data Translation and MetraByte boards, which are popular in many university labor atories, are supported. GEN ESIS is priced competitively with other control packages such as LABTECH Control. !CONICS offers a university discount. A very Winter 1992 inexpensive demo version is available which is limited to 20 algorithm blocks. APPENDIX 3: Using the Strategy Builder The start-up of the column is achieved using a sequencing algorithm block which is not illustrated in Figure 2 The se quencer is designed with a series of states. Eight states are used in our start-up sequence, along with an emergency state. Control events may be associated with each state. Sixteen bits are associ ated with e ach state which may also trigger events. The status of each bit may change with a given state of the sequencer. In our app lication, the sequencer is used to control the display of control screens which provide instructions for start-up, enable alarms and warning messages which alert the students when a step has not been completed, and provide restrictions on the type ofsteam flow control for a given state. After an algorithm block has been placed on the screen, the details regarding the block may be entered by clicking th e mouse on the block When a block is selected, a parameter window pops up on the screen. Parameter values for input ranges, output rang es, alarms, controller settings, etc., are entered in this win dow. All of the values entered in the blocks are in engineering units making the programming quit e simple The instrum e nt range and operating range may differ For example, our thermo couple inputs provide an instrument range of 0-500 C, but the measurements are only e xpected in the 15-ll0C range. B y speci fying the operating range the graphical use of the acquired data is kept on a meaningful scale. GENESIS scales all data internally for communication between blocks. Following specification of the blocks arrows are added be tween blocks, using the mouse, to specify the flow of information. For example, in Figure 2 the reboiler level ( REBLVL ) is obtained from the interface card ( RTI815 ) At the time the arrow is con structed, windows pop up to facilitate specification of the input channel of the card which carries the signal. REBLVL serves to convert the raw signal to engineering units OVERFLW provides a digital signa l to OVR, which communicates with the hardware card. TMPSET is an examp l e of a block which does not obtain its value from the interface card The initial value ofTMPSET can be entere d at the time the strategy is constructed but the value may be changed during runtime from the computer screen. This is done by associating a particular cursor location on the control screen with TMPSET during graphical construction of the control screen The graphics builder is used after the strategy has been developed to design the computer-control screens which display data from the strategy during runtime Careful design of the control loop and graphic display should precede programming to minimize modifications. Alteration of the strategy block connections or graphic display requires l eaving the runtime system and reentering the strategy-builder program. Our system runs on an older 286 IBM AT which was slow in loading the large strategy builder and in saving the compiled control program "code. On the other hand most of the param eters for the algorithm blocks may be changed during runtime eliminating the need for reentering the strategy builder if the structure of the strategy is correct This makes controller tuning relatively simple. The revised parameters are stored on the hard disk and by simp ly renaming a file after completion of the run the revised parameter file becomes the default file The details of some of the strategy algorithm blocks are not described com pletely in the GENESISdocumentation, but many of the capabili ties exceed the needs for most applications. In general, a phone call to the company quickly resolves any confusion GENESIS offers a training course which is helpful in learning the capabili ties of the software. 0 43

PAGE 46

Ndi classroom ) CHEMICAL ENGINEERING DESIGN Problem Solving Strategy RICHARD TURTON' RICHARD C. BAILIE West Virginia University Morgantown, WV 26506-6101 A significant fraction of our undergraduates ac fi cept industrial employment that requires deal ing with problems associated with an operating sys tem (pro duction ). These responsibilities depend on an ability to both predict problems before they occur and to diagno se problems after they occur, in addi tion to designin g and d eveloping new processes. Prob lems must be solved under the constraints imposed by physical equipment and the existing processes. Neither course work nor the typical capstone design course makes a conscious effort to develop such abili ties in the student. To address this problem, the chemical engineer ing department at West Virginia University has ini tiated a process in the senior design course that focuses on the development of the necessary skills for solving problems which production engineers may expect to confront. 11 2 31 The program introduces a series of comprehensive problems, extracted from a single process flowsheet, into the two senior design courses. The problems are worked out individually under the same rules that apply in the AIChE stu dent-contest problem wherein the stu d ent presents Richard Turton received his chemical engineer ing degrees from Nottingh am University (BSc, 1977) and Oregon State University (MS 1979; PhD 1986). He is currently an Associate Profes sor at West Virginia University and his primary interests lie in fluidization research and under graduate and graduate education 44 Richard C. Bailie holds C hE degrees from Iowa State University ( PhD) Wayne State University (MSChE) and Illinois Institute of Technolog y ( BS ChE) and is currently a Professor at West Virginia University His interests are in f/uidization a n d energy utilization and he has published man y articles and a book in these areas He is currently working on a second book on the teaching of design in the undergraduate curriculum. Copyri g ht ChE Divi sio n, ASEE 1 992 TABLE 1 Unit Function: Intellectual vs. Industrial UNIT INTELLECTUAL CONCEPT 1 Visualization plus Analysis 2 Analysis plus Synthesis 3 Synthesis plus Optimization INDUSTRIAL APPLICATION Trouble Shooting R etrofit New Process D esign a written report and a one-hour oral report on each problem. The student is "escorted" through the tran sition from a detailed consideration of a single unit operation to management problems that can include several interrelated unit operations. This is accom plished by providing intensive individual and group feedback after each assignment. Major adjustments in student attitudes are some times necessary. It is difficult for the student to go from solving narrow, focused, single-answer prob lems that involve no creativity to global, multi answer problems that require considerable creativ ity. They find that the well-entrenched strategies for problem solving that worked for them in the past are no longer effective. DISCUSSION The general content of the comprehensive prob lems are summarized in Table 1. It shows the skills that are necessary for solving the problem and the area of its industrial application. The order of the problems is significant. The skills required in a pre vious problem must be mastered in order to solve each new problem. Each problem develops a single skill while introducing a new skill that is the basis for the following problem. For example, the first problem develops an ability to visualize-and intro duces analysis The second problem assumes the ability to visualize, develops an ability to analyze and introduces synthesis. In terms of industrial ap plication the three problems will cover trouble-shoot ing, retrofit operations, and new process design. The problems are introduced to the student Chemical Engin ee r ing Education

PAGE 47

through industrial "role -playing ." The student (a nov ice engineer) joins a company and is assigned to a plant site where three assignments are given that cover the areas shown in Table 1. The stu dent is given a packet of materials that includes a process flowsheet and flow table, a description of changes that have been made to the process, thermodynamic data, and various other items of information describ ing the process and the company that may be useful. The process flowsheet is special. It represents a poor process design (it was built many years ago) under current conditions and includes several poor design areas that could be improved. Some defects have been distorted (like big ears and long noses ) to make them stand out and be easy to detect. The special flowsheet represents a "caricature flowsheet" and is an aid to the early development of visualiza tion skills. In addition to poor design, several proc ess modifications (operating conditions) have been made in response to changes in constraints and availability of new technology. In the process of solving problems, students 1 ) will identify all of the "caricature problems" and offer solutions, 2) will identify many other problems, not highlighted for identification and offer solutions, and 3) will create new processes that improve the original process and eliminate the poor engineering of the original system. All solutions must be justified by economic considerations. All three projects are associated with the producEQU IPMENT LIST AB-1 Qu e n ch Tower AB-2 Off Gas Absorber C-1 Inl e t Air Blower E -1 Reactor Coo l a nt Exchange r E-2 Absorber Product Cooler E-3 Solvent Separator Condenser I E-4 Solvent S e parator R ebo il er 1 E-5 Waste Water Separator Co nd enser E-6 Waste Water Separato r Reboiler I Air E-7 Product Separator Condenser BTH -----; E-8 Produ c t Separator R e boiler E-9 In c in e r ato r Waste H eat Boiler H-1 Off Gas In cinerator M-1 Solvent-Produ c t Mix e r S-1 Solvent-Produ c t Liquid Separator T-1 Solvent S e parator I T-2 Wast e Water Separator T-3 Product Separator UT ILITY LIST BFW Boiler Feed Water CW Cooling Water CWR Cooling Water Return CR Co ndensat e Return DI Deionized Water NG Natural Gas STM Steam Propylene The assignments and experiences of our 1988 class are used here us an example of how the comprehensive problems along with the "caricature flowsheet," are used to introduce and develop skills. tion of the same chemical commodity. Since each successive problem requires mastery of the previous problem, it has been necessary to provide extensive individual and group feedback during the oral and after each project. STUDENT ASSIGNMENTS The assignments and experiences of our 1988 class are used here as an example of how the com prehensive problems, along with the "caricature flow sheet," are used to introduce and develop skills. Fig ure 1 is a caricature" flowsheet for the production of acrylic acid. It was created from a 1986 AIChE stu dent contest problem on the production of acrylic acid via partial oxidation of propylene. A packet of material ( about twenty pages ) was provided that contained information concerning the instability of the acrylic acid with temperature ( even with the addition of an inhibitor, it will rapidly po lymerize at above 90 C when present in large con centrations). Some additional information on cur rent operations is also given. NG /Clp 1 To stack STM E-9 AIR ~ I j H-l BFW DI M-l I CWR Detail -AACWR Treatment Acetic Acid (IT STM \!,J.. cR Acrylic Acid Figure 1. Acrylic Acid Process Flowsheet Wint er, 1992 45

PAGE 48

ASSIGNMENT 1 The plant faces a serious prob l em Our largest customer cancelled its order. We shipped off-spec material ( a slight discoloration in the normally clear acrylic acid). It damaged their process and their product was worthless (colore d ) The potential loss to our company is in excess of $500 000/day and this is the economic incentive that calls for an immediate resolution. The assignment to be completed in one week is 1 ) to identify the cause of the off-spec material, 2 ) to recommend action to be taken to correct the situation 3) to present a written report and 4 ) to make an oral presentation to the company president There are several potential explanations for the off-spec material There is evi d ence to support some of the explanations, while others are no more than speculation Table 2 gives a partial list of some of the explanations. The y can be separated into three groups : 1 ) global 2 ) numerical and 3 ) research. An example from each group follows and possible solu tions are included in Table 2. Global problems require that the student con sider items o u tsi d e of the process unit for an expla nation For example the off-spec material may have resulted from lowquality feed There feed and ruined their product. This could be the same problem we have We could ha v e r ec ei v ed tainted feed material. We have evidence to support this explanation. Students not only fail to make a c o n n ection betwee n these two sit u ations but they a l so are not rea d y to accept the ex pl anation. A limited number of explanations can be verified numerically ( the purpose of this problem was to limit the number of necessary calculations ) The refriger ated water ( at 10 C ) to E7 ( Figure 1 ) wa s replaced with cooling water ( at 30 C ) Many of the students noted this and suggested that it might have caused the temperature of the acrylic acid to reach +90 C causing polymerization When asked to determine this temperature many of the students had a vague idea of how to approach the solution, but only one in twenty was able to fully analyze the problem and calculate the temperature The level of analysis re quired to solve the problem is similar to t h at used by a plant engine e r. We expect our students to under stand and be able to employ s uch an ana ly sis. Figure 2 gives the temperature-dut y profile for the d esign case ( all this information was available to the st u dent ) The temperature of the cooling water is known an d is plotted on the same figure. For the same recycle ratio, the condenser duty is constant TABLE2 is evidence in the ma terial packet given to the student that sup ports this ex pl anation: the amount of fee d has increase d ( fee d may not be pure pro pylene ); the amount of steam generated in the waste heat boiler in creased ( the non-pro pylene material in the feed may not react in the reactor and burns in the afterburner ) ; the non-propylene mate rial was colored Possible Explanations for Off-Spec Product This problem was intentionally set up for the student to dis cover. Poor feed is in fact the most common cause of an off-spec product We shipped our cu s tomer tainted 4 6 P o ss ibl e C au se PRE-PROCESS Feed is n o t pur e PROCESS Evi d e n ce 1 In c r ea s e d fe e d r a t e to maint ain output 2. In crea s e d st e am produ c t io n in wast e h ea t boil e r 3. N e w s uppli e r 4 Tempera tur e an d p ress ur e of s t o r age vesse l h as c h anged R eac tor sys t e m 1 In crease i n fee d ra t e an d l ower co n ve r s ion 2. L o w e r se l ec tivi ty S e parati o n sys t e m s 1. High te mp e r a tur e in Chan ge from refrig e rat e d to c oolin g w a t er p urifi ca t io n co lumn s in co nd e n se r E 7 2. A b s orb e r m al fun c ti on Increase in fee d r a t e and s t ea m p rodu c ti on in was t e h ea t b o il er. POST PRO C ESS Produ c t s to rage Inhibit o r n o t add e d t o p ro du ct in sto r a ge Produ c t tr an s port a t io n Dirt y tank c ar s or p ro du c t ex po se d t o air and s unli g ht C u s tom e r er ror Mi s handlin g o f m a t e rial b y c u s t o m e r OTHER Produ c t not dis c olor e d Ch ec k a nalytical e quipm e nt t es ting pro ce dur es, an d rec o r d s S olu ti on Ch ec k fee d for impuri ti es an d r e turn t o th e s uppli e r R e pl ace spe nt ca t a l yst R e turn to refr i ge r a t e d wat e r Increase the flow of water to absor b ers. C h ec k i nhi bi t or and i nj ect i o n sys t em C h eck tran spor t a t io n co mp any ? R e run sa mpl es in our l a b Chemical E ngineering E ducation

PAGE 49

To achieve a constant duty Q = U A T 10 Where A is constant and U is assumed constant the value ofT 1 0 must be constant. If the curves for the refrigerated water coolant are shifted upward until the low tem perature line corresponds to cooling water, the proc ess temperature at the top becomes 46 C. This is the equilibrium temperature for the exiting acetic acid. Figure 3 provides the vapor pressure curve for the distillate and the bottoms that were obtained from the vapor pressures of the pure components ( and Raoult's law ) From Figure 3, for a temperature of 46 C the condenser pressure is 6.2 kPa. The pres sure drop in the column is largely the result of the liquid level on each plate and the number of plates Since this remains constant ( at 15.9 kPa ), the bot tom pressure is 22.1 kPa. From Figure 3 this corre sponds to a temperature of 95 C. This is well over the 90 C limit for polymerization Some students elected to look for information outside of the material packet that was provided and came across the information that iron enhances po lymerization Iron may have come in with the feed and been the cause of the off-spec material. 46 I 46 u 30 z 26 ... p. El ... Cooling Water (New Conditions) \ I Refr i gerated Water (Design Conditions) j 35 26 15 10 DUTY (Q) Figure 2. T e mp e ratur e vs du ty d ia g ram fo r p r odu c t o ve rh e ad co nd e n se r 2 2.1 k.Pa ... ., ., ... p.. ... 'o 1. 0 p. "' > I I I I I I A crylic A ci d Prod u c t : ( 99 7. pure ) i T~~t P~~~;,Ct I Bottom Produ c t "": (at 95 c) "' ; 0 .1 +---~-~~ ---,---~ ..-~ 2 0 40 60 80 100 Temperatur e C) Figure 3. Vapor pr e ssur e c u rves for a cry li c and a ce ti c a c id produ c t s tr e am s. W i n te r 199 2 Student performance on thi s assignment was poor. Unfortunately, once a single explanation was identified, most students ended their search. No con sideration was given to the economic loss of$500,000/ day when they made their recommendations. Rec ommended changes were sequential and no thought was given to making simultaneous changes to save time and money. Assignment 2 The plant was returned to the original feed mate rial refrigerated water was returned to condenser E 7 and the catalyst was replaced. The high product quality returned and the immediate problem was solved The next assignment was to improve the plant profitability There are recent changes in the value of steam and credit for excess steam will no longer be given. The first step was to identify the major costs under the current conditions or base case. Figure shows the costs of feed material s and utilities tha were calculated from information provided It iden tifies the major costs and shows that the overwhelm ing cost is propylene There was no information on the catalyst that had been replaced. A recommenda tion that serious consideration be given to the devel opment of a new catalyst with higher conversion and selectivity is appropriate. The next largest cost is steam. If large savings are to be realized the purchase of steam must be curtailed. Two methods of reducing the steam costs could be reducing the use of steam or generating steam in the process The acrylic acid reaction is exothermic and the heat generated is removed with cooling water in E-1 (another example of the carica ture nature of the problem ) Replacing E-1 with a steam boiler eliminates the need to purchase steam 8 8 7 7 G) :,.. CD .. 8 cri ., 6 G) G) ... .; p. .., en .. l 5 G) s 5 .., ., .. ., ::J G) 0 4 G) .. je: 0 4 A .., G) G) G) u .... .., A .. ., G) 0 3 0 3 i,:, >, ; G) :!:; A "t:I p.. .0 .9 2 G) 0 ] 2 0 N Sl .. 0 a p., u 1 0 .; A 0 0 Cost of Utilities Cost of Raw Materials Figure 4. Op e ratin g cos t s for t h e d es i g n c as e 47

PAGE 50

Table 3 lists some of the options for using the steam produced to reduce the costs of utilities shown in Figure 4. Student performance on this assignment im proved dramatically from the first assignment. While the assignment was more difficult and the changes were subtle, the students uncovered and corrected many process changes that improved profitability The changes were ranked by incremental return on investment. Assignment 3 This assignment requires the design of a new plant based on new technology The students are free to be as innovative as possible under the given structure. There are new catalysts that provide very high selectivity at low conversion; they are tempera ture sensitive and ideal for a fluid bed. The single reactor system is replaced by a two-reactor system that can operate at different temperatures and pres sures. The students are provided with detailed ki netics along with a computer program that considers both the kinetics and a fluid-bed model. This pro gram allows computation of the reactor size ( given the conversion and operating conditions ). It is clear that the new catalysts provide increased selectivity and, with a recycle stream, the possibility of high overall conversion. Even though the stu dents are provided with the computer program, they will likely fail if they pro ceed to obtain the optimum design by a random approach. Using the kinetic expressions or perform ing a few well-chosen simulations will provide a feel for the effects of operating variables. Solely from the reaction kinetics point-of-view, astounding savings are possible by increasing the recycle. However, once the cost of compressing the recycle gas is considered, the advantages rapidly diminish and a sharp opti mum is reached. In the old plant without recycle, large amounts of steam were introduced into the reactor. With the recycle stream this is not neces sary. The students did well on this assignment. Figure 5 is a typical flowsheet for section -AAof Figure 1, using new technology. It is noticeably different from the old flow diagram. The plant costs are estimated along with the predicted return on the investment. CONCLUSIONS Teaching the student about acrylic acid is not the objective of this course. It is only a vehicle to demon strate how to approach a complex open-ended prob lem While it is true that every process is different it is equally true that the methodology of visualizing, 48 analyzing, and synthesizing are universal and trans ferrable to other processes. The goal of these problems is modest. The stu dent is not expected to master the strategies needed to be effective in working problems of the type out lined. We only hope to introduce the student to these types of problems and to move them to the break point on the learning curve where the rise becomes rapid with a little more experience. We feel that if we have accomplished this, the process has been successful. The final conclusion deals with student motivation. Motivation provides both a high driving force and a low resistance to learning, enabling the student to learn more rapidly. One student's view of the sequence of problems is as follows : On S e pt em b e r 15th 1988 we were pres ented with our first in a se ries of design proje cts. It was apparent that our experience in the classroom had to be challenged by the real-world situation. Instead of performing a "c ookbook TABLE3 Possible Uses for Excess Steam in the Acrylic Acid Plant 1 Place a trim coo l er in se ries with th e waste heat boiler and produce just eno ugh steam for the process. 2. Use any excess steam in a turbine to power the feed air compressor and reduc e the overall power consumption in the plant. 3. Place a preheater on the feed or air streams into the incinerator and use excess steam to reduce th e consump tion of natural gas 4 Use incinerator effluent to pr e heat incinerator feed streams and a trim waste -h eat boiler to make steam balance within the plant. TO H-1 To M-1 Propylene BFlf '-=..2_ CWR steam Fig ur e 5. Modified section (detail -AA-) from Figure 1. Chemical Engineer ing Education

PAGE 51

type problem we were supplied with a minimal amount of information Assumptions answers and suggestions for rather vague situations had to b e made. Th e time to t es t our knowledge in a r e al world sit uation was suddenly upon us. As th e semester progres sed, man y hours of s l ee p were lost, many times tempers b eca m e s hort and often the most docile stud e nt ex hibit e d impatient b e havior. Th e e ffort put forward wa s not wasted though ; the increase in knowledge g r e atly outweigh e d th e amount of time put forth towards the proj ec ts It is for the above reasons that I am pleased to pr ese nt this final in a se ri es of three d es ign projects. The knowl e dge which has accumulated over th e past two semesters will be of great importance in the yea r s to co m e. An expanded version of this paper and copies of the original problems are available from the au thors. ACKNOWLEDGEMENTS The authors would like to acknowledge that the concept of these individual projects or "majors" has been in existence at WVU for almost twenty years and that many faculty have contributed to its suc cess. Most recently, we would like to recognize the contributions of A.F. Galli and W.B. Whiting. REFERENCES 1. Bailie R.C., The Evolution of a Design-Based Curriculum in Ch.E.," Eng. Educ. July / Aug ( 1990 ) 2 Bailie, R.C. "Mining the AlChE Contest Problem for All It s Worth," Proceedings of the 1990 ASEE Annual Conference, Toronto Canada ; p. 506 3. Turton R. and R.C Bailie The Senior Design Problem in Chemical Engineering : A Programmed Approach ," presented at the 1989 AlChE Annual Meeting San Francisco November ( 1989) 0 REVIEW: Chemical Kinetics Continued from page 33. Chapter 5 with catalysis. The former is conventional in its coverage, while the latter introduces autocat alysis and oscillating reactions (a subject of much current interest in chemical engineering) in addition to homogeneous catalysis heterogeneous catalysis and enzyme reactions These first five chapters, two hundred pages in length, comprise nearly half the book. Chapter 6, a brief discussion of the transition from the macroscopic to the microscopic level, sets the stage for the remainder of the text. It introduces the description of reactive collisions in terms of colli sion cross sections and relates the cross section to the rate coefficient. Also introduced at this point are the quantum state description ofreactions (state -to state kinetics ), microscopic reversibility and detailed balance, and a short discussion of the relationship between macroscopic and microscopic kinetics. Winter 1992 Chapter 7 discusses the role of the potential en ergy surface in governing the outcome of reactive events. Succeeding chapters take the microscopic approach to kinetics, covering the kinetics of ele mentary reactions, and chemical dynamics. A treat ment of bimolecular collision dynamics and molecu lar beam scattering is followed by another chapter on experimental methods, this time for state-to-state kinetics. Chapters on transition-state theory ( includ ing variational transition-state theory ) and unimol ecular reactions are followed by one on dynamics in solution and at interfaces. Chapters 13 and 14 cover the advanced topics of the information-theoretic ap proach to dynamics, including surprisal anal ysis and the master equation. The book concludes with a chap ter on applications: atmospheric chemistry and hy drogen and methane combustion. This book has been used for the past two years in a one-quarter course given to chemical engineering graduate students at the University of Minnesota. In the ten weeks of lectures it has not been possible to cover the entire book. The chapters on reactions in solution and dynamics in solution, catalysis (w hich is covered in another course ), information theory, and the master equation have been omitted and supplementary material on bimolecular collision dy namics, chain reactions, and the kinetics of combus tion reactions has been incorporated. It has proved to be a satisfactory text, treating the subject at a level suitable for graduate students and giving more comprehensive coverage ( particularly of modem de velopments ) than many other texts For undergraduates, portions of the book could be employed as the basis for a series of lectures on kinetics that would introduce students to material of interest to engineers in a more modern vein than most chemical engineering kinetics texts. For ex ample, at Minnesota juniors are given approximately six weeks of lectures on kinetics during the spring quarter The first five chapters of this book plus some selected material on rate theory from the chap ters on collision dynamics, transition state theory, and unimolecular reactions (and perhaps atmospheric chemistry and combustion reactions), would provide good reading material to accompany the lectures. Students mastering this material would have an excellent foundation for future work in kinetics and reaction engineering. Chemical kinetics in an enormously broad area. It is difficult to find a text that gives good treatment of the fundamentals of the subject, as well as cover age of subjects of interest to chemical engineers. This book is one that succeeds and can be recom mended. 0 49

PAGE 52

ifii classroom ) EDUCATION IN PROCESS SYNTHESIS Application to Inorganic Processes J.M. GUTIERREZ, J. GIMENEZ, M.A.AGUADO Universidad de Barcelona 08028 Barcelona, Spain T he first task in process design is selecting the raw materials and the reactions and opera tions needed to manufacture the desired product Process synthesis accomplishes this task by gen erating different process alternatives and selecting the best ones. Taking into account the voi d existing in the lit erature about inorganic process synthesis, this pa per is an attempt to show, in a simple and summa rized manner, how the principles of process synthe sis can be applied to inorganic processes. These prin ciples can also be applied in other fields of chemical engineering. Process synthesis uses heuristics, or rules-of thumb" based on experience for the generation and selection of process alternatives. Education in proc ess synthesis implies that students will acquire, in a systematic way, the necessary experience to enable them to apply heuristics, or even to invent" them when the situation demands it. Douglas L 1 21 pro pounded a systematic application of heuristics to process synthesis of organic processes and also indi cated the possibility of using the procedure in chemi cal engineering instruction. The application of process synthesis methods to inorganic processes requires specific considerations due to the different chemistry and properties of inor ganic compounds. However, it seems possible to give heuristics which will permit the student to acquire systematic knowledge about the processes. Raw materials are very important in inorganic processes. Nat u ral compounds are the raw material for one (or a very reduced number) of the derived compoun d s of every element, an d from this (or these) all the other compounds of the element are obtained The natural raw material for an element and its derived compounds is selected according to the eleCo p yrig ht C hE D iv i sion of ASEE 1 992 50 ment concentration, ease of mineral extraction, ease of se p aration of the desired compounds, associated impurities etc The most frequent natural raw ma terials are sulfi d es, oxi d es and hy d roxides, chlorides, carbonates and silicates. The next step in process synthesis of inorganic processes is the selection of the reaction path. Three types of reactions can be considered in inorganic processes: redox, displacement (exchange of ions be tween compounds), and change of crystal structure. For redox reactions, the general rules are : oxida tions must be made at the beginning of the process; reductions will, if possible, be the latest steps of the process; and electrochemical reactions will be the last step. The oxidant or reductor selection can be made using Ellingham's (or a similar) diagram and taking into account the price of the reactants and the ease of reaction-products separation. Air for oxi dations, and H 2 C, or CO for reductions, will be consi d ered first. Displacement reactions are closely Jos e M Gut ie rrez is a professor of chemical engineering at the University of Barcelona (Spain) where he received his PhD in 1984 His major research focus has been i n the areas of colloid chemistry and in chemi c al pro c esses including synthesis and safety. Ja i me G i menez is a professor of chemical engi neering at the University of Barcelona (Spain) where he received his PhD in 1985 His major research focus has been in the areas of kinetics and catalysis and in chemical processes in cluding synthesis and safety M i guel A Aguado is an assistant professor of chemical engineering at the University of B arce l ona (Spain) He received his PhD in 1990 from the University of Barcelona His major re search focus has been in the areas of kinetics and catalysis and in chemical processes includ ing synthesis and safety Ch e mi c al En g in ee ring Education

PAGE 53

related with separation steps and are generally used when a precipitate or an easily separable gas is formed. When a change of crystal structure is re quired, the corresponding reaction will be the latest step in the process. Following the scheme given for organic processes the next step in process synthesis is to decide the species allocation i e. the route of each species through the process Thus quantity and reactive n e ss of impurities permit one to decide if they are processed or not. If there are ions in the solution, every valuable product or by-product and every group of species with waste destination and similar solu bility will constitute an outlet. Recirculation of the mother liquor with a purge to avoid the build-up of impurities can be convenient The general structure of the s eparation system depends upon the phases leaving the reactor In every case, first separation is a phase split, and after that the stream of each phase is driven to the corre sponding separation system ( solid, liquid, or gas ). From each separation system streams of different phases are normally produced and driven again to the corresponding separation system. The same heuristics as in organic processes can be applied when choosing the separation task Crys tallization and precipitation operations followed by solid-liquid separations ( sedimentation or filtration by example ) are used in many cases. In inorganic processes there are normally not a great number of separations and the sequencing problem is not diffi cult. In all cases, heuristics such as the removal of corrosive and dangerous species first removal of the most abundant species, making difficult separations last etc ., can be applied The last step in process synthesis is the inte gration of operations. In the inorganic processes the same general criteria are valid as in organic processes. It can thus be concluded that the student s education in process synthesis can consist of trans mission of the necessary experience together with heuristics or rules-of-thumb. This instruction in pro cess synthesis can be an advantageous substitute for the exha u stive descriptions of chemical processes which is given to students in subjects such as indus trial chemistry. REFERENCES l. Douglas, J M. Hierarchic a l Decision Proc e dur e for Proc e ss Synth e si s," AIChE J. 31 25 3 ( 1985 ) 2. Douglas J M. Co n ce ptual D esign o f C h e mi c al Pro ce s ses, McGraw-Hill N e w York (1 988 ) 0 Wint e r 1992 LETTER: Langmu i r 's Isotherm Continued from page 23 And from Eqs. ( 4 ) to ( 6 ) = K c 1 J -I, ( K-c ) = K c -0 I,K-c( e m J m J J K f j i = l I I J J J i = l I I Solved for 0 J K.c 0. = -~ J ~ J -J m 1 + L (K i c J I =l The concentration of j on th e ad s orbent i s propor t ional to the surface frac ti on occupi e d b y j so that k K c k 0 J J J qj = j j = m 1 + I, (K i ci) i = l ( 7 ) where k. is another constant This is of th e form ofEq. ( 1 ), with k.K corresponding to a and the K to the b. J J J l I This derivation also gives the same result as Langmuir 's for adsorption of a dissociating molecul e, for instance of H 2 as 2 H occupying two sites Langmuir-type adsorption i s seen to result from ideal thermodynamic behavior that i s, from the ab s ence of any specific interactions that would result in activit y -coeffi cient corrections or call for the definition of additional species. This is a more germane reason than are the ad sorption and desorption rates. That adsorption equilib rium has a basis in thermodynamics will be easier for the student to accept and will convey better insight into wh y many adsorbents behave in this manner and wh y some do not. When I went to school mor e years ago than I care to count textbooks still derived the mass-action law for reac tions with the kinetic argument of equal forward and reverse rates at equilibrium-one of them even with the reaction Cu 2 + + 6NH 3 H [cu(NH 3 \ J2 + postulating a single-step mechanism with hepta-molecu lar formation and decay into seven fragments! Happily, we have long outgrown such nonsense and have put the mass-action law for reactions on a sounder basis. Is it not time that we accord the Langmuir isotherm the same courtesy? Frie d rich G. Helfferich Th e P e nn sy l v an ia S ta t e Un ive rsit y 1. Langmuir I. J Am. Chem. Soc. 38 2221 ( 1916 ) 2. For instance see Fogler H.S ., El e m e nt s of Ch e mical Reaction Engine e ring Prentice-Hall Englewood Cliffs NJ, p. 241 ( 1986 ) 0 51

PAGE 54

Award Lecture COMPUTING IN ENGINEERING EDUCATION From There, To Here, To Where? Part 2: Education and the Future BRICE CARNAHAN University of Michigan Ann Arbor MI 48109 W hat have engineering schools / colleges been do ing about the remarkable computing develop ments chronicled in Part 1 of this lecture? With the arrival of the first campus computers in the mid-1950s, many faculty researchers gravitated to computing centers to solve their equation sys tems. The computer has been central to creation of new research disciplines ( e.g., numerical fluid me chanics ) and reorientation of old ones ( such as pro cess analysis and design ) The availability of com puting resources has caused a major shift in the way research is done. On the other hand, it would be inaccurate to claim that undergraduate education has been radi cally transformed by the computer. The impact of computing on engineering education has been much less profound than research. No paradigm shift away from the traditional lecture-recitation-laboratory-tu torial format with the textbook as principal instruc tional resource, has yet occurred. Nevertheless, there has been a gradual infusion of computing work into the curriculum since the early 1960s. THE ACADEMIC COMPUTING ENVIRONMENT The computing experience of our undergraduates has been controlled to a considerable extent by their access to computing facilities: mainframes for re search only in the 1950s ; mainframes accessible for instruction ( initially with batch processing and later with timesharing from hard-wired terminals ) in the 1960s; mainframe time-sharing from remote termi nals, wide-area networks, and decentralized mini computer facilities in the 1970s; personal computer clusters, local area networks, student-owned com puters, and network-accessible supercomputers in the 1980s; workstations, heterogeneous internets, distributed services ( client-server model), intensely graphical interfaces and applications, and parallel Part 1 of thi s le c tur e app e ar e d in th e la s t is s u e of CEE : Vol. 25 ( 4 ), (1991 ), pag e 21 8. Co p y r ig h t Ch E D ivision of ASEE 1 992 52 processing in the 1990s. Every engineering school now has computing facilities for all its undergradu ates though the quantity and quality of the hard ware, software and services vary widely from school to school. A pertinent statisticl 11 : in 1989 there were more than 200,000 university-owned microcomput ers in the US, and a significant fraction of them were in engineering school clusters. EARLY DEVELOPMENTS In 1959 Don Katz ( then chairman o f the Chemi cal Engineering Department at Michigan ) foresaw the tremendous impact that computing would have on engineering practice. He convinced the Ford Foun dation to support a feasibility study of broad-scale integration of computer use into undergraduate en gineering curricula that would make recommenda tions prepare teaching materials and train faculty In a three-year period over two hundred faculty from nine engineering disciplines and s ixty-five en gineering schools participated in the variou s activi ties of the Michigan project. They jointly produced more than 120 completely documented computer problems/programs for classroom use that were widely distributed to other faculty. My first contact with this project occurred in the summer of 1959 when Don offered me a full-time job with his project. My acceptance put my thesis on hold and delayed my PhD by an unconscionable num ber of years But I have never regretted that deci sion. It provided opportunities that I would not have otherwise had and steered me toward an academic career which has brought me much pleasure. Principal recommendations of the project: 1 Train faculty to use computers 2. Provide ("free") time-shared computing services to all students 3. Require a computer-programming course 4. Teach numerical and optimization methods 5. Integrate computing assignments into all engineering, science, and design courses 6. Stress design-like (now called "open-ended") problems throughout the curriculum Ch e mi c al Engine e ring Edu c ation

PAGE 55

Some of these recommendations are a bit dated but most are still on the mark WHAT SHOULD STUDENTS KNOW? According to ABET guidelines, our current engi neering students must 12 l 1 ) have access to computing facilities 2 ) use computers for technical calculations, problem solving, data acquisition and processing, process control, and design and 3 ) be able to apply digital computing techniques to specific problems in a disciplinary program In addition, c hemical engi n ee ring students must be able to 1 ) program in a high-level language, 2 ) use software packages for analysis and design, 3 ) use computing facilities such as editors, and 4 ) simulate engineering problems My own view is that students should be able to formulate algorithms, be familiar with a procedural computer language, and be able to understand and write simple programs in that language. In addition I believe that graduating engineers should have competence in workstation-centered ac tivities having little to do with traditional program writing-in particular Electr o nic communication Preparation of most written materials and reports with editing, word-processing page layout or desktop docu ment preparation and publishing software Drawing, drafting plotting and CAD tools Spreadsheeting for technical and financial analysis Direct access to engineering and other information resources ( e .g ., property and equipment data bases, engineering abstracts library c ard cata l ogues on-line display of journal articles and technical archives) Signal processing and statistical analysis of machineacquired experimental data Symbolic mathematical software Numerical methods and mathematical software Powerful analysis simulation and synthesis packages for solution of complex discipline-oriented engineer ing problems Interpretation of computer-generated visual images as important aids to understanding computed results physical phenomena, and engineering systems Group project work in which the computer is a major factor in effective collaboration Computer-assisted instruction/training (multi-media) Are faculty implementing such recommendations so that our students graduate with computing skills essential for successful careers? A recent ( unpub lished ) CACHE survey of three large petroleum/ chemical companies suggests that industry views recent graduates as having acceptable-to-good com puting knowledge and abilities. My own view is that while our students do graduate with significant com puting skills, we are closer to implementing the let ter than the spirit of even the ABET requirements Win te r 1992 particularly when it comes to integrating computing work into our core engineering science courses BUILDING A NETWORKED INFRASTRUCTURE I believe that a collegeor university wide net worked computing infrastructure is an essential in gredient if we are to be successful. Departmental resources alone will be inadequate for the job that needs to be done. Since both our college and depart mental administrations at Michigan are committed to "computing as a vehicle for significantly improv ing the quality of instruction I would like to relate some of what we have been and are now doing to make this happen The impetus came first in 1983 when our dean ( now President of the University ) brought a group of faculty ( myself included ) together to plan the complete" integration of "computing" into the life of the college ( target date, 1993 ) After lengthy argument and discussion our principal con clusion was : The personal computer/workstation connected to the rest of the world through a hierarchical heterogeneous multi vendor network will be c entral to the e ngineering comput ing paradigm well before the turn of the century. We decided that a networked personal computer/ workstation infrastructure was the first prerequi site for effective implementation of the broad goal of "computing in the curriculum In 1984 the college began to implement the plan b y Startin g to build a college network c onnected to but otherwise independent of the university's central main frame computing facilities Creating a college network management structure that would foster personal academic and research use of the new facilities. This included a full-time network staff and an Executive Committee representing the fac ulty An "Applications Sector" concept for supporting g eneral-purpose commercial productivity and disci pline-oriented professional software ( e .g. CAD, design control ) was also put into place Putting a networked personal computer on the desk of every faculty member in the college. The dean sent out the word that electronic mail should be the principal means of communication for the faculty and staff Putting in place a long-range financial plan. The col lege made a firm commitment to maintain state-of-the art facilities replacing outdated equipment and soft ware with the current "best" affordable Since 1984 our network has grown to a size and complexity that even we did not envision at the outset. Here are some facts and figures on its cur rent status: 1. 100 mbp s F DDI opti ca l fib e r t o k e n rin g b ac kbon e 2. Tw o IBM T o k e n R i n g La n s i n t e rfa ce d to th e b ac kbon e 3 O ver th irty E th e rn e t s ( at l eas t o n e i n each e n gineering bui l in g) in t e r face d t o the backbone 53

PAGE 56

4. Three bridged Apollo workstation 15 mbps op t ica l-fib er token rings gatewayed through Ethernets to th e FDDI backbone 5 Subsidiary LANS ( e.g., office Apple Talk networks) gatewayed to building Ethernets 6. About 2 000 attached machines distributed as follows: 650 in 18 open (24 hour s a da y ) co mputing clusters princi pally for undergraduates 700 in departmental t eac hin g laboratories and in research l aborator i es (principally for graduate s tud ent and faculty re search) 65 0 on fac ult y and professional staff desks, and for co ll ege and departmental adminis trativ e and clerical s taff s. The public workstations directly available to our undergraduate students include the following mix: 125 IBM PS / 2 ( 386DX and 386SX ) 250 Macintosh II (vario us model s) 140 Apollo workstations 50 DEC workstations 50 Sun workstations 20 H ew lett Pack a rd workstations 20 IBM ( RS6000 ) workstations Because all of the high-performance workstations use variants of UNIX for their operating systems, the actual number of computing nodes available to our students from these open machines is much larger than 650 Computing tasks can be distributed over a large number of network workstation nodes ( e.g., 75 RS6000s, several hundred Apollos, many high-per formance computation servers) operating in client server mode. The network also supports a state-of the-art visualization laboratory THE FIRST COMPUTING COURSE With the infrastructure in place, we began the transformation of our traditional programming course into a "computing" course to prepare our undergrads for immersion in our networked computing environ ment Jim Wilkes and myself are responsible for the required computing course for all of our undergradu ates, including those in the Computing Science Divi sion of our Electrical and Computer Engineering Departments Our premises for switching from the traditional programming course format were: 54 1. Engineering computing involves much more than tradi tional programming 2. The networked workstation will be central to engineering computing in the 1990s; every student and practicing engineer will have personal access to one 3. Central mainframes will function principally as informa tion servers to networked clients 4 The predominant operating system characteristics of the future will : support multi-tasking; be multi-user and dis tributed; have windowed graphical user interfaces; sup port pointing and other non keyboard devices. 5. Application programs will have a uniform interface across hardware platforms with: common menu appearance ; common text and graphical element selection; on-line, context-sensitive help 6. The network will support workstation access to any net work file, information, computation server; there will be a common file storage format for all servers. 7. Future workstations will support rapid and natural ma nipulation of text, graphics, sound, video, and other me dia. 8. Cooperative activity will be supported for both communi cation and computation (e.g., group report, data-base, spreadsheet access/generation, design project work--even homework) We decided (after substantial and continuing con sultation with college faculty ) to include the follow ing topics in the first computing course: General computer "literacy" Hardware and network facilities that will be used through out the student's academic career Basic productivity software that will be used later on, regardless of discipline (word processing; drawing, draft ing plotting; spreadsheeting; symbolic math; data-base manipulation; data acquisition/analysis) Programming in a procedure-oriented language Solution of problems designed to expose students to topics from different engineering disciplines Participation in a group activity (e.g. cooperative design and implementation of a problem solution on the com puter) To make the balance of our activities a bit clearer the course syllabus is shown in Table 1. In truth we have not yet been able to include the last two topics without consuming more than a fair ( 3 credit-hour ) share of the student's time during the term. It is likely that we will eliminate some of the introduc tory material and include one or both of these topics in future terms since our students are arriving with more knowledge of computers and more experience in computer programming with each passing year. IMPACT ON ENGINEERING COURSES Almost all of our faculty have taken the produc tivity enhancing services of the networked personal computer/workstation (mail, wordprocessing, desk top publishing, and even spreadsheeting and draw ing applications) in stride much more easily than I would have thought ten years ago Our students have taken to network services like fish to water especially the major productivity tools like word pro cessing, plotting, and spreadsheeting. But what about the impact of computing and the networked environment on engineering course con tent and teaching methodology? In upper-level pro fessional design courses the computer already has had a pervasive influence In effect, most of the college's design and control courses have become computer-aided courses. The computer has also had a significant impact in our undergraduate labora tory courses, which now emphasize experimental Chemical Engineering Education

PAGE 57

design and data analysis. Unfortunately there is still a reluctance by many faculty to truly integrate computing" into the middle level engineering science courses. Historically, our computing activities in these courses have been simi lar to those at many other colleges-hit-and-miss, with little course-to-course coordination A few courses such as our senior/graduate numerical meth ods offering depended heavily on computing assign ments while many courses ignored the computer altogether or had only an occasional computer as signment ( typically the writing of a program ) Around 1986 our department adopted a policy that every core course would involve at least two "computing" problems and at least one "open-ended problem. The computer problems started out as pro gramming assignments for solving assigned analy sis problems, but over the years the trend has been toward use of application packages for solving more realistic (nonlinear) technical problems. THE COMPUTER AS TEACHER There is still a vast potential for enhancing our engineering science courses with computer simula tions and computer-assisted instruction. Bob Seader 21 points the way with the following comment: The microcomputer is the most powerful learning device since the printed textbook If the computer is used prop erly, individualized interactive learning is economically possible for the first time. Substituting "networked workstation" for "mi crocomputer, I agree with Bob. The networked work station with high-quality color graphics ( and in the future, CD-ROM video and sound ) and access to a variety of information servers has the potential of being the near-perfect vehicle for bringing unique "open-ended" problems to individual students who TABLE 1 Freshman C o mputing Course Syllab u s 2-hr classes TOP I C 1 G e n e ral introdu c tion to c omput ers, th e PS /2 2 OS /2, PM i nt e rfac e fil es 2 Word pro ces sin g/ T e xt Editin g 1 N e tworks 1 Internet hardwar e/ software s e rvi ces 1 Ele c tronic mail, on-lin e s e rvic e s of th e univ e rsit y library 9 Programming language (FORTRAN-7 7 or PASCAL) 1 Th e M a cinto s h make the technical decisions while the computer generates alternative solutions and displays them in easily understandable ( principally graphical ) forms. Unfortunately, the creation of good instructional software is very demanding of faculty time, student manpower, and money There is unlikely to be any significant effort outside the universit y to generate such software because there is little prospect of fi nancial return. And since the reward structure at least at research universities discourages such ac tivity, there is little prospect of major faculty efforts to develop instructional software except in associa tion with popular textbooks or when supported by outside funding. Two of our faculty ( Scott Fogler and myselD b gan to work on instructional modules for personal computers several years ago Scott developed a set of several stand-alone modules for his reaction engi neering course. They were well received b y students and have been acquired by many schools that adopted his reaction engineering textbook At about the same time several students and I developed a CAI authoring package and several chemical engineer ing and numerical and optimization modules for stand-alone PCs More recently, Scott received a NSF curriculum improvement grant and has undertaken develop ment of several ( four to six each ) instructional mod ules for introduction to chemical engineering ther modynamics, reaction engineering separations and fluid mechanics. Many of these will be released to the chemical engineering community through CACHE in the next year or so. In the meantime I supervised development of a system-management tool called MicroMENTOR that supports networked delivery of all IBM-oriented soft ware (including Scott s modules ) used in our department. MicroMENTOR allows individual faculty to install software and other instruc tional materials directly on network servers without going through network administra tors to control access to the materials by their students to automaticall y collect s tudent com ments/criticisms and to gather statistics on user activity and software use I believe that the concept of departmentally-controlled in3 Graphi c s (dr a win g, drafting, plottin g, CAD) Claris CAD Delta Graph 1 Non-numeri c al m e thods (sortin g, se arching symbol manipulation structional software embedded in a more glo bal environment of network services will prove to be an effective way to deliver educational 1 S y mboli c math e mati cs Mapl e 2 Spr e ad s h e eting simple financi a l ana l ys i s tool s EXCE L 1 Simple databa se tools ..1... D a t a ac qui s ition an d anal ys is 27 TOTAL Winter 1992 software to our students Two other NSF-sponsored curriculum im provement grants promise to deliver very high 5 5

PAGE 58

quality products that will be widely distributed (through CACHE) at low cost. One project (co-di rected by Rex Reklaitis and Bob Squires at Purdue) is developing comprehensive simulation models of industrial processes The simulations are very flex ible, involve professionally produced videos, and could be used effectively in a simulation lab, design course, or engineering-science courses. The second project (directed by Bruce Finlayson at the University of Washington) focuses on numeri cally intensive solution of general fluid mechanics, reaction, and heat and mass transfer models: its major features are extensive graphical input and high-quality color display for the visualization of computed results In addition, a few faculty scattered across the country are producing very useful instructionally oriented software. I hope that the recent showcasing of some of this software at the national AIChE meet ing in Los Angeles will be an annual event, giving faculty an opportunity to demonstrate their pro grams/lessons to a nationwide faculty audience THE FUTURE What about the future? The track record for long range predictions about computing is only fair at best. Nevertheless, I'd like to end with some predic tions about the future of computing and the impact of computing on engineering education in the short (five to ten years) term. By 1996 1. Microprocessors will be the dominant computing technology; differences in personal computer and workstation character istics (and probably technology) will virtually disappear. Most functions of the simpler microcomputers will be incorporated into a single chip 2. Typical workstation memories will be 16 MBytes; t ypica l operating speeds will be 50-100 MIPS. 3. The dominant mode of professional and academ i c engineer ing computing will be networked and distributed with main frames functioning principally as large-scale networked infor mation servers 4. Character-based applications will disappear; essentially all applications will operate in graphical mode. 5. "Three-dimensional" graphical interfa ces will be available (no glasses). Flat panel displays will b e widely available. 6. The visualization capabilities of th e newest workstations will be widely used for research and educatio n 7. Multimedia workstations will be both available and afford able for instruction, but only in limit ed numbers; high-quality multimedia instructional software will b egin to appear on eng ineering campuses 8. With the advent of digital high-definition television, there will b e a significant merging of video and computing technol ogy. Some campus workstations will have a camera and mi crophone as peripherals. 56 9. New authoring tools will make it much easier to create visu ally attractive and interesting interactive instructional mod ul es (e.g., with animation and video). Standards will be in place for capturi ng and encoding audio and visual informa tion New image compression algorithms will reduce hard ware demands for visual-image storage. Hypertext (non-se quential information representation) tools will play an impor tant role in authoring software. 10 It will still be time-consuming and expensive for fac ult y to design, c r eate, and test discipline-oriented instructional soft ware. Some co ll eges will begin to recognize faculty-authoring efforts as a valued contr ibution at merit salary time (but not as criteria for promotion or tenure). By2001 1 Procedure-language programming will decrease in importance and will probably not be required of students. Object-oriented and visual-programming tools will fill the void left by the decline of algorithmic languages. 2 Interactive instructional modules will have a sign ifi cant im pact on the education and co ntinuing e duc ation of engineer ing students and professionals 3. Workstations operating in th e 1000 MIP/100 MFLOP range will be available on the desktop (and in a briefcase) for about the price equivalent to a current high-end personal computer. 4 Students will take notes on pen-based notebook com put e rs ; in-class transponders will allow immediate storage of figures, assignments etc ., in these comp ut e r s 5. Network transmission rates will increase severa l fold to many gigabit/sec Much of the paperwork (cla ss handouts, home work assignments, collections of assignments) associated with current eng in ee ring classes will be handled e l ectronically using network services. (So far the co mput er h as hardl y pro duced the paperless society-but I think this will happen anyway.) 6. A digital cellular wireless network will allow access to the international Internet from virtually anywhere in the world. 7 Many technical publications will be electronically accessible, with new mechanisms for copyright, royalties, e tc. in place. 8. Spoken-language interfaces will begin to appear. 9. Parallel processors on a s ingl e chip will be common in per sonal workstations. 10 Microcomputers will compete effectively with supercomputers in the solution of many computat ional int ensive problems. Supercomputers will be able to maintain superior speeds but only by increased parallelism They will remain, relatively speaking quite expensive. 11. Massively parallel machin e s will b e available and widely used for engineering research Many will operate at Teraflop speeds. Programming tools and application software will b e slow to develop, but in place and very effective. 12. Some faculty will still not be using comp uters, except grudg ingly (E-mail only?). And finally, no one will remember these predic tions. I will not be asked to comment on them-and a good thing too! REFERENCES 1. Seader, J D., "Education and Training in Chemical Engi neering Related to the Use of Computers ," Comp. Chem. Eng. 13, 377 ( 1989) 2. Seader J.D., "A Brief History of Computing in Chemical Engineering, first Katz lecture, Chemical Engineering De partment, University of Michigan ( April 1990 ) 0 Chemical Engineer i ng Education

PAGE 59

AUTHOR GUIDELINES This guide is offered to aid authors in preparing manuscripts for Chemical Engineering Education ( CEE ) a quarterly journal published by the Chemical Engineering Division of the American Society for Engineering Education ( ASEE ) CEE publishes papers in the broad field of chemical engineering education. Papers generally describe a course, a laboratory a ChE department a ChE educator, a ChE curriculum, research program, machine computation special instructional programs, or give views and opinions on various topics of interest to the profession Specific suggestions on preparing papers TITLE Use specific and informative titles They should be as brief as possible consistent with the need for defining the subject area covered by the paper AUTHORSHIP Be consistent in authorship designation. Use first name second initial and surname. Give complete mailing address of place where work was conducted. If current address is different, include it in a footnote on title page. TEXT Manuscripts of less than twelve double-spaced typewritten pages in length will be given priority over longer ones. Consult recent issues for general style. Assume your reader is not a novice in the field. Include only as much history as is needed to provide background for the particular material covered in your paper. Sectionalize the article and insert brief appropriate headings TABLES Avoid tables and graphs which involve duplication or superfluous data. If you can use a graph, do not include a table. If the reader needs the table, omit the graph. Substitute a few typical result s for lengthy table s when practical. Avoid computer printouts NOMENCLATURE Follow nomenclature style of Chemical Abstracts ; avoid trivial names. If trade names are used define at point of first use. Trade names should carry an initial capital only with no accompanying footnote. Use consistent units of measurement and give dimensions for all terms. Write all equations and formulas clearly and number important equations consecu tively ACKNOWLEDGMENT Include in acknowledgment only such credits as are essential LITERATURE CITED References should be numbered and listed on a separate sheet in the order occurring in the text COPY REQUIREMENTS Send two legible copies of the typed ( double-spaced ) manuscript on standard letter-size paper. Clear duplicated copies are acceptable. Submit original drawings ( or clear prints ) of graphs and diagrams, and clear glossy prints of photographs. Prepare original drawings on tracing paper or high quality paper; use black india ink and a lettering set. Choose graph papers with blue cross-sectional lines ; other colors interfere with good reproduction. Label ordinates and abscissa s of graphs along the axes and out s ide the graph proper. Figure captions and legends may be set in t y pe and need not be lettered on the drawings Number all illustrations consecutively. Supply all caption s and legends typed on a separate page.

PAGE 60

N FCFCC>S V\TO:R::K.. Long before most people even recogni7.Cd the problems CH2M HILL was developing innovative solutions for the world 's changing waste water management energy, transportation and laboratory needs It's a rac e against time now But our over 40 years experience in environmental consulting engineering make s it a race we're well positioned to win. Challenging projects .. early significant respon s ibility ... the opportunity to work with top professionals in a creative, yet stable, environment ... there are many excellent reasons to begin your career with CH2M HILL. Perhaps the most important is the role you'll have in shaping the engineering for a new world. With 60 offices heading more than 4,000 projects CH2M HILL has U.S openings in the following disciplines : Civil Engineering (including specialization in environmental, structures, water resources, transportation, geotecbnical, ports, harbors and hydraulics) Mechanical, FJectrical, Agricultural and Chemical Engineering Geology and Geological Engineering Planning Scienc:es Economics Computer Engineering Requirements include a BS degree in engineering from an ABET Engineering program A Master's degree is preferred for most specialtie s. As a member of our employee-owned corporation you ll enjoy a competitive starting salary, attractive bonuses and flexible benefits We invite you to learn more about CH2M HILL's current staffing needs by sending your resume and salary history in confidence to : Staffing Manager CEE; 291, CH2M IULL, P.O. Box 2211ll, Denver, CO 80222-9998. -If you have a PC and a modem find out more about CH2M HILL and other opportunities we have available. Just dial (603) 432 2742 pre s s "return twice and enter "water when prompted for a password. An Equal Opportunity Employer Pure Challenge