Chemical engineering education

http://cee.che.ufl.edu/ ( Journal Site )
MISSING IMAGE

Material Information

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

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:
oclc - 01151209
lccn - 70013732
issn - 0009-2479
Classification:
lcc - TP165 .C18
ddc - 660/.2/071
System ID:
AA00000383:00119

Full Text
















































































L 19












ACKNOWLEDGEMENT


DEPARTMENTAL SPONSORS

The following 154 departments contribute to the support of CEE with bulk subscriptions.

If your department is not a contributor, write to
CHEMICAL ENGINEERING EDUCATION,
c/o Chemical Engineering Department University of Florida Gainesville, FL 32611-2022
for information on bulk subscriptions


University of Akron
University of Alabama
University of Alberta
University of Arizona
Arizona State University
University of Arkansas
Auburn University
Brigham Young University
University of British Columbia
Brown University
Bucknell University
University of Calgary
University of California, Berkeley
University of California, Davis
University of California, Irvine
University of California, Los Angeles
University of California, San Diego
University of California, Santa Barabara
California Institute of Technology
California State Poly Institute
California State University
Carnegie-Mellon University
Case Western Reserve University
University of Cincinnati
Clarkson University
Clemson University
Cleveland State University
University of Colorado
Colorado School of Mines
Colorado State University
Columbia University
University of Connecticut
Cooper Union
Cornell University
Dartmouth College
University of Dayton
University of Delaware
Drexel University
University of Edinburgh
University of Florida
Florida Institute of Technology
Florida State/Florida A&M University
Georgia Institute of Technology
University of Houston
Howard University
University of Idaho
University of Illinois, Chicago
University of Illinois, Urbana
Illinois Institute of Technology
University of Iowa
Iowa State University


Johns Hopkins University
University of Kansas
Kansas State University
University of Kentucky
Lafayette College
Lakehead University
Lamar University
Laval University
Lehigh University
Loughborough University
Louisiana State University
Louisiana Technical University
University of Louisville
Lowell University
Manhattan College
University of Maryland
University of Maryland, Baltimore County
University of Massachusetts
McGill University
McMaster University
McNeese State University
University of Michigan
Michigan State University
Michigan Technical University
University of Minnesota
University of Minnesota, Duluth
University of Mississippi
Mississippi State University
University of Missouri, Columbia
University of Missouri, Rolla
Montana State University
University of Nebraska
University of New Hampshire
University of New Haven
New Jersey Institute of Technology
University of New Mexico
New Mexico State University
North Carolina A & T University
North Carolina State University
University of North Dakota
Northeastern University
Northwestern University
University of Notre Dame
Technical University of Nova Scotia
Ohio State University
Ohio University
University of Oklahoma
Oklahoma State University
Oregon State University
University of Ottawa
University of Pennsylvania
Pennsylvania State University


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












EDITORIAL AND BUSINESS ADDRESS:
Chemical Engineering Education
Department of Chemical Engineering
University of Florida
Gainesville, FL 32611
FAX 904392-0861

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
Angelo J. Perna
New Jersey Institute of Technology
Stanley I Sandler
University of Delaware
Richard C. Seagrave
Iowa State University
M. Sami Selim
Colorado School of Mines
James E. Stice
University of Texas at Austin
Phllip C. Wankat
Purdue University
Donald R. Woods
McMaster University



Summer 1993


Chemical Engineering Education


Volume 27


Number 3


Summer 1993


DEPARTMENT
154 M.I.T.'s School of Chemical Engineering Practice,
John I. Mattill

EDUCATOR
160 Michael B. Cutlip, of the University of Connecticut,
Lucinda Weiss

TEACHING
164 Seven Rules for Teaching, R. Byron Bird

DESIGN
166 The Technically Feasible Design, TW Fraser Russell, N. Orbey

CLASSROOM
170 Process Control Education: A Quality Control Perspective,
Pradeep B. Deshpande

188 Integrating Communication Training into Laboratory and Design Courses,
Karen R. Pettit, Richard C. Alkire

200 Grand Words, But So Hard to Read! Diction and Structure in Student
Writing, Aloke Phatak, Robert R. Hudgins

204 Safety and Writing: Do They Mix? Robert M. Ybarra

216 Computing Teaching with Fortran 90, lan Furzer

220 Simulation in the Chemical Engineering Classroom, Wallace B. Whiting

RANDOM THOUGHTS
176 Teaching Teachers to Teach: The Case for Mentoring, Richard M. Felder

CLASS AND HOME PROBLEMS
178 When is a Theoretical Stage Not Always a Theoretical Stage? W.E. Jones

LABORATORY
184 A Comprehensive Process Control Laboratory Course, P. T. Vasudevan

194 Experience With a Process Simulator in a Senior Process Control
Laboratory, Suresh Munagala, Daniel H. Chen, Jack R. Hopper

OUTREACH
210 A Unit on Acid Rain in a High School Outreach Program,
John A. Marsh, Michael A. Matthews

BOOKREVIEWS
182, 183, 199


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 ofFlorida, Gainesville,
FL 32611. Copyright @1993 by the Chemical Engineering Division, American SocietyforEngineering
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.
153















M. I. T.'s

SCHOOL OF

CHEMICAL ENGINEERING PRACTICE


The Powerful Potential of Alumni Support
...or....How Its Graduates Matched Their Enthusiasm with Their Money

JOHN I. MATTILL
Massachusetts Institute of Technology
Cambridge, MA 02139

In January of 1980, Selim Senkan and J. Edward Vivian wrote an up-beat description of MIT's School
of Chemical Engineering Practice (SCEP) for this journal.m What they did not emphasize in that paper
was that this unique educational program, then in its 64th year, had some threatening liabilities as well
as the assets they so ably celebrated. When some of those liabilities materialized in the next decade,
graduates of the Practice School, many of them among the top leadership of the U.S. chemical industry,
moved aggressively to support this unique concept in chemical engineering education. Indeed, the School
today is an example of the influence that alumni can have on professional education in chemical engineer-
ing.
To tell this story is, in fact, to tell a brief history of the School. What follows is a radical condensation of
the history prepared by the author as a complement to SCEP's 75th anniversary celebration in 1991.121


From the beginning of instruction at the Mas-
sachusetts Institute of Technology in 1865,
there was an option in "practical and indus-
trial chemistry," and by 1888 it had become the
nation's first four-
year curriculum in
chemical engineer- -
ing. Beginning in
1884, its head was
William H. Walker,
an entrepreneurial
analytical chemist
trained at Penn
State and the Uni-
versity of Gottingen.
One of Walker's major concerns in teaching chemi-
cal engineering was to help students understand
how chemistry was different when scaled up to in-
dustrial dimensions. For many years, he gave his
students a sense of the industrial environment by


taking them on week-long tours of major chemical
plants in the Northeast. But by 1914 enrollment
had become so large that the difficult logistics of
such tours proved insoluble, and they were termi-
nated.


The School of Chemical Engineering
Practice was conceived jointly in 1915
by William Walker (L) and Arthur D.
Little (R), whose own career proved
that even young people without chemi-
cal degrees could contribute signifi-
cantly to the chemical industry.
Photographs courtesy M.I.T. Museum


Walker, how-
ever, remained
concerned about
how to best intro-
duce his students
to chemistry in
industry. His
friendship with
Arthur D. Little,


who had entered
M.I.T. in 1881 to study industrial chemistry, was to
eventually lead him to the solution to that problem.
Little never finished the four-year curriculum-fi-
nancial needs and his impatience with academics
led him to go to work in 1884 as assistant chemist in


Copyright ChE Division ofASEE 1993


Chemical Engineering Education


[e, departmentt


154




































Photograph courtesy M.I.T. Museum
In the early years, the Practice School was concerned as
much with teaching the scale as the sophistication of in-
dustrial chemistry. These students in the School's first
post-World-War-I class were photographed at the Revere
Sugar Refinery near Boston.

a small paper mill in Rhode Island where, despite
his modest academic credentials, he almost single-
handedly perfected the plant's sulphite papermak-
ing. Soon thereafter, Little became a pioneering and
successful consultant, and he presently was chosen
for membership on M.I.T.'s Corporation-its board
of trustees.
As Little's success demonstrated, the country's fast-
growing chemical industries were desperate for tech-
nical help-just as Walker was desperate to give his
students industrial experience. The logic was ines-
capable, and Little and Walker devised an elegantly
simple exercise of it. M.I.T. would establish branches
(they were called "stations") at several chemical
plants. Faculty would be augmented so that two
teachers could be in residence at each station, and
groups of Master's students would visit the stations
to learn plant operations under supervision of the
resident faculty. In between terms and during sum-
mers the resident faculty would work on technical
problems for the host companies; they would likely


Summer 1993


As Little's success demonstrated, the country's
fast-growing chemical industries were desperate
for technical help-just as Walker was desperate
to give his students industrial experience. The
logic was inescapable, and Little and Walker
devised an elegantly simple exercise of it.


be the companies' most sophisticated research and
development people, and if they needed even more
expertise, it could be obtained from M.I.T. colleagues.
The companies would meet the stations' operating
costs, and M.I.T. would pay the faculty salaries.
Little solicited a $300,000 gift (a prodigious sum
in 1916 dollars) from George Eastman to build the
needed stations. As it turned out, the companies in
their enthusiasm built the needed stations them-
selves (offices, libraries, adjacent small laboratories),
and Eastman's gift became a useful nest-egg for the
Practice School. The scheme quickly drew the ap-
proval of M.I.T.'s faculty, administration, and Cor-
poration, and a communication to the London Times
Engineering Supplement applauded the experiment
for chemical engineering students "who have no
doubt found that dexterity with flask and test tube
does not create precisely the self-confidence needed
by the chemist who is working with, say, 25,000
gallons of acid in a digester."[3'
During their six weeks at each station, the stu-
dents' assignments included creating and drafting a
plant flow sheet, laboratory exercises using the
plant's test equipment, lectures by the faculty and
selected company staff followed by a series of "home
quizzes," and group work on several plant problems
that typically involved measuring the effects of
changes in one or more process parameters. Each
student served at least once as a project leader for a
group of three to five colleagues, and each group
was required to make formal presentations of project
plans and progress reports in addition to verbal and
written final reports. Students often had to devise
and build the test equipment they needed, and ev-
ery student worked on at least one problem that
required taking data for a 16-to-24-hour period.
Alumni complain that there was never enough time
to do everything. But the faculty were unrespon-
sive; they wanted the program to replicate the char-
acteristics of professional work-the pressure for re-
sults, the need to innovate technical methods, and
the problems of group leadership, project planning,
and technical reporting. Alumni almost without ex-
ception suggest that this mission was accomplished;

155









the School gave students confidence and
a powerful enthusiasm for the profes-
sion they were entering.
Some interesting comments by stu-
dents in the Practice School's first class
in 1917 are:
* Every member of the group is impressed
with the change from the theoretical view-
point of the classroom to the practical
viewpoint of the course.
To say a Guy-Lussac tower is so many
feet high is one thing but to climb it is
another.
We are gaining an interest in our work
that has never been equalled, and we are
gaining a friendship with men of impor-
tance in our profession.
As late as 1949, Gerald Lessells, now
retired, was having experiences that
were typical of SCEP's earliest years:
"We had been working since eight the
previous morning, getting ready for a
stream-flow measurement in a high-pres-
sure steam line. After machining our own
orifice and setting up for pressure-drop
measurements, we stood aghast in the
small hours of the next morning as our
sole achievement was to blow the mer-
cury in the manometer into the steam
line. We quit, almost in tears. But we
finished successfully the following day.
That was forty-one years ago, and I still
can remember the frustration, and later
the sense of fulfillment, when we reached
our goal."
Since then there have been evolution-
ary changes. Today, the Practice School
programs focus almost entirely on prob-
lems suggested to the resident faculty
by company technical personnel who
then become the students' consultants
on the projects. Students' reporting ses-
sions are, in effect, plant seminars at-
tended by both company personnel and
M.I.T. representatives; Practice School
alumni are especially enthusiastic about
their experiences in preparing and pre-
senting reports, the final versions of
which ended up in the host companies'
proprietary files. Practice School faculty
have no roles in companies' research ex-
cept to help identify technical problems
suitable for student projects and to help


Photograph courtesy Oak Ridge National Laboratory, from M.I.T. Museum
Reporting project results to colleagues and company hosts is cited by
alumni as an important contribution of the Practice School. Shown
above is Elsa Kam-Lum, the second woman to attend the Practice
School, at the Oak Ridge Station, 1973.

students fulfill the
companies' needs.
The students, whose
role is likened to that
of outside consultants,
are unpaid-which .
gives them license
to continue the tra- -
ditional complaints
about intense pressure
and overload. But they
continue to draw con-
fidence from their work
(which typically be-
comes company prop-
erty) and to take great
pride in it. Projects of
Practice School stu-
dents at American
Cyanamid's Bound Photograph courtesy Bethlehem Steel Corp., from M.I.T. Museum
Brook plant between A typical student-faculty ratio and relation-
1962 and 1967 were ship: Professor George Huff ('82) with three
said to have saved that students at the Bethlehem Station (1982-
company an average of 1984).
$160,000 a year, and
savings of millions of dollars are attributed to two student projects
at Dow Chemical Company in the 1980s.
In the 1960s, as the Practice School passed its 50th birthday,
however, a host of problems began to press on it:
This was a time of growing emphasis on "engineering sci-
ence," especially at M.I.T. Practical experience such as
emphasized by the Practice School was out of style, and many


Chemical Engineering Education










students (and some staff as well) thought SCEP was
irrelevant. Enrollment fell.
* In order to attract students, M.I.T. asked the host
companies for help with the students' expenses at the
stations, and companies subsequently agreed to
provide funds that could be awarded as fellowships to
cover tuition and a part of living expenses. But tuition
was rising faster than the rate of inflation, and
companies found these rising commitments onerous.
* Despite these stipends for their semester at the


Photograph courtesy Esso, from M.I.T. Museum
Professor Warren K. Lewis ('05), who taught at M.I.T. from 1908
until well beyond his official retirement in 1948, regularly vis-
ited the stations-shown here at the Bayway Station in 1959.


stations, students found the Practice School an
expensive option. Away from the campus for one
semester, they were poor candidates for on-campus
research or teaching assistantships that were avail-
able to most other graduate students.
The Practice School was clearly a cost center for M.I.T.
as well as for its host companies. With two members of
the faculty at each station, Practice School students
enjoyed the Institute's lowest student-faculty ratio,
and higher housing costs resulted from the arrival of
women and married students. The Eastman funds
were long gone.
More and more foreign students came to M.I.T., and
far more in proportion than American students sought
out the Practice School as a way to learn about
American industrial practice. But to the companies
foreign students were vexatious-unlikely to be
available for employment after graduation and very
likely to carry American methods back home to
overseas competitors.
For all these reasons, by the late 1970s SCEP began to
look to the M.I.T. administration more like a liability than
an asset, and its termination seemed likely.
But its alumni had not yet been heard from, and almost
Summer 1993


from the year of its founding the Practice School
was distinguished by the enthusiasm of its
former students-an esprit probably greater
than among the alumni of any other graduate-
level program at M.I.T. Fully ninety percent of
the funding for the department's new building
in Cambridge, dedicated in 1976, had come from
Practice School alumni or companies that they
founded. When queried in 1991 (in anticipa-
tion of the School's 75th anniversary), an ex-
traordinary number of them wrote enthusias-
tic recollections, saying that their Practice
School experiences had been pivotal in shaping
their careers.
Peter Melnick ('52, Hercules, Inc.) credited
SCEP with "a hands-on practical experience
that opened up the real world of industrial
manufacture, revealing how everything in en-
gineering is tied together." Ralph Landau ('41)
said, "I never worked so hard in my life, but I
really learned how to concentrate and get a job
done under forced draft." Vernon Bowles ('33)
remembers the Practice School as "the great-
est experience of my educational encounters."
Because they were prominent in the profes-
sion, SCEP alumni were prominent in the coun-
cils of M.I.T.-including especially the
Corporation's Visiting Committee to the De-
partment of Chemical Engineering. Unmoved
by an estimate that $180,000 a year might be
needed to overcome the problems that beset
the School, they stonewalled any suggestion of
terminating what Professors Senkan and Vivian
had called "a continuing catalyst in engineer-
ing effectiveness."[41
One of the trump cards was played by Charles
Reed ('37), whose doctorate in chemical engi-
neering from M.I.T. had not included SCEP
experience. In 1977, as General Electric's se-
nior vice president for corporate technology, he
had invited SCEP to open a station at GE's
chemical plants in Waterford and Selkirk, New
York; he thus rescued the Practice School from
the embarrassment of a two-year search to re-
place its station at Bound Brook, New Jersey,
terminated when American Cyanamid found
the escalating costs too high. Upon hearing the
project reports by the first class at the
Schenectady station, Reed wrote M.I.T. that
the students "did an extraordinarily good job of
presenting (their) results and recommendations.
I was really delighted." Returning to his office
from Schenectady after a similar session the
157










The decisive event was the commitment by John Haas ('42), then vice-chairman of Rohm and Haas,
to head a fund-raising effort among companies in which SCEP alumni held major posts.
Haas had come to M.I.T. from a liberal arts background, and he says "I didn't
know what a reactor was until I went to the Practice School."


next year, Reed reported to M.I.T. President Jerome
B. Wiesner, "I was tremendously impressed with
the great range and high quality of the projects
being worked on .... (The students') studies have
resulted in recommendations expected to (yield) sav-
ings of $400,000 to $700,000 a year .... (The Prac-
tice School provides) a most important type of expe-
rience that many of us wish we could have had at an
early age. In my opinion, this is really unusual and
highly valuable graduate education."
The decisive event was the commitment by John
Haas ('42), then vice-chairman of Rohm and Haas,
to head a fund-raising effort among companies in
which SCEP alumni held major posts. Haas had
come to M.I.T. from a liberal arts background, and
he says "I didn't know what a reactor was until I
went to the Practice School." He, Landau, and twelve
other prominent alumni, establishing themselves as
"Friends of the Practice School," in 1980 completed
a $600,000 fund for fellowships for Practice School
students while in Cambridge. Donor companies
received only one "perk" as an incentive-they
had first review of the resumes of all M.I.T. chemi-
cal engineering graduate students about to receive
their degrees.
But the Friends' fund was a wasting grant that
would soon enough be exhausted and plunge SCEP
back into uncertainty. So in 1981, Haas made a new
proposal. The Phoebe Hass Charitable Trust, he said,
was considering a $500,000 grant to M.I.T. If he
persuaded his M.I.T. undergraduate classmates to
match that gift for their 40th M.I.T. reunion, would
the Institute commit the resulting $1 million Class
of 1942 Professorship to a member of the faculty
who would have the goal of stabilizing SCEP opera-
tions within five years? After some frustrating
months of indecision, the Institute administration
accepted this proposal, and Jefferson Tester ('71)
was recruited from Los Alamos to be Class of 1942
Professor and director of the Practice School. Tester
never studied in the Practice School, but he had
served two years as a station director after complet-
ing his doctorate at the Institute, and his enthusi-
asm for the Practice school was unbounded.
During his first year as director, Professor Tester
Changed the SCEP curriculum so that the School


At the Practice School's 75th anniversary celebration in
1991 are (left to right) David Koch ('63) of Koch Industries,
Inc., Jean Leinroth ('48), director of summer stations at
Syntex Chemicals and Chevron, and Professor Jefferson
Tester ('71), Practice School director from 1980 to 1989.

could serve three groups of students: outstanding
undergraduates who would study for five years at
the Institute, including one term at the Practice
School, and receive both bachelor's and master's
degrees; M.I.T. doctoral students, who would study
for a one term at the Practice School in order to gain
a sense of industrial practice available to few ScD
and PhD candidates; and graduate students who,
after completing undergraduate degrees elsewhere,
would come to M.I.T. for master's degrees in chemi-
cal engineering practice, studying for two terms in
Cambridge and one summer at the Practice School
stations and thus making SCEP a year-round
activity.
Raised the salaries of Practice School station
directors so that they related not to faculty salaries
at M.I.T. but to industrial salaries for people of
comparable experience in the plants in which they
served.
Increased the budgets of SCEP's stations to include
travel and some of the professional/social occasions
that animated the Practice School of the 1930s and
1940s, when the Eastman funds had been available.
Raised the visibility of the Practice School by a
variety of strategies that reflected Tester's confi-
dence in and enthusiasm for the program.
Worked with alumni and M.I.T.'s fund-raising
apparatus to catalyze two separate fund-raising
efforts. The first reactivated the Friends of the


Chemical Engineering Education




























Photograph buy Carole Williams, from Chevron Focus
Shown here on their first day at the Richmond (California) Station,
students tour the Chevron Refinery, 1989.

Practice School organization (Robert Richardson, '54, then execu-
tive vice-president of Du Pont, became chairman) to fund for
several more years the corporate-sponsored fellowships first
established in 1980. The second-and far more ambitious-effort
was to raise from individual donors (corporate gifts were not
solicited) an $8 million endowment to permanently underwrite
fellowships for SCEP students during their Cambridge studies.
This task was accepted by the Corporation's Visiting Committee,
whose chair was Jerry McAfee ('40), retired chairman and chief
executive officer of Gulf Oil Company.
Though it is far easier in the telling than it was in the doing,
the final result was celebrated late in 1990 when the endowment
was completed with a major gift from David H. Koch ('63) execu-
tive vice president of Koch Industries, Inc., leading to the
School being renamed in his honor. "There was nowhere else in
my M.I.T. experience," Koch told me, "where I had the chance
to test my technical abilities, and I figured any educational
experience that was this powerful for me might be of similar
value to others."
With the endowment complete, the David H. Koch School of
Chemical Engineering Practice entered the 1990s with its an-
nual funding of about $1.3 million coming roughly in equal parts
from endowment income, host companies, industrial fellowship
grants renewing those obtained by the Friends, and M.I.T. re-
sources. The endowment income and industrial grants cover sti-
pends for Practice School students while studying in Cambridge;
the host company funds are used by M.I.T. for fellowships for
students at the stations, and Institute funds cover SCEP faculty
salaries and benefits and administrative expenses.
As of 1993, the David H. Koch School operates year-round
stations at Dow Chemical Company and neighboring Dow-Corn-
ing Company, Midland, Michigan, and Merck and Company's
pharmaceutical operations at West Point, Pennsylvania. Annual
enrollment is typically between thirty and forty, and the waiting
Summer 1993


list extends well into 1994. Each student
spends eight weeks at each station, nor-
mally working on two four-week projects
in two different groups. Thus each stu-
dent has the experience of group leader-
ship once during his or her term at the
stations.
Essentially all M.I.T. Master's candi-
dates in chemical engineering attend the
Koch School, and two-thirds of all Doctor's
candidates do so. Its director is T. Alan
Hatton, Chevron Professor of Chemical
Engineering at M.I.T., whose enthusiasm
for the Practice School was developed as
a station director during the summers of
1983 and 1984.
Perhaps the best recent summary of the
School's status was given by Professor Jef-
frey Feerer, associate director of SCEP
from 1989 to 1992, at a 1990 conference
on national materials policy: "For almost
seventy-five years this chemical engineer-
ing internship program has directly trans-
ferred innovation and technology from the
universities to the production floor, and it
has educated chemical engineering stu-
dents to the specialized and complex prob-
lems of chemical manufacturing. In doing
so, it has provided a unique link between
the narrowness of graduate chemical en-
gineering education and the breadth of
activities in which chemical engineers par-
ticipate in the workplace.
"The Practice School is today more
vibrant than at any time in its his-
tory, thanks in part to a legion of
alumni/ae who celebrate the value that
the Practice School experience has had in
their careers."'

REFERENCES
1. Senkan, Selim M., and J. Edward Vivian, "MIT
School of Chemical Engineering Practice,"
Chem. Eng. Ed., 14, 200 (1980)
2. Mattill, John I., The Flagship: The M.I.T.
School of Chemical Engineering Practice, 1916-
1991, Dept of Chem. Eng., M.I.T. (1991)
3. Quoted in Tech. Rev., 18, 885, November (1916)
4. Senkan and Vivian, op cit., 200
5. Feerer, Jeffrey, "The M.I.T. Practice School: A
Unique Partnership of Education and Indus-
try." Unpublished paper for the Workshop on
Education for National Materials Processing
Efficiency, 11th Biennial Conference on Na-
tional Materials Policy, Federation of Materi-
als Societies, June (1990) 0










IM educator
---


Michael B. Cutlip

of the University of Connecticut

by Lucinda Weiss*


he teach-
ing "bug"
bit Mike
Cutlip at an early
age. He was tak- _
ing sophomore V
calculus at Ohio
State University .-
in 1960 when his
teacher, Marg-
aret Jones, in-
spired by the way
he explained
problems at the
board, asked him Cutlip and his father, Sidney,
to teach an intro- years
ductory math
class. He did such a good job that he was sub-
sequently hired as a "student assistant" to help
teach algebra and trigonometry at OSU. On spring
breaks, when others headed to Fort Lauderdale for
fun and games, he went home to Milford, Ohio, a
suburb of Cincinnati, and taught math or science at
the high school as a substitute teacher when the
opportunity arose.
"I liked it. It kind of got me started in teaching,"
he recalls.
It began a career devoted to chemical engineering
education in which Cutlip has combined a research
interest in catalytic reactions with computer-based,
self-paced educational programs for chemical engi-
neering students. He has coauthored personal com-
puter software that is presently used by more than
one hundred chemical engineering departments, and
he is president of CACHE Corp., a non-profit orga-
nization based at the University of Texas, Austin,
which promotes the application of computing tech-
nology to chemical engineering education.
When Cutlip started taking chemical engineering
courses at OSU, the slide rule still represented the
* 128 Davis Road, Storrs, CT 06268
160


Teprf
of te


prevailing tech-
nology. He ma-
jored in chemi-
.- cal engineering
h in a five-year
program that
combined the
bachelor's and
master's de-
grees. Joseph H.
Koffolt was the
department
chairman at the
time, and he was
renting (so far) over seventy-five known for re-
aching. membering all
his student's
names and calling them "his jewels." Cutlip still
runs into his former classmates on occasion. "There
are still a few gems around," he says.
But an even earlier influence on his decision to
teach was his own father. Sidney Cutlip was the
high school principal and was Mike's algebra teacher
at Milford High. He, too, had started teaching at
nineteen-in a one-room school that included all
eight grades. He worked on his bachelor's degree
during the summer months and eventually earned
his master's degree. Now eighty-five and retired, his
teaching career spanned forty-four years.
"I have a lot of respect for my father," Cutlip says.
Cutlip's high school chemistry teacher, Mary
Moore, encouraged him to pick a profession such as
chemical engineering. "Initially, I thought of high
school teaching, but she encouraged me to consider
other options as well," he says. One of those options
was aeronautical engineering, but Cutlip particu-
larly liked freshman chemistry and had the lucky
foresight to reject aeronautical engineering as too
dependent on government grants.
He also decided to study for a PhD so that he


Copyright ChE Division ofASEE 1993
Chemical Engineering Education









could eventually do research or teach, hoping to even-
tually expand his horizons beyond the Midwest. Dur-
ing one spring break at OSU he fortuitously visited
the University of Colorado where Max S. Peters was
the new Dean of Engineering. Peters was very sup-
portive of young Cutlip's goals and later became
Cutlip's PhD adviser.
But it was a ski-run down Aspen Mountain that
clinched Cutlip's decision to attend graduate school
at the University of Colorado. Cutlip and a fellow
OSU chemical engineering student, Alkis
Constantinides (now a chemical engineering profes-
sor at Rutgers University) had learned to ski on
Ohio's Mt. Mansfield. He says, "For a guy who had
skied before but who had only experienced a vertical
drop of about 150 feet, skiing from the top of Aspen
Mountain was an unforgettable experience!"
Cutlip did both his graduate and post-graduate
work at Colorado, studying the catalytic properties
of polymeric materials and looking for new catalytic
materials. Cutlip points out that his adviser, Peters
(who continued as Dean and who became vice presi-
dent and then president of AIChE in 1967 and 1968),
was good at delegating responsibility and often asked
the graduate students to help write reports and de-
velop proposals in addition to conducting their re-
search. "What I learned from him was organiza-
tion," Cutlip says.
Peters also helped Cutlip find his first teaching
position. In 1968 Cutlip became assistant professor
at the University of Connecticut, and was later
appointed associate professor, then professor-and
for "nine long years," he jokes, he was department
head. He stepped down from that administra-
tive post in 1989.
New England was quite a change of pace for Cutlip
and his wife, Susan, who grew up in Denver and
met Mike at the University of Colorado. On his first
visit to UConn, Cutlip was surprised by its rural
character. "I thought it would have been paved from
New York all the way up through Connecticut," he
says. When he got back to Colorado from that initial
visit, he told Susan that he had driven around the
area a little bit, "but I didn't see downtown Storrs."
He didn't realize that "downtown" Storrs was a small
grocery store, a movie theater, and a traffic light!
The chemical engineering department at UConn,
then headed by Leroy Stutzman (now retired), was
still relatively young, having seen its first students
graduate in 1963. Cutlip pursued his interest in
reaction engineering, applying catalysis to air pollu-
tion control. Among his projects at UConn has been
a long-standing interest in the electrochemical pro-
Summer 1993


Cutlip developed a self-paced, mastery-oriented
course in reaction engineering for seniors.
[They] could come into the computer lab
at any time, go through the tutorials at
their own pace, and then take assignments
and quizzes as they were ready for them.

cesses in hydrogen/oxygen fuel cells and in looking
for ways to improve their efficiency. As a result of
this interest, he is heavily involved with the Pollu-
tion Prevention Research and Development Center,
a new center in UConn's Environmental Research
Institute that is supported by the EPA.
When he came to UConn, he was encouraged by
Stutzman, who was a consultant and later a board
member at Control Data Corporation, to research
the potential of computers in education. Stutzman
introduced him to CDC's PLATO (Programmed Logic
of Automated Teaching Operations) system, and they
looked into using it for self-paced, individualized
instruction. "I feel fortunate that I've been able to
do both lab research-hard-core research-and edu-
cational research projects simultaneously," Cutlip
said. "There are not a lot of environments where
that can be accomplished."
Cutlip developed a self-paced, mastery-oriented
course in reaction engineering for seniors. The stu-
dents could come into the computer lab at any time,
go through the tutorials at their own pace, and then
take assignments and quizzes as they were ready
for them. They could send and receive messages
from the teacher and work out intricate problems
using a touch-sensitive screen. Cutlip points out,
"We could keep track of each student and pose prob-
lems for them. We could tailor each problem for
each student."
Between the computer time involved and the hir-
ing of programmers, it turned out to be a very ex-
pensive developmental project, but it determined
Cutlip's focus as a teacher. "I definitely like the
concept of self-paced learning and mastery, giving
the student a choice of educational activities. It's
quite different from a one-way lecture where the
student is passive," he states.
Computer-based instruction also helps chemical
engineering students deal with the demands of their
curriculum, he believes. The chemical engineering
curriculum tends to be one of the most rigorous at a
university, he notes, involving extensive math, engi-
neering, and chemical engineering courses, with a
continually growing load of course work being
squeezed into the same amount of time. He feels
that requiring students to have a personal computer
161









to use for tutorials would enrich their understand-
ing of the course work.
Cutlip believes that the growing use of CD ROMs
for personal computers will accelerate the develop-
ment of computer-based learning methods. "We can
get 650 megabytes of information on that little disk.
It's a real educational challenge-what to put on it
to enhance the educational process." He adds that
CD ROMs could also be used for process design pack-
ages, for a physical properties data base, or for in-
structional modules that would serve as enrichment
activities; health and safety training is another po-
tential use. He points out that as students become
more computer literate earlier in their education,
computer-based course work will become a much
more common approach to teaching.
One way to involve computer-based learning in
chemical engineering education was developed by
Cutlip and co-researcher Mordechai Shacham of
Ben Gurion University in Israel. They developed a
numerical analysis package, POLYMATH, that al-
lows chemical engineers to use various numerical
methods to solve chemical reaction problems on a
variety of personal computers (Cutlip has both an
IBM and an Apple in his office and uses both ac-
tively). Since it was developed ten years ago,
POLYMATH has been refined and improved in sev-
eral versions and is now widely used by chemical
engineering departments. It is site licensed, so it is
easily shared throughout a department by both fac-
ulty and students.
"In my view, if we provide students with the right
general-purpose software, we can propose more re-
alistic problems for them to use in their homework,"
Cutlip says. The homework could be more represen-
tative of actual industrial problems because the com-
puter would allow the use of more complex numeri-
cal methods, and theory could then be taught in
more detail. He adds, "It's a big opportunity for
enhancing the educational process."
POLYMATH is distributed by CACHE, which de-
velops educational products and supports projects
that promote computer-aided chemical engineering
education. One of CACHE's biggest successes in
its twenty-three years, says Cutlip (who is in the
first year of a two-year presidency of the non-profit
organization), has been the introduction of
computer-aided process design (via FLOWTRAN)
with the help of the Monsanto Chemical Corpora-
tion. Recent CACHE activities include participation
in three curriculum development projects (with Na-
tional Science Foundation support) at the Univer-


Cutlip and one of his former graduate students, Angelos
Efstathiou, who is now doing teaching and research
at the University ofPatras in Greece.

sity of Michigan, Purdue University, and the Uni-
versity of Washington. Cutlip also is leading
CACHE's effort to produce the first CD ROM for an
educational discipline and has encouraged a new
effort in advanced computing.
Cutlip views the computer as a tool-much as the
slide rule and the electronic calculator which pre-
ceded it-that aids students in mastering the mate-
rial but which does not diminish the role of the
teacher. He is actively involved with his students;
they pop in and out of his office constantly. He greatly
admired his own teachers, such as Peters, who were
accessible to students even when they had heavy
administrative responsibilities, and he has vowed to
continue that tradition.
During the past semester Cutlip taught a gradu-
ate chemical reaction engineering course and the
undergraduate laboratory. He is also the AIChE
student chapter advisor for the second time in
his UConn career (he served for six years the last
time) and recently helped the students organize
and host the New England Regional AIChE student
chapter meeting.
He has been active in the local AIChE chapter
himself, serving as chairman and vice chairman of
the Western Massachusetts Local Section (which in-


Chemical Engineering Education









cludes eastern Connecticut) and winning its Dia-
mond Jubilee Award in 1983. He won the Ralph R.
Teetor Award for Outstanding Teaching Record from
the Society of Automotive Engineers in 1974 for his
work in helping engineering students at UConn de-
velop and engineer a prototype catalytic converter
device for an urban car (before these were required
on automobiles).
Cutlip has also been awarded fellowships to study
and teach in the United Kingdom and in Japan. In
1990 he went to Japan to work on fuel cells and also
gave seminars at six universities on interactive nu-
merical methods in chemical engineering education.
At his host institution, Yamanashi University, he
held afternoon teas for Japanese students, giving
them an opportunity to practice their conversational
English. He observes that students in Japan have
more day-to-day direction from their faculty advis-
ers but less interaction in lectures than do their
American counterparts.
He has taken two sabbaticals at Cambridge Uni-
versity in England (1974 and 1983) and regards
Cambridge as his "second academic home." He first
worked there with Dr. Nigel Kenney, and they have
since shared post-doctoral researchers, graduate stu-
dents, and research projects using computers in ca-
talysis-related work. One result of their work, using
gradientless reactors to study the oxidation of CO
and hydrocarbon mixtures, was that they were
among the first to determine that these reaction
systems would oscillate or change with time in a
repetitive way. They published and described math-
ematically and physically why this happened.
Another project that Cutlip developed at UConn
as a result of this collaboration involved periodic
reaction operation and the finding that the reaction
rate can be enhanced by feeding in one reactant and
then another, alternately instead of all at once.
While at Cambridge, Cutlip was a member of
Fitzwilliam College, participating in college activi-
ties and learning how the English university system
works. He found that the colleges allow faculties of
different departments and interest to interact and
get to know one another's fields.
The Cutlips also enjoyed living in a small village
near Cambridge, where their son attended school
and where Cutlip was elected to the PTA/school com-
mittee. Susan, who is an accomplished musician and
plays the violin in chamber groups and orchestras
in eastern Connecticut, was invited to play in both
the Cambridge University Orchestra and the Cam-
bridge Symphony while they were there.


Sabbaticals are one of the really great benefits of
a university position, offering an opportunity to see
how educational systems work and how societies
function in other countries, Cutlip observes, adding
"Chemical Engineering is a pretty small group of
professionals at the academic level, and sabbaticals
let you get to know people around the world." His


Cutlip and some of his Japanese students.


most recent sabbatical was at the University of
Adelaide in Australia in 1989, working with John
Agnew, who is the chemical engineering department
chairman there and who also worked in Cambridge
with Kenney.
Before leaving for Australia, Cutlip stepped down
from a nine-year tenure as department head at
UConn. Although he jokes about the demands of
that job, he confesses he "really enjoyed it-it's
one of the most, if not the most important job at
the university." It requires interfacing with so
many varied groups, he notes-parents, prospec-
tive students, faculty, students, graduates, deans,
administrators, business and industry, govern-
ment, and accreditation bodies. "You are at the
center of the network that really makes the univer-
sity go," he says.
The chairmanship also offers an opportunity to
have an impact on the curriculum and to recruit
and nurture new faculty. One of the interesting con-
trasts of the job, he says, is trying to get the faculty
to pull together as a cohesive group, yet encourag-
ing faculty members to work on their own projects
independently. "You have to be very persuasive in
order to get all this to work," he declares.
Ultimately, Cutlip's primary career interest in
chemical engineering has been teaching. As one who
began teaching at nineteen and who is now fifty-
one, he is likely to surpass his father's record in
front of the blackboard. Or, perhaps more aptly, in
front of the computer screen. 0


Summer 1993









MR on teaching...



SEVEN RULES FOR TEACHING

R. BYRON BIRD
University of Wisconsin
Madison, WI53606-1691

after forty years of being a university professor, I would like to think that I have learned something
about the art of teaching. From time to time younger colleagues and teaching assistants have asked
me for advice on how to teach. My suggestions to them can be summarized in terms of three
DON'Ts, three DO's, and one REMEMBER.

DON'T... SHOW OFF
Many teachers, either intentionally or unintentionally, seem to enjoy inflating their egos by
trying to impress the class with their own brilliance or with their own recently acquired
knowledge. Although a few particularly gifted students may be challenged by this display
of erudition, most students will be confused or disgusted-or both. A teacher can be colorful
without showing off, and colorful, lively teachers are much appreciated. It is important that
the teacher conduct the course at a level commensurate with the students' background.
Teachers also have to be very careful not to be condescending.

... BLUFF
Beginning teachers often feel embarrassed when they are asked questions that they can't
answer. Feeling that their lack of knowledge may make them appear inadequate, they then
try to escape by bluffing. Sooner or later their dissembling will be discovered, and the
students will lose respect for them. It is far better to admit ignorance and promise to find
out the answer to the question by the next class meeting. A challenging question from a
student can often be a wonderful learning experience for both the teacher and the class.
Keep in mind, too, that there may well be one or more members of the class who are
actually brighter and better informed than the teacher; such students will find the teacher's
bluffing contemptible.

... INTIMIDATE
All of us have at one time or another been victims of a tyrannical teacher who appears to
derive pleasure by making students feel uncomfortable or inadequate. Such teachers create
a hostile atmosphere and thereby make the learning process difficult. Students will be
hesitant to ask questions if they are told that their questions are ridiculous. Students
should be encouraged to ask questions, and they should be answered patiently and care-
fully. The class should be reminded of the Japanese proverb: Kiku wa ichiji no haji, kikanu
wa matsudai no haji: To ask is a moment of shame; not to ask is an eternity of shame.
Teachers should be demanding, but they should not embarrass or ridicule students either
in public or in private. They should always treat students with courtesy and respect.


DO... KNOW WHAT YOU ARE GOING TO TEACH
It may seem unnecessary to remind teachers that they should know the subject material
thoroughly. It is not enough just to have read the textbook before going to class. We all
164 Chemical Engineering Education









know that even the best textbooks contain misprints, factual errors, and unclear passages.
The good teacher will have read other textbooks, some primary sources, some review ar-
ticles, or perhaps some recent research papers in order to have a depth of understanding
well beyond that needed for the classroom presentation. He will also spend some time
thinking independently about the subject material in order to develop a deeper understand-
ing and even novel viewpoints. This is time consuming, but ultimately very rewarding.

... KNOW WHY YOU ARE GOING TO TEACH IT
Students have to be motivated in order to learn new material. If they know why they should
learn a particular subject and how they can apply the newly learned material, they will be
more enthusiastic and receptive. It is therefore essential that the teacher be well aware of
the scientific and engineering relevance of each topic; if a topic is not important, then it does
not merit inclusion in the syllabus. It is also very important to discuss how the topic being
presented is related to subject material in other courses in the curriculum. Many teachers
are discouraged when students seem to be unaware of the connections between different
courses, and yet they do very little themselves to emphasize these connections. Since stu-
dents do have problems with carry-over between courses, it is quite appropriate to take a few
minutes to review a topic from another course by prefacing the comments with something
like "As you will recall from your course in thermodynamics .."

... KNOW HOW YOU ARE GOING TO TEACH IT
It is not enough to master a topic before teaching it. Considerable thought must also be given
to the mode of presentation-questions and answers, discussion of homework problems,
visual aids, and library assignments are just a few of the many alternatives to straightfor-
ward lecturing. The sequencing of the material also requires careful consideration; for
example, should one start with a general statement and then give illustrative examples, or is
it preferable to give some examples first and then proceed to a general statement? Symbols
and notation should be carefully chosen for the optimum mnemonic value. Making a subject
easy to learn requires originality and artistry. One of a teacher's most important jobs is to
figure out how to take a massive amount of difficult material and present it in an orderly,
easy-to-understand way. An excellent motto for a teacher is: "Eschew obfuscation!"


REMEMBER .. .THE TEACHER'S JOB IS TO SERVE THE STUDENT
Students pay money for being taught, and teachers receive money for teaching. The teacher
has a contractual obligation to provide the best possible guidance to those who are entrusted
to him. This includes high quality lecturing, careful mentoring, career guidance, and in some
instances a willingness to help with personal problems. It also includes the maintenance of
standards and informing students frankly and honestly when their performance is unsatis-
factory; the teacher does not help students by being a crowd-pleaser or by rewarding poor
performance. The teacher's responsibility does not stop with the end of the semester, or even
with the student's graduation. Years later the teacher may be called upon to provide help in
connection with a former student's job application, his aspiration to a position with more
responsibility, or his consideration for an award or prize. The student-teacher relation can
evolve through the years into a lasting friendship, with all the rewards that such a relation-
ship implies.


I have arrived at the above simple rules after many years of classroom teaching, student advising, and
textbook preparation. At various times I have broken all of the above rules, and I have suffered the conse-
quences. The rules are hard to follow, but it helps to have some guidelines.
(I would like to thank Professor C.G. Hill, Mr. Atul M. Athalye, and Mr. Peyman Pakdel for constructive comments.)
Summer 1993 165










design


THE TECHNICALLY FEASIBLE


DESIGN
TW FRASER RUSSELL AND N. ORBEY
University ofDelaware
Newark, DE 19716

Introductory Note
It is frustrating to attempt to capture effective classroom experiences in an article, but I am tempted to
try again to do so because this past year I had the very rewarding experience of supervising a DuPont
Teaching Fellow* (Ms. Linda Broadbelt) while team teaching a junior level chemical engineering
course in reaction and reactor design with Dr. N. Orbey, a visiting professor at the University of
Delaware from Middle East Technical University (Turkey). Their enthusiasm for the "technically
feasible design" approach has prompted this paper. It is my hope that it will encourage classroom
experimentation and help educate students about design problems.
TW Fraser Russell -


Chemical engineers design, build, operate, and
modify process equipment, or carry out the
research necessary to do so more creatively
and more efficiently. Not all chemical engineers are
directly involved in the art and science of design,
but all chemical engineers are exposed more or less
effectively to various aspects of design in the educa-
tional programs in our universities. Indeed, it is
part of our profession's criteria for accreditation, as
shown by Section IV.C3(a) of "Criteria for Accredit-
ing Programs in Engineering in the United States":
(IV.C.3(a)) Engineering Design
(a) Engineering design is the process of devising a system, compo-
nent, or process to meet desired needs. It is a decision-making
process (often iterative), in which the basic sciences, mathematics,
and engineering sciences are applied to convert resources opti-
mally to meet a stated objective.
and by the "Program Criteria for Chemical and Simi-
larly Named Engineering Programs":
Engineering Design. (Amplified criteria section IV.C.2.d(3))
The various elements of the curriculum must be brought together in
one or more capstone engineering design courses built around com-
prehensive, open-ended problems having a variety of acceptable
solutions and requiring some economic analysis.
These legal sounding criteria, which attempt to
define design content, do not give us any insight
into the value of design as a tool for making courses
more intellectually challenging or more interesting.

* The E.I. duPont de Nemours and Company's DuPont Teaching
Fellows Program is designed to encourage graduate students to
become interested in university-level teaching.


In fact, the "Chemical Engineering Criteria" which
calls for a "capstone" design course has been inter-
preted by some educators as allowing them to ig-
nore design until the final year of a four-year pro-
gram in chemical engineering.
We tend to educate in the early years of the cur-
riculum by using ideal technical problems in our
courses. The ideal technical problem is one in which
all the information is given and for which a single
correct answer is most frequently obtained by solv-
ing an equation or sets of equations. Much effort is
expended, both by professors in class and by stu-
dents doing homework, on mathematical manipula-
tion. While this serves a purpose in that it helps
teach problem-solving methodology, it tends to pro-
TW Fraser Russell is the Allan P. Colbum
Professor of Chemical Engineering and Director
of the Institute of Energy Conversion, a labora-
tory at the University of Delaware devoted to thin-
film photovoltaic research. He received his BSc
and MSc degrees from the University of Alberta
and his PhD from the University of Delaware, all
in chemical engineering. He is the author of over
seventy technical publications and is coauthor of
two texts.


Nese Orbey received her BS and MS in chemi-
cal engineering from Middle East Technical
University (METU), Turkey, where she is cur-
rently an associate professor, and her PhD
from McGill University, Canada. She served
as Visiting Associate Professor at the Univer-
sity of Delaware from 1990 to 1992. Her re-
search interests are in the area of polymer
rheology.


Copyright ChE Division ofASEE 1993
Chemical Engineering Education










duce students who do not understand that engineer-
ing problem solving and/or the creation of engineer-
ing opportunity must go beyond routine mathemati-
cal manipulation.
This serious difficulty can be avoided and stu-
dents can be introduced to the art of engineering
earlier in the curriculum if faculty would re-
quire that the students produce a "technically
feasible design" rather than a single-answer solu-
tion to a problem.

TECHNICALLY FEASIBLE DESIGN
A technically feasible design is one which defines
the size of a piece of process equipment to meet a
stated goal, and in so doing initiates an analysis of
the factors affecting optimal design. It could be speci-
fication of the volume for a reactor, the total area
for an exchanger, or the height and diameter of a
separation unit. The use of a technically feasible
design can be illustrated for chemical engineering
students by considering a simple problem in chemi-
cal reaction engineering.
Chemical engineers frequently become involved
in a reactor-process design problem at an early
stage, i.e.,
How can our firm safely make a product ("D") for which there appears
to be a good market at a fair profit?
While this is the type of problem that we would like
a chemical engineer to be able to solve, it is too
open-ended for students in their first few years of
study. It is time-consuming and difficult even for
many faculty to do. It also disrupts the logical flow
of subject matter to introduce the issues of market
development, competitive market-share pricing, and
capital and operating cost estimating which are nec-
essary to solve the problem.
The following is a simple, technically feasible de-
sign problem that can be presented to the student:
Our firm has determined that we can sell 1260 metric tons/year of
product "D," a raw material for the manufacture of an important
fiber. "D" has a molecular weight of 50 and the reactor is assumed to
operate 24 hours a day, 350 days a year.
Our laboratory has studied the homogeneous liquid phase reaction
which produces "D"
A+B- D
This reaction can be carried out isothermally in an excess of B with
the kinetics determined as follows:
rA = kcA
k = 0.005 min-
The simplest possible technically feasible design
can be completed if the student is told at this point
to assume that the reactor will be a continuous flow
stirred tank (CFSTR) with a feed stream concentra-
Summer 1993


We tend to educate .. .by using
ideal technical problems in our courses.. .in
which all the information is given and for which a
single correct answer is most frequently obtained
by solving an equation or sets of equations.

tion of A, CA = 0.2 g-moles/liter.
We have tested this problem with chemical engi-
neering students in courses such as "Introduction to
Chemical Engineering Analysis" and "Chemical En-
gineering Kinetics" and with a great many non-
chemical engineers (mostly chemists and other en-
gineers) in professional society-sponsored courses
throughout the Delaware Valley. So far, over two
thousand students have been asked to carry out this
technically feasible design in class.
At the stage in any course when we introduce this
exercise, the students are capable of deriving the
required mass balances:


species A
species D


0 = qCAF qCA kCAV
O=O-qCD+kCAV


The technically feasible design is required for
C, = 0.2 g-moles/liter
k = 0.005 min1
total production of 1260 metric tons/year
(qCD = 50 g-moles/min, or 2.52 x 107 g-moles/year)
We ask students to carry out this exercise during
class so we can observe their thought processes and
can thus generate more effective discussion. When
the exercise is introduced, the students are told that
the design will be considered complete when the
reactor volume, V, has been determined.
In order to maximize the educational gain for both
the instructor and the students, the class should
work unaided on the design for about thirty min-
utes, with each student attempting to obtain the
reactor volume, V. We have found that students
rarely obtain the reactor volume on their own with-
out additional class discussion. A walk around the
classroom, observing how the students attempt to
carry out this very simple design, is most instruc-
tive. They will manipulate and remanipulate Eqs.
(1) and (2) in an effort to obtain V. It has never been
clear to us why almost all students do this, since
counting unknowns and equations clearly shows that
one variable in addition to those given must be speci-
fied. (Students should have done enough algebraic
manipulations by this time in their academic lives
to be thoroughly familiar with solutions of such a
simple system of equations.)
167










Students are often reluctant to complete the tech-
nically feasible design by selecting values for the
variables, q or CA, probably because they have been
taught to solve problems in which they had to de-
rive and manipulate equations to obtain a solution.
A very simple design decision (i.e., select a value for
CA, the exit concentration of raw material A, and
determine reactor size V) turns out to be foreign to
the student's whole experience in problem solving.
To achieve a technically feasible design by assum-
ing a value for q or for CA, it is convenient to rear-
range Eqs. (1) and (2). The most effective way to
compute a reactor volume, V, is with Eq. (2):

V= (3)
kCA
V= 50
0.005 CA

Since CA can only vary between CA = 0.2 g-moles/
liter and 0, the student can quickly obtain a
technically feasible design. For example, if
CA = 0.1 g-moles/liter, then V = 100,000 liters.
If the students are encouraged to experiment with
the set of equations, about a third of them will even-
tually derive Eq. (3). Others will assume a value for
q, calculate CD from qCD = 50 g-moles/min, obtain CA
from CA CA = CD (the addition of Eqs. 1 and 2), and
then solve for V using either Eq. (1) or Eq. (2). This
more involved approach has the disadvantage that
limits on the value of q are not as obvious as limits
on the value of CA. For instance, if q is assumed to
be 200 liters/min
CD = 50/200 and CA = 0.2 0.25 = 0.05
Obviously, CA cannot be negative, so q must be
greater than 250 liters/minute for a technically fea-
sible design (q qC D/CA).
The problem is discussed in more detail in Intro-
duction to Chemical Engineering Analysis, I1 and the
role of the technically feasible design in initiating
an analysis of the factors affecting optimal design is
illustrated in Table 1.
It is very important to again stress that almost all
the educational impact of the technically feasible
design concept is lost if students do not have an
opportunity to work on the problem by themselves
in a classroom setting, with an instructor who is
willing and capable of initiating discussion. Table 1
shows that the optimal size of a reactor cannot be
considered without also considering how unreacted
A is separated from product D. It also shows stu-
dents how the reactor analysis affects the down-
stream process design. A large reactor with a small
168


TABLE 1

q C, CA V 0=V/q
(liters/min) (g-moles/liter) (g-moles/liter) (liters) (mins)
250 0.200 0 0
300 0.167 0.033 303,000 1000
400 0.125 0.075 133,000 333
500 0.100 0.100 100,000 200
800 0.0625 0.1375 72,600 90.7
1000 0.0500 0.1500 66,600 65
2000 0.0250 0.1750 58,100 27
4000 0.0125 0.1875 53,200 13.3

TABLE 2
CQ(g-moles/liter) C,(g-moles/liter) t(min) batches/year V(liters)
0.01 0.19 599 700 190000
0.05 0.15 277 1269 133000
0.10 0.10 139 1945 130000
0.15 0.05 57.5 2839 178000


concentration of A in the effluent (high conversion)
costs more than a small reactor with a large concen-
tration of A in the effluent (low conversion). If a
customer can use D with a small amount of A
present, then it might be possible to eliminate an A-
D separation unit which requires an expensive reac-
tor. To reduce reactor costs one must pay for the
capital and operating costs of the separation unit.
An "optimal" design is discussed in Introduction to
Chemical Engineering Analysis. Also, a process de-
sign game that has been widely used and which
very effectively illustrates the economics and intro-
duces the concept of competition is described in a
paper titled "Teaching the Basic Element of Process
Design with a Business Game."[21
The CFSTR technically feasible design problem
can be used with students at any level in the cur-
riculum (when providing Eqs. 1 and 2, we have even
used it with high school seniors and first-semester
freshmen). We expect University of Delaware chemi-
cal engineering majors to be able to derive Eqs. (1)
and (2) after their sophomore year.
The design problem is used throughout the junior-
level chemical engineering kinetics course, and we
require that the students carry out a commercial-
scale technically feasible design for the following
reactor design situations:
CFSTR isothermall single reaction)
batch reactor isothermall single reaction)
semi-batch reactor isothermall single reaction)
tubular reactor isothermall single reaction)
CFSTR and tubular reactor isothermall series-par-
allel reactions)
CFSTR (non-isothermal)
Chemical Engineering Education









EXAMPLE: BATCH REACTOR
The technically feasible design for the batch reac-
tor requires significantly different thinking, even
though the same problem is addressed. The stu-
dents are expected to derive the pertinent material
balances

dC -kCA (4)

dC kCA (5)

and solve the differential equations

In(A -kt (6)

CD=CAo-CA (7)
At this stage the problem differs from the CFSTR
example in that the volume V for a technically fea-
sible design cannot be directly obtained since the
material balance equations for the batch reactor do
not contain a volume term. Reaction time, t, must
be obtained from Eq. (6) and then used to obtain the
reactor volume.
Again, students need to work on their own in a
classroom setting and must be given an opportunity
to discuss the design with the instructor. Most
students have difficulty obtaining the reaction time
despite having encountered a similar situation
with the CFSTR design. Equation (6) has two un-
knowns: CA and t. The value of t can only be solved
as a function of CA, and any pair of t-CA is one
solution leading to a technically feasible design.
Students must assume a value for CA just as
they did in the CFSTR example. For example, if
CA = 0.1 g-moles/liter
t = 138.6 min and CD = 0.1 g-moles/liter
Students must also make additional judgments to
obtain a technically feasible design. The total yearly
production is known and the volume of the reactor
is related to the reaction time.
2.52 x 107 = (VCD) (Batches/year) (8)
Both CD and the number of batches per year that
can be processed depend on the reaction time (Eqs.
6 and 7).
CD = CA(1 ekt) (9)
To find the total time in hours to process a batch,
time for charging raw materials, removing product,
and cleanup must be considered in addition to the
reaction time. We can then obtain V from Eq. (8) by
assuming there are 350 x 24 hours in a year. The
results of some sample calculations for technically
feasible values of V are given in Table 2, assuming
Summer 1993


that the time for charging raw materials, removing
product, and cleanup is two hours.
Table 2 also provides information for a discussion
of the important factors in any optimal design. The
optimal size of the reactor depends on downstream
processing. In the case of a CFSTR, volume decreases
monotonically as conversion decreases (see Table 1).
Batch processing is a labor-intensive process. At very
low conversions (CA = 0.15) with low reaction time,
the time required for charge and cleanup (two hours)
is almost twice that of the reaction time. The reac-
tor volume is thus greater than for CA = 0.10.
In the CFSTR, low conversions had the advantage
of low reactor capital cost and the disadvantage of
high separation costs. In the batch process, low con-
version leads to the double disadvantage of high
reactor capital cost and high separation costs.
In Table 2, batches/year are given, but in class-
room discussions either batches/day or batches/shift
can be computed to promote discussion on the is-
sues of labor requirements and costs.

CONCLUSIONS
The importance of design in chemical engineering
education has been effectively taught to both chemi-
cal engineers and chemists by requiring students to
complete simple, technically feasible designs in class.

ACKNOWLEDGMENTS
The authors would like to thank the Department
of Chemical Engineering for supporting Dr. N. Orbey
and for encouraging experiment in chemical engi-
neering education. Ms. L. Broadbelt was a valued
teaching colleague, and the E.I. duPont de Nemours
and Company is acknowledged and thanked for the
creation and sponsorship of the E.I. duPont Teach-
ing Fellows Program that provided her support.

NOMENCLATURE
C species concentration, g-moles/liter
k specific reaction rate constant, min'
q volumetric flow rate, liters/min
rA rate of reaction, g-moles/liter, min
t time, minutes
V reactor volume, liters
Subscripts
A,D chemical species
F feed condition
o initial condition
REFERENCES
1. Russell, T.W.F., and M.M. Denn, Introduction to Chemical
Engineering Analysis, John Wiley & Sons (1972)
2. Russell, T.W.F., and D.S. Frankel, "Teaching the Basic Ele-
ment of Process Design with a Business Game," Chem.
Eng. Ed., 12, 18 (1978) O
169










ISI classroom


PROCESS CONTROL EDUCATION

A Quality Control Perspective


PRADEEP B. DESHPANDE
University ofLouisville
Louisville, KY40292


here are a number of issues that warrant an
examination of the current undergraduate
course in process control. One is the notion of
statistical quality control being used in industry.
Statistical process/quality control (SPC/SQC)
concepts"i grew out of the discrete manufacturing
environment as a response to competitive pres-
sures, and current efforts aim at the notion of zero
defect. In recent years, SPC concepts have also made
their presence felt in the continuous-process indus-
tries,[2"" and concepts such as control charts, control
limits, common causes, and special causes are now
commonly used as measures of product quality vari-
ability as well as to detect problems and take correc-
tive actions. Automatic correction when a defect is
detected is beyond the scope of SPC, however, so the
word "control" in the acronym is somewhat mislead-
ing. For the buyers of products from processing in-
dustries, SPC measures represent proper evidence
of the variability of product quality, and these mea-
sures often form the basis of purchasing contracts.
One consequence of the success of SPC is that excel-
lent communication appears to exist between statis-
ticians and company management.
Similar communication among process control pro-
fessionals and management, however, appears to be
lacking, and one of the contributing factors is con-
trol "jargon." The control engineer speaks in terms
of servo and regulatory responses, input suppres-



Pradeep B. Deshpande is Professor and
former Chairman of the chemical engineering
department at the University of Louisville,
where he also directs a Center for Desalina-
tion. He has over twenty-two years of aca-
demic and full-time industrial experience and
is author, coauthor, or editor of four textbooks
in process control and seventy papers.
Copyright ChE Division ofASEE 1993


Upper Control Limit


as





S ...........................................
24

Lover Control Limit

Time
Figure 1. Typical control chart

sion and penalty parameters, model uncertainties,
exponential filters, and robustness. While there is
no implication that these are unimportant, to tie
them to product quality (the primary concern of man-
agement) is often difficult. A proper understanding
of the role of both statistical process control and
engineering process control in achieving product
quality would improve communication between stat-
isticians, management, and control specialists, and
would considerably enhance the ability of control
specialists to have a stronger impact on process and
plant operations. Students should be aware of the
importance of this kind of interaction.
Consider the Shewhart control chart"6' of a hypo-
thetical discrete parts manufacturing process, shown
in Figure 1. In discrete-parts manufacturing, the
adjacent data points are assumed to be statistically
independent of each other. In fact, the data from a
process under statistical control are deemed to fol-
low a Gaussian (normal) distribution.
The mean of the data is the center line on the
Shewhart control chart. In light of the assumption
of normality, then, 99.73% of the data will lie within
3o limits from the mean; the 3a limits are the so-
called statistical control limits. If the data points lie
within the control limits and are more or less ran-
Chemical Engineering Education










The control engineer speaks in terms of servo
and regulatory responses, input suppression and
penalty parameters, model uncertainties, exponential
filters, and robustness. While there is no implication
that these are unimportant, to tie them to product
quality is often difficult.

domly distributed, the variability is deemed to have
been caused by common causes. Within the context
of process control, common causes are those random
disturbances whose detrimental effect upon product
quality cannot be eliminated by any kind of control
action. If a non-random pattern is detected, although
the data points are within the control limits (e.g.,
seven points in a row are above/below the central
line, fourteen points in a row alternate up and down,
etc.), there may be an assignable (or special) cause
that should be investigated. Points outside the con-
trol limits are also said to have been caused by
assignable causes. Assignable causes need to be
investigated and corrective action must be taken.
In contrast, the data points on a chart similar to
Figure 1, representing the quality variable from a
continuous process and plotted as a function of the
sampling interval, are invariably autocorrelated.
Furthermore, the center line is the set point and not
the mean of data points. The integral action in the
controller will insure that the quality variable will
return to the set point for certain types of distur-
bances. Thus, the closed-loop responses in the CPI
do not obey statistical control concepts well. Recent
research, however, has led to methods that can be
used to analyze the autocorrelated data and make
them amenable to statistical monitoring.
Process control practitioners have often observed
that a good control algorithm is one which shifts
much of the variability of an output onto the input,
i.e. the manipulated variable.[6" In fact, an algorithm's
performance is frequently measured in terms of its
ability to shift the entire variability from the output
to the input under ideal conditions.
An illustrative example of a heat-exchanger sys-
tem taken from Downs and Doss[61 is shown in Fig-
ure 2. In this instance the control algorithm at-
tempts to hold the exit temperature as closely as
possible to the set point by suitably manipulating
the flow of the heating medium. The ability to de-
liver offset-free performance is a key requirement in
controller design. It has been pointed out that in
industrial situations, manipulated variables often
have their own processing units, and transferring
an excessive amount of variability to them may not
always be the the best approach. Thus, a control law
Summer 1993


Outlet
Temperature
Cooled Process Variotion
Stream

Figure 2. A control algorithm shifts the variability of
the output onto the input.

should be designed so that only as much variability
is transferred to the input as is necessary to pro-
duce a product of acceptable quality.
The foregoing discussion points out the need for
suitable control laws and appropriate statistical tools
which can handle autocorrelated output data for
statistical monitoring purposes. We recently pre-
sented a unifying perspective that combines the best
features of engineering process control (EPC) and
statistical process control (SPC) for achieving total
quality control of continuous process systems. In
the following paragraphs we will briefly review the
unifying methodology for EPC and SPC and will
reveal what new material should be added to the
traditional process control course to make it more
effective in meeting the needs of industry.

UNIFYING METHODOLOGY FOR ENGINEERING
AND STATISTICAL PROCESS CONTROLm
We propose a two-part procedure for achieving
total quality control of continuous processes. In Part
1, we design a suitable control law to hold the out-
put (reflecting the quality variable) within specifica-
tions in the presence of load disturbances and mod-
eling errors. In Part 2, we analyze and massage the
autocorrelated output data so as to render them
amenable to statistical monitoring. The usual SPC
rules may then be applied to keep the continuous
process under statistical control.

Part 1
Engineering Control Systems Design
Many approaches to control are described in the
literature. Just which approach to use depends on
171


Inlet
Temperature
Variation

Reactor Feed from
Upstream Process


Variations Ci


Hot Process
Stream









the type of process and on the personal preference of
the designer. The basic requirement is that the con-
trol law must hold the quality variable within speci-
fications in the presence of disturbances and model-
ing errors. In light of our introductory discussion, it
is obviously desirable that the control law contains
tuning constants which can be adjusted to improve
quality or to reduce costs, in terms of the manipu-
lated variable movements, consistent with client
specifications. We will here review two approaches
to control: PID control with feedforward control and
dead-time compensation, and stochastic control.
The Standard Approach The standard PID con-
troller continues to be the most popular in industry.
the ideal PID controller is described by the transfer
function

G,(s)= Kc(l+-l +DS (1)

There are several approaches to tuning this type
of controller. Some of them involve open-loop test-
ing, while others are based on closed-loop experi-
mentation. The settings that result are meant to
satisfy certain specified optimization criteria, such
as minimum ISE (integral of the squared error),
quarter decay amplitude ratio, etc.
The system performance deteriorates as dead-time
in the loop increases. The notion of dead-time com-
pensation is to remove the dead-time from the
system's characteristic equation so that the system
performance improves. There are two ways to achieve
dead-time compensation: one (attributable to O.J.M.
Smith"'1) is called the Smith predictor, while the
other (attributable to C.F. Moore"91) is called the ana-
lytical predictor.
In real-life applications, disturbances are invari-
ably present. The controller must compensate for
the negative effect of the disturbances on the pro-
cess output. In those applications where disturbances
can be measured, the notion of feedforward control
may be employed. Figure 3 depicts the block dia-
gram of the closed-loop system, showing PID control
with dead-time compensation and feedforward con-
trol. The Smith predictor approach for dead-time
compensation is shown in Figure 3 for illustrative
purposes; Moore's analytical predictor may be em-
ployed instead if so desired. The arrangement shown
in Figure 3 is industry standard. Blocks to imple-
ment PID control, lead lag, and dead-time compen-
sation come standard with modern distributed con-
trol systems.
Stochastic Controller DesignO11'" We know that
for identical model structures (Gp and GL) and iden-
172


tical closed-loop performance specifications (e.g.,
minimum variance), there is really no difference be-
tween the design of feedback controllers for deter-
ministic or for stochastic disturbances.[121 It is nev-
ertheless important to be familiar with how control-
lers are designed for stochastic disturbances, be-
cause with this approach the closed-loop data can
lend itself directly to statistical monitoring.
Consider a single-loop linear system that is per-
turbed by stochastic disturbances. A stochastic dis-
turbance, called noise, is obtained by passing white
noise, at (having zero mean and a constant variance
02a), through a suitable model structure such as a
first-order lag, an integrating type load such as a
ramp, etc. The disturbance model structure is se-
lected so that it is representative of the real-life
situation. In fact, plant testing with PRBS (pseudo
random binary sequence) signals followed by time
series analysis can help identify the models that
would be needed for designing the type of controller
being discussed in this section. The purpose of the
exercise is to design a control law which will mini-
mize the variance. The output of the system, Ct, is
related to the manipulated variable, Mt, and the
noise, Nt, according to

Ct = MtF- + Nt (2)

where F is the time delay in terms of number of
sampling periods.
Equation (2) can be equivalently written as

W(z-1)
Ct+F+1 = -- Mt + Nt+F+1 (3)
t+1 (z-1)

For minimum variance control, C+F+l, must be set to


Figure 3. The standard approach.
Chemical Engineering Education









zero. Equation (3) then gives

8(z-1)
Mt = (z-- t+F+1 (4)

The control law given in Eq. (4) cannot be imple-
mented since it requires knowledge of N,, F+1 sam-
pling instants into the future. Since future informa-
tion can only be forecasted, Eq. (4) must therefore
be written as

8(z-1) ^
Mt t+F+1 (5)
t )(z-1)

where the caret A denotes an estimated value. For
illustrative purposes, we will assume that the noise
model is adequately described by

e(z-1)
Nt = (-)vd at (6)

Usually, the parameter d = 0, 1, or 2. It permits the
designer to describe non-stationary types of distur-
bances. Following the procedure for forecasting the
disturbance (see Reference 7 for details) leads to the
control law

S8(z-1) L(z-1) 1
V(z) = (((z) eltzil)t (7)

where

et =Ct Cset
et
Equation (7) is the minimum variance stochastic
control law for single-loop systems. The choice of
minimum variance (deadbeat control) invariably
leads to excessive manipulated variable movements,
but this difficulty can be overcome by incorporating
a filter ahead of the controller. Here, the filter con-
stant can be adjusted to improve quality or to re-
duce costs. The procedure has been shown equiva-
lently leading to the IMC scheme[131 with a filter.[121
It can be shown[12] that substitution of Eq. (7) back
into Eq. (2) for a case where F = 0 gives
Ct = at (8)
That is, the closed-loop output data are distributed
according to a normal distribution, having zero mean
and a constant variance. Thus, the output data can
be used directly in preparing control charts
(Shewhart, CUSUM, etc.). It should be pointed out,
however, that for processes with dead-time under
minimum variance control, the data points every F
sampling intervals need to be used for control chart-
Summer 1993


ing since the autocorrelation reduces to zero at lag F
in this instance. As previously pointed out, mini-
mum variance cannot often be specified because it
leads to excessive movements of the manipulated
variable. Furthermore, the quality requirements in
specific situations may not call for such tight con-
trol-in which case it would be wiser to select tun-
ing constants that will dampen the oscillations. Un-
der these situations, industrial experience suggests
that the output data will be autocorrelated.[14"151 The
question remains: how can the autocorrelated data
be massaged so that SPC rules can be applied? We
take up this topic in the following section.

Part2
Statistical Monitoring
We assume here that the feedback controller has
been designed and the tuning constants have been
selected properly. Thus, we can surmise that the
process operates under the command of the selected
controller, producing a product of acceptable quality
in the presence of load disturbances and modeling
errors. We want to apply SPC techniques to main-
tain the continuous process under statistical con-
trol. We can assume that the variance is greater
than the minimum and that the output data are
autocorrelated. Attempts to apply the traditional
SPC rules will result in false signals due to the
highly autocorrelated nature of the data; that is, no
assignable causes would be found.
Problems arising due to autocorrelation can be
overcome in one of two ways. In the first approach
an autocorrelogram is prepared, showing how the
autocorrelation coefficient reduces with increasing
sampling intervals."16] From such a plot, a sampling
interval may be selected for which the autocorrelation
coefficient is sufficiently small. A control chart can
then be prepared using the selected sampling inter-
val to which SPC rules may be applied as usual to
detect the presence of assignable causes and to main-
tain the process under statistical control. A poten-
tial drawback of this approach is that the selected
sampling interval may be too large, meaning that
the process could go out of control before the next
data point becomes available.
In the second approach, the thrust is to fit an
appropriate time series model to the observations
and then apply the control charts to the stream of
residuals from the model.'5m Thus, if C, represents
the observation, and Ct represents the predicted
value obtained from an appropriate model fitted to
past data, then the residuals et = Ct -Ct, represent-
ing the prediction error, will behave as independent
and identically distributed random variables. Several
173










time-series models have been suggested for this pur-
pose. One is an autoregressive integrated moving
average (ARIMA) model that is of the form

(p(z l)Vd Ct = Oq(z-)at (9)
Another basis is the exponentially weighted mov-
ing average (EWMA) statistic. In this instance, the
sequence of one-step ahead forecast errors

Et = Ct -Ct(t- 1) (10)

are deemed to be independently and identically dis-
tributed and may be used to prepare control charts
to which SQC can be applied as usual. Here, Ct (t 1)
is the forecasted value of C, made at time instant
t-1. The EWMA approach is said to have computa-
tional advantages over the exact ARIMA approach,
but the former is adequate when the observations
are positively autocorrelated and the process mean
does not drift too quickly.
Having reviewed the unifying procedure for total
quality control in continuous process industries, we
will now discuss the issue of fault diagnosis and the
corrective measures that can be invoked to remedy
the situation. We assume that the designer has
access to the run-time charts and the appropriate
control chart pertaining to the quality variable
under assessment.
A variety of assignable causes can lead to out-
of-control points on the control chart. For some,
the remedy is in the domain of instrumentation
and control, while for others the remedy may lie
elsewhere. Some commonly encountered assign-
able causes in the domain of instrumentation and
control are
* Malfunctioning control valve and/or sensor
Changes in dynamic process parameters such as gain,
time constants, and dead-time due to equipment foul-
ing, catalyst decay, etc.
Increasing system nonlinearities.

PROPOSED ADDITIONS TO COURSE CONTENTS
A number of excellent textbooks for undergradu-
ate process control are available (a sampling is
included in references 17 through 20 at the end of
this article), and instructors typically cover a num-
ber of standard topics in the course (i.e., the
material in Chapters 1-16 of Process Dynamics and
Control"8l). In light of the foregoing discussion, we
feel the following material can also be added to
the course contents.
> Introduction to Process Control It must be emphasized


in the introductory chapters that the fundamental objec-
tive in process control is to produce products of a specified
quality. Other aspects (such as maximizing throughput,
environmental considerations, and safety) are extremely
important, but the student should not lose sight of the
fundamental objectives.
- Statistical Process Control A new chapter on statistical
process monitoring should be introduced. Students need
to understand the assumptions inherent in SPC-namely,
normality of quality data. SPC measures (such as
Shewhart and CUSUM charts) and concepts (such as com-
mon causes and assignable causes) need to be discussed,
and the commonly used rules to detect out-of-control sig-
nals should be outlined.
- Feedback Controller Design During the discussion of
the trade-offs between responsiveness and robustness in
the controller design section, the instructor should intro-
duce the new perspectives on trade-off between quality
and costs, and the discussion should include a number of
control laws that have the desirable properties. Since the
design of control algorithms will require an appreciation
of feedforward control and dead-time compensation, these
concepts will also have to be introduced if they have not
yet been covered.
- Process Identification The specified closed-loop perfor-
mance can best be achieved when the process model accu-
rately reflects the industrial plant. Pseudo random binary
sequence (PRBS) testing is widely used by industry to
identify plant dynamics. Time series analysis of the
input-output data leads to transfer function models;
step response models can also be evaluated. Because of
predictable time limitations for instruction, we suggest
that canned software packages be used to demonstrate
the concepts.
- Introduction to Stochastic Control As previously men-
tioned, deterministic and stochastic design procedures will
lead to the same control law for identical performance
specifications and model structures. It is nevertheless de-
sirable to expose the student to the basics of stochastic
control. The important lesson here is that under ideal
conditions, the closed-loop output obtained under mini-
mum variance control has a normal distribution, and there-
fore it is directly usable in preparing control charts for
statistical monitoring purposes. Here too, the instructor
can highlight the trade-offs between quality and costs.
)- Unifying Methodology for EPC/SPC The instructor
should warn of the problems associated with the use of
autocorrelated data in the CPI in preparing control
charts-namely, that numerous false signals are likely to
result. The procedure for massaging the autocorrelated
data to make them amenable to statistical monitoring
should be discussed.
>- Fault Diagnosis The last item concerns what to do
when the presence of assignable causes is detected. Ex-
pert systems are being used in some applications to de-
duce what actions to take when an assignable cause is


Chemical Engineering Education










detected. Again, due to time limitations, only an introduc-
tion to expert systems can be given here.

A study of the foregoing topics, together with the
standard material currently covered, would lead to
a more effective process control course.
While we have essentially focused on single-loop
systems in this paper, the ideas can be extended to
multivariable systems as well. A suitable (multiva-
riable) controller would be needed, however, in
order to maintain each quality variable of the
multivariable system within specified limits. The
discussion on statistical monitoring would remain
unchanged.

CONCLUSIONS
We have offered some comments on the under-
graduate process control course and have shown how
the unifying methodology for engineering and sta-
tistical process control brings attention to the topics
that should be studied to gain a fundamental under-
standing of engineering process control from a qual-
ity control perspective. We hope that the material
presented here will be helpful to other process con-
trol instructors.

NOMENCLATURE
a normally distributed random variable
C controlled variable
E error (set point measured value)
E, forecast error, Ct Ct (t 1)
F delay expressed as number of integer sampling
periods
G transfer function
K gain
k kth sampling instant
L load
M manipulated variable
N noise
R set point
s Laplace transform operator
z z-transform operator
Subscripts
c pertaining to controller
D pertaining to derivative mode in Eq. (1)
I pertaining to integral mode
L pertaining to load
P pertaining to process
t pertaining to time
A estimated value
Greek
6(z-1) polynomial in z1
O(z-1) polynomial in z 1
8(z') polynomial in z-'
o(z-') polynomial in z'-


~p autoregressive polynomial of order P,
(1+ ,lz- + z2 +.. +...+pZ-p)
8 moving average polynomial of order q,

(1+81z-1 + 2-2 +...+qZ-q

Et prediction error, Ct Ct
V backward difference operator (1 z-')
o standard deviation
o dead-time
T characteristic time constant
( damping coefficient
REFERENCES
1. Deming, W.E., Quality, Productivity, and Competitive Posi-
tion, Center for Advanced Engineering Study, Massachu-
setts Institute of Technology, Cambridge, MA (1982)
2. Jacobs, D.J., "Watch Out for Non-Normal Distributions,"
Chem. Eng. Prog., 86, 11, 19 (1990)
3. Levinson, W., "Understand the Basics of Statistical Process
Control," Chem. Eng. Prog., 86, 11, 28 (1990)
4. Block, S.R., "Improve Quality with Statistical Process Con-
trol," Chem. Eng. Prog., 86, 11, 38 (1990)
5. Grant. E.L., and R.S. Leavenworth, Statistical Quality Con-
trol, 5th ed., McGraw-Hill Book Co., New York (1980)
6. Downs, J.J., and J.E. Doss, "Present Status and Future
Trends: A View of the North American Industry," Proceed-
ings of the Fourth Int. Process Control Conf., San Padre
Island, TX, February (1992)
7. Deshpande, P.B., R.E. Hannula, M.A. Bhalodia, and C.W.
Hansen, "Achieve Total Quality Control of Continuous
Processes,"Chem. Eng. Prog., 89, 7 (1993)
8. Smith, O.J.M., "Close Control of Loops with Dead Time,"
Chem. Eng. Prog., 53, 217 (1957)
9. Moore, C.F., "Selected Problems in Design and Implemen-
tation of Direct Digital Control," PhD Thesis, Louisiana
State University (1969)
10. Astrom, K.J., Introduction to Stochastic Control Theory,
Academic Press, New York (1970)
11. Box, G.E.P., and G.M. Jenkins, Time Series Analysis: Fore-
casting and Control, Holden Day Publishers, Oakland, CA
(1976)
12. MacGregor, J.F., "On-Line Statistical Process Control,"
Chem. Eng. Prog., 84, 10 (1988)
13. Garcia, C.E., and M. Morari, "Internal Model Control: 1. A
Unifying Review and Some New Results," Ind. Eng. Chem.
Proc. Des. & Dev., 21, 308 (1982)
14. Allison, P., and N. Dumont, "Effects of Correlation on Sta-
tistical Process Control Charts," Pulp and Papermaker, 5,
(1990)
15. Montgomery, D.C., and C.M. Mastrangelo, "Some Statisti-
cal Process Control Methods for Autocorrelated Data," J.
Quality Tech., 23, 3 (1991)
16. Yeager, R.L., and T.R. Davis, "Reduce Process Variation
with Real-Time SPC," Hydrocarbon Proc., 71, 3 (1992)
17. Coughanowr, D.R., Process Systems Analysis and Control,
2nd ed., McGraw-Hill Book Co., New York (1991)
18. Seborg, D.E., T.F. Edgar, and D. Mellichamp, Process Dy-
namics and Control, John Wiley & Sons, Inc., New York
(1989)
19. Smith, C.A., and A.B. Corripio, Principles and Practice of
Automatic Process Control, John Wiley & Sons, Inc., New
York (1985)
20. Stephanopoulos, G., Chemical Process Control, Prentice-
Hall, Inc., Englewood Cliffs, NJ (1984) 0


Summer 1993










Random Thoughts...




TEACHING TEACHERS

TO TEACH

The Case for Mentoring


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


eaching-like medicine, auto mechanics,
professional basketball, and chemical en-
gineering-is a craft. There are distinct
skills associated with its practice, which people
are not born knowing. Some people are naturals
(in education, the so-called "born teachers") and
seem to develop the skills by intuition; most are
not, however, and need years of training before
they can function at a professional level. Doc-
tors, mechanics, basketball players, engineers,
and teachers at the K-12 level routinely get such
training-but not college professors, most of
whom get their PhDs, join a faculty, and set off
to teach their first course without so much as
five seconds on how one does that.
Not realizing that there are alternatives, new
professors tend to default to the relatively inef-
fective teaching methods they experienced as stu-
dents. Although they work hard to make the
course material as comprehensible and interest-
ing as they can, many of them consistently see
only glazed or closed eyes during their lectures,
terrible test grades, and evaluations suggesting
that the students liked neither the course nor
them. Some of them eventually figure out better
ways to do their job; others never do and spend
their careers teaching ineffectively.
The absence of college teacher training is not
an unrecognized problem, and at least some in-
stitutions are trying to address it. Various schools
offer graduate courses on teaching, hold faculty
teaching workshops lasting anywhere from one


Richard M. Felder is Hoechst Celanese
Professor of Chemical Engineering at North
Carolina State University. He received his
BChE from City College of CUNY and his
PhD from Princeton. He has presented
courses on chemical engineering principles,
reactor design, process optimization, and
effective teaching to various American and
foreign industries and institutions. He is co-
author of the text Elementary Principles of
Chemical Processes (Wiley, 1966).

morning to several days, and provide teaching
consultants to critique end-of-course evaluations
and videotaped lectures. Although such programs
are worthwhile and should be standard on every
campus, there are limits to what they can accom-
plish. You can't turn someone into a skilled pro-
fessional in a one-semester course, much less in a
three-day workshop or a two-hour consultation.
True skill development occurs only through re-
peated practice and feedback.
Fortunately, the resources needed for effective
training of college teachers are readily available
on every campus. Most academic departments
have one or more professors acknowledged to be
outstanding teachers by both their peers and their
students. They have learned how to put together
lectures that are both rigorous and stimulating
and homework assignments and tests that are
comprehensive, challenging, instructive, and fair.
They have found ways to motivate students to
want to learn, to co-opt them into becoming ac-
tive participants in the learning process, to help
them develop critical and creative thinking and
problem-solving abilities.
Unfortunately, under our present system, fac-
ulty members may collaborate on research but


@ Copyright ChE Division ofASEE 1993


Chemical Engineering Education









generally don't even talk to each other about teach-
ing. Most professors must therefore plod through
the same lengthy trial-and-error process when
learning how to teach, seldom benefitting from
the knowledge and experience of their colleagues.
Here is a proposal for what I believe might be a
better way.

> All new professors should team-teach their
first two courses with colleagues who have
earned recognition as excellent teachers
and who agree to function as mentors.
The first course would begin with the men-
tor taking most of the responsibility for
laying out the syllabus and instructional
objectives, planning and conducting the
class sessions, and constructing the home-
work assignments and tests. Both profes-
sors would attend most classes and have
regular debriefing sessions to go over what
went well, what didn't go so well and
why, and what to do next. The protdge
would gradually take over more of the
course direction, ending up with primary
responsibility by the end of the course.
> In the second course, the protege would
take sole responsibility for planning and
delivering the course. The mentor (who
may be the mentor from the previous se-
mester or a different professor) would
function entirely as a consultant, observ-
ing class sessions and participating in
debriefing meetings.
> When planning teaching assignments, the
department head should recognize that
team-teaching a course and serving as a
mentor to a new instructor is a heavier
time burden than simply teaching a course
alone, and should provide a suitable re-
duction in the mentor's other responsi-
bilities. Ideally, the mentor would get ad-
ditional compensation, such as a summer
stipend, release time, or a travel grant.

The potential benefits of this plan are evident.
New professors would get a jump-start on learn-
ing their craft rather than having to rely entirely
on painfully slow self-teaching. The experience
would likely energize the mentors as well, stimu-


lating them to reexamine and improve their own
teaching as they provide active guidance to their
junior colleagues. The overall quality of the
department's instructional program would inevi-
tably improve.
Caution, however-mentoring is also a craft,
with its own assortment of skills and pitfalls. As
it happens, teacher educators have explored this
subject for decades and have developed a variety
of methods to make mentoring successful.* If you
find yourself serving as a mentor, formally or
informally, consider the following guidelines:

When you teach, you often do subtle things
that you learned by experience, and you also
occasionally make errors in judgment when
handling classroom situations. The inexperi-
enced observing protdge is likely to miss it all.
Go over items in both categories during
debriefings.
When protdgds get into trouble in class, fight
off the temptation to rescue them immediately.
Instead, prompt them in debriefings to figure
out for themselves what went wrong and how
to fix it.
Offer suggestions, not prescriptions. What
you lay out for protdgds explicitly is unlikely to
stick. What they discover for themselves with
your help, they will own.
Don't try to turn your protdges into clones of
you. Instead, help them find the teaching style
best suited to their own strengths and person-
alities and encourage them to develop and per-
fect that style.

Only one step remains to complete the process.
When a department colleague-perhaps one of
your proteg6s-starts to win teaching awards, talk
her into serving as a mentor for the next faculty
hire. When she protests that she doesn't know
how, pass along this column and add that while
she's figuring it out you'll be happy to be her
mentor. 0

* I am indebted to Dr. Rebecca Brent, my mentor on
all matters related to teacher education, for many of
the ideas that follow. See also T.M. Bey and C.T.
Holmes, Mentoring: Contemporary Principles and Is-
sues, Reston, VA, Ass'n. of Teacher Educators (1992)


Summer 1993










M 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, MI 48109-2136.


WHEN IS A THEORETICAL STAGE

NOT ALWAYS A

THEORETICAL STAGE?


W. E. JONES
University of Nottingham
Nottingham, England NG7 2RD

he study of continuous distillation is one of
the cornerstones of an undergraduate course
in chemical engineering. Indeed, the pre-
sentation of the McCabe-Thiele construction is
a well-rehearsed routine. Therefore an exercise
which makes more experienced undergraduates
reconsider widely used textbook assumptions and, in
addition, to think more broadly about the subject is
very useful.

BACKGROUND
The McCabe-Thiele analysis starts with a diagram
much as the one shown in Figure 1.[1"4 The reboiler
of Figure 1 must be a kettle (see Figure 2) in order
to phase-separate vapor and liquid and meet the
requirement that vapor return and bottoms be in
equilibrium. Indeed, some undergraduate textbooks
provide this detail.'"57
Kettle reboilers, however, are not widely used be-

Warren Jones holds BSc and PhD degrees in
chemical engineering from the University of
Nottingham and is a registered Chartered En-
gineer. He has wide-ranging interests in both
front-end process and detailed plant design,
developed initially through nine years experi-
ence with a major engineering and construc-
tion company. Teaching responsibilities include
several design courses, process economics,
and engineering thermodynamics.
Copyright ChE Division ofASEE 1993


CooL on


DL~stLLLate







m
---> Bottoms


LiquLd F
Ln Sump


Figure 1

cause they have some significant disadvantages. The
most commonly cited disadvantages are:[8'91
The high cost of the large shell to permit vapor-
liquid phase separation and provide bottoms surge
volume.
Their tendency to collect dirt and to foul.
A liquid pool submerging the reboiler tubes, to a
first approximation, operates uniformly at the bot-
toms composition and hence boils at the highest
temperature; this narrows the temperature driving
force and leads to increased heat transfer area.
It is important to note that kettle designs per-
mit high vaporizations-up to 80% of the liquid
from the bottom tray of the column. Columns can
Chemical Engineering Education


I









thus be designed with liquid/vapor traffic ratios as low as 1.25.
Thermosyphon reboilers are much more commonly used. They
may be either vertical units with vaporization inside the tubes, or
horizontal units with vaporization inside the shell. Thermosyphon
designs rely on a head of liquid (achieved by precise positioning of
the reboiler relative to the column) and careful piping design to
force the return of a two-phase mixture to the column.
Thermosyphons are popular because:E8'91
Their design normally allows for a high process flowrate through
the unit; this is beneficial in reducing fouling and promoting a
high heat transfer coefficient.
Both phase separation and surge volume are moved to the column
sump, where these functions are more easily managed.
Boiling occurs over a range of temperatures, and hence there is an
improved driving force for heat transfer.
A cheaper overall construction is obtained because of the above
factors.


Once- through
OperaLton
n

SBot
L.
Twc



RebotLei
Surge Votume


Re-cLrcuLatL on
Operation


Bottoms


Figure 3a Figure 3b
Summer 1993


One particular feature of
thermosyphon design is a strict upper
limit on the per-pass vaporization
achievable, normally about 25% of the
process liquid entering the unit with
vaporization in the range of 5-15% be-
ing normal. Indeed, a high percentage
of liquid in the two-phase return is an
advantage because it ensures that the
reboiler tubes are kept "wet," thus im-
proving heat transfer and minimizing
fouling. If a thermosyphon is connected
to a column in an arrangement analo-
gous to that for a kettle (contributing
one theoretical stage with exiting va-
por and liquid in equilibrium), then Fig-
ure 3a applies. But the upper limit on
per-pass vaporization now poses a se-
vere difficulty for the distillation de-
sign because liquid/vapor traffic ratios
below 4.0 are not possible. Hence, once-
through operation is only generally suit-
able for reboiled strippers.
Far more flexibility is achieved by
recirculation operation (see Figure 3b).
Now a low liquid/vapor traffic ratio can
be achieved in the column because the
process flowrate through the reboiler
is boosted by recirculation of part of
the returning liquid, keeping the per-
pass vaporization to acceptable levels.
There are two disadvantages, however:

1. The vapor rising from the reboiler
toward the bottom tray is not in
equilibrium with the bottoms liquid.
Hence, the recirculation reboiler has a
separation performance that is
equivalent to less than one theoretical
stage (this answers the question posed
in the title of this article).
2. A portion of the bottoms liquid will
experience a long residence time in the
column sump, resulting in repeated
contact with the hot tube surfaces of
the reboiler-this may promote
fouling.

Despite these disadvantages,
recirculating reboilers are widely used,
and adapting the McCabe-Thiele con-
struction is a challenging exercise for
more experienced undergraduates.
This exercise is useful, not only be-
cause it brings real industrial practice


Heating
MedLum


LLquLd PooL
tn Sump


II Bottoms
Figure 2









to the students' attention (often significantly different from
the theory presented in textbooks), but also because it may
be widened to include discussion of other aspects, such as
Kettle versus thermosyphon reboilers
Vertical versus horizontal thermosyphons
Exchanger cleaning
Thermosyphon operation
Distillation column elevation
Space requirements for different reboilers
Detailed design of distillation column internals
This article concentrates on the McCabe-Thiele con-
struction. Normally, it is helpful for the instructor to sum-
marize some of the discussion contained in the "background"
section above before students commence with the following
"problem."

Problem Statement
The McCabe-Thiele construction for distillation is presented
as taking credit for one theoretical stage in the reboiler. This
is based on the traditional use of a kettle reboiler (see Figure
2). Industrial practice is often different, however, and in-
volves using either a once-through or a recirculation
thermosyphon (see Figure 3).

(a) Which of the two reboiler arrangements shown in Figure
3 is not equivalent to a theoretical stage? Give your
reasons.
(b) The flows and more volatile component (MVC) composi-
tions in the bottom section of a distillation column fea-
turing a recirculation thermosyphon are shown in Fig-
ure 4. Derive the equation
Lxn + (R V)xr (
Xb= (L+R-V)
and use the result to assist sketching recirculation re-
boiler operation on a McCabe-Thiele diagram. Comment
on the asymptotic behaviour when R is very large com-
pared to L and V, and when R = V.
(c) A two-component mixture, having relative volatility of
2.0, is distilled to produce a bottom product with xb = 0.1.
If the liquid/vapor traffic ratio below the feed is 1.5, and
15% of the reboiler feed is vaporized per pass, draw a
detailed McCabe-Thiele construction (making the usual
simplifying assumption) for the recirculation reboiler and
bottom tray.
(d) Express the separation performance as a liquid-phase
Murphree efficiency.
(e) Devise column sump internals which give the advantage
of recirculation but ensure that the reboiler behaves as a
theoretical stage.


Solution
(a) The arrangement of Figure 3b is not
equivalent to a theoretical stage as ex-
plained in the "background" section.
(b) Equation (1) is very simply derived as a
MVC mass-balance around the column
sump and may be interpreted as express-
ing bottom product composition as a
blend of bottom tray liquid and reboiler
return liquid. Figure 5 shows the re-
quired McCabe-Thiele construction
where recirculation operation is repre-
sented as a partial step (rather like the
construction taking into account tray ef-
ficiency) between operating and equilib-


Figure 4


Figure 5

Chemical Engineering Education


Tray L





Tray n-1
Tray n









rium lines. Point (x,, y,) must lie on the operat-
ing line, and (x,, y,) on the equilibrium line; Xb
lies between them in accordance with Eq. (1)
and controls the step size.
When R is very large compared to L and V, then
Xb approximates x and the system behaves as a
theoretical stage.
When R = V (i.e., the minimum reboiler
feedrate), then Xb equals x, and no separation is
achieved.
(c) The general equation for the operating line, us-
ing the symbols of Figure 4, is
L B
Yi+1= V Xi xb (2)
Equation (2) may be applied at any elevation
below the feed. Hence

yr =Lxn-VXb (3)
The reboiler return comprises vapour and liquid
in equilibrium. Hence y, and x, are related by
axr
Yr l+(a 1)xr (4)

Recirculating systems have an additional degree
of freedom-namely the process flowrate, R,
through the reboiler-and it is this variable that


Figure 6
Summer 1993


fixes the size of the partial step. An extra equa-
tion is required for the analysis, and it is ob-
tained by a MVC mass-balance around the re-
boiler


Rxb = (R V)xr + Vyr


Xb = (1- f)Xr +fYr (6)
where the reboiler fraction vaporization, f =
V/R, is a design parameter.
To perform the construction, Xb is known,
but three compositions (x,, x,, y,) are unknown.
There are three equations in these unknowns,
however-namely Eqs. (3), (4), and (6). Using
the data in the Problem Statement,

L-=15, so B=0.5, and f= 0.15
V V
we obtain
Yr = 15 xn-0.05 (7)

Yr (8)
1+xr
0.1= 0.85xr +0.15Yr (9)

Eliminating y, between Eqs. (8) and (9) gives a
quadratic in x, having feasible solution x, =
0.08885. Figure 6 shows the completed construc-
tion where y, = 0.1632 and x, = 0.142.
(d) The normal convention is to express stage mass
transfer efficiency in terms of vapor composi-
tion change. In this instance, however, there is
no vapor feed to the stage, and the efficiency
must be expressed in terms of liquid composi-
tion change,

E=(xn-xb x100= 0.142-0.1 x 100=79% (10)
Xn -xr 0.142- 0.08885
Alternatively, if Eq. (1) is subtracted from x, =
x., the efficiency of the reboiler can be shown to
be
E R-V (1)
100 L+R-V ( '
Thus, knowing that L/V = 1.5 and V/R = 0.15,
then L/R = 0.225, and we again obtain E = 79%.
Equation (11) would be useful if the equilib-
rium data is not expressed in a simple form
permitting analytical solution. Trial-and-error
solution could be used to find x, and x, such
that the predetermined efficiency is obtained.
(e) This part is a challenging exercise for students,
since the problem is at two levels: first, finding
a philosophy, and second, devising equipment


Bottom Tray


0 20 Rebo i ler



























Figure 7
to implement the philosophy. In summary, the
philosophy is to preferentially route the bottom
tray liquid to the reboiler and sufficient reboiler
return liquid to the bottoms. Excess reboiler
return liquid (not needed as bottoms) joins the
bottom tray liquid as reboiler feed. One way of
implementing this on larger diameter columns
is illustrated in Figure 7; note that the down-
pipe from the trap-out mixes the bottom tray
liquid into the excess reboiler return liquid well
below the overflow level-this ensures the over-
flow is predominantly reboiler return liquid.
With this type of design, the vapor return and
bottoms are in equilibrium so the system acts
as one theoretical stage.

INDUSTRIAL PRACTICE
The loss of part of a theoretical stage is widely
recognized in industry, but response varies consid-
erably. At one extreme some companies do not take
credit for the separation due to the reboiler (treat-
ing it as a safety factor) unless it is a kettle when
full credit is taken for one theoretical stage. At the
other extreme, others simply count the reboiler (ir-
respective of type) as one theoretical stage-but then
a conservative tray efficiency applied to the column
hides any shortcoming. Based on the approach in
this article, it seems that a recirculating reboiler
generally has a separation efficiency of at least 60%.
Hence, even though one might not wish to actually
perform the construction as described, it would be
quite safe to credit the reboiler with a separation
equivalent to an efficiency of 50%.

REFERENCES
1. Henley, E.J., and J.D. Seader, Equilibrium-Stage Opera-
tions in Chemical Engineering, John Wiley & Sons, New


Two-Phase
Return



ReboiLer




Bottoms


York, p. 323 (1981)
2. King, C.J., Separation Processes, McGraw-Hill Book Co.,
New York, p. 216 (1971)
3. Treybal, R.E., Mass-Transfer Operations, 3rd ed., McGraw-
Hill Book Co., New York, p. 373 (1980)
4. Van Winkle, M., Distillation, McGraw-Hill Book Co., New
York, p. 195 (1967)
5. Coulson, J.M., and J.F. Richardson, Chemical Engineering,
Vol. 2, 3rd ed., Pergamon Press, Oxford, England, p. 422
(1978)
6. McCabe, W.L., and J.C. Smith, Unit Operations of Chemi-
cal Engineering, 3rd ed., McGraw-Hill Book Co., New York,
p. 549 (1976)
7. Smith, B.D., Design of Equilibrium Stage Processes,
McGraw-Hill Book Co., New York, p. 127 (1963)
8. Jacobs, J.K., Hydrocar. Proc. and Petrol. Refiner, 40(7), p.
189 (1961)
9. Palen, J.W., "Introduction to Shell-and-Tube Reboilers," page
3.6.1-1 in Hewitt, G.F. (ed), Hemisphere Handbook of Heat
Exchanger Design, Hemisphere Publishing Corporation
(1990) 0


[Mn book review

CHEMICAL ENGINEERING:
Vol. 1. Fluid Flow, Heat Transfer and
Mass Transfer
by J.M. Coulson and J.F. Richardson, with J.R.
Backhurst and J.H. Harker
Pergamon Press, Headington Hill Hall, Oxford, OX3
OBW, United Kingdom; $48 (paperback) (1990)

Reviewed by
Chang-Won Park
University ofFlorida

This book is an undergraduate text on the unit
operations of chemical engineering and is published
in the United Kingdom. Since the first edition was
published in 1954 it has been revised three times,
updating the material as significant developments
in chemical engineering have been made. The mate-
rial is divided into thirteen chapters: the first eight
chapters are on fluid flow, and the following two
chapters are devoted to heat and mass transfer,
respectively; Chapter 11 gives a brief overview of
the boundary layer theory, and in Chapter 12 the
molecular diffusion in momentum, heat, and mass
transfer is described; finally, humidification and wa-
ter cooling are treated in Chapter 13.
The level of treatment seems adequate for under-
graduate students who have an elementary knowl-
edge of material and energy balances but who may
not have taken a course in transport phenomena.
This book uses slightly different nomenclature than
textbooks published in the U.S., but not to an extent
Chemical Engineering Education









that will cause a serious problem. Compared to Unit
Operations of Chemical Engineering (4th edition),
by McCabe, Smith, and Harriott (which may be
the most widely used textbook in the U.S. on
the subject), this book contains more example prob-
lems, more figures, and more pictures (especially for
the section on fluid flow), which may be an impor-
tant feature for a textbook. It also contains more
detailed design aspects of pumps, flow meters, heat
exchangers, etc.
Units and dimensions are briefly covered in Chap-
ter 1, followed by elementary thermodynamic prin-
ciples in Chapter 2. Chapter 3 describes the flow in
pipes and channels, including an adequate level of
description for non-Newtonian behaviors. The flow
of compressible fluids is described in Chapter 4.
Starting with the flow of gas through a nozzle or
orifice, the unique features of compressible fluid flow
are well enough described so that students can com-
prehend the subject matter without too much diffi-
culty. Multiphase flow, which is important in many
areas of chemical engineering but which is a diffi-
cult subject to handle at the undergraduate level, is
treated in Chapter 5. The empirical developments of
liquid-gas and fluid-solid systems are described, in-
cluding up-to-date literature references. Chapters 6,
7, and 8 describe flow measurements, liquid mixing,
and pumping of fluids, respectively.
Chapter 9 is devoted to heat transfer. This long
chapter covers the fundamentals of conduction, con-
vection, and radiation as well as heat transfer in-
volving a phase change and heat exchangers. The
material and the level of treatment are similar to
those of many other undergraduate textbooks. Fi-
nally, mass transfer is treated in Chapter 10. How-
ever, only the fundamentals of mass transfer are
described in this volume-the various mass transfer
processes such as distillation, liquid-liquid extrac-
tion, and gas absorption are covered in Volume 2 of
the Chemical Engineering Series, Particle Technol-
ogy and Separation Processes. The flow past im-
mersed bodies, including fluidized beds and packed
beds, is also covered in Volume 2.
In summary, this volume can serve as an excel-
lent textbook or a principal reference for an unit
operations course on fluid flow and heat transfer.
Chapters 1 through 9 constitute an adequate amount
of material to be covered in one semester, and the
many example and homework problems contained
in this volume are a useful feature. The treatment
of mass transfer, however, is rather brief and a sepa-
rate volume must be used if a course is to be de-
voted to mass transfer operations. O
Summer 1993


M book review

CHEMICAL ENGINEERING:
Vol. 2. Particle Technology and
Separation Processes, 4th ed.
by J.M. Coulson, J.F. Richardson,
J.R. Backhurst, and J.H. Harker
Pergamon Press, Headington Hill Hall, Oxford, OX3
OBW, United Kingdom; 968 pgs, $51 (paperback) (1991)
Reviewed by
Benjamin J. McCoy
University of California, Davis
This book, nearly 1000 pages in length and weigh-
ing over four pounds, is a bargain at $51.00. Part of
a six-volume introduction to chemical engineering,
this particular volume covers separation and par-
ticle processes. The rule-of-thumb that any book in
its 4th edition is worth knowing applies in this case.
The authors have prepared a carefully written and
judiciously planned book which is rewarding to read
and study. Material from the 3rd edition has been
revised, reordered, and rewritten, and new chapters
have been added on adsorption, ion exchange, and
chromatographic and membrane separations.
With the need for chemical engineering to expand
beyond its traditional central role in the petrochemi-
cal industries, this book provides a satisfactory back-
ground for the particle and separation technologies
important to biotechnology, biomedical applications,
materials science, and environmental engineering.
It will serve nicely as either a handbook on the shelf
of the practicing chemical engineer or the teacher of
chemical engineering, or as a textbook used in a
course on separations and applied mass transfer.
The prerequisite courses are introductory physics,
chemistry, and calculus. The book has an adequate
supply of homework problems and an abundance of
cited references for the researcher.
The authors have maintained a commendable bal-
ance of practical engineering and mathematical fun-
damentals. Necessary for application to industrial
applications, the book is well stocked with photo-
graphs, diagrams, and explanations of equipment.
The book supplements the now-usual mass, momen-
tum, and energy transport phenomena approach and
includes thermodynamics of adsorption, physics of
particles, and dynamics of chromatography. Treat-
ing the essential physical processes, the authors
present concise derivations of mathematical rela-
tionships that succinctly capture the significant, ba-
sic quantitative concepts.
Continued on page 199.
183










laboratory)


A COMPREHENSIVE

PROCESS CONTROL

LABORATORY COURSE


P. T. VASUDEVAN
University of New Hampshire
Durham, NH 03824

Process Control and Simulation is a 4-credit
course in the department of chemical engi-
neering at the University of New Hampshire.
Classroom lectures (three hours per week) are
supplemented with a two-and-a-half hour labora-
tory session held once a week. The course is tradi-
tionally taught in the spring semester of the senior
year. The topics include, but are not limited to, dy-
namic behavior of chemical engineering processes
described by differential equations, feedback control
concepts and techniques, stability analysis, and ad-
vanced control techniques.
Students are usually divided into groups of two or
three, and each group is required to do six different
experiments over the course of the semester. These
experiments are designed to expose the students to
the practical aspects of almost all the theoretical
topics covered in class. The basic materials and
equipment are supplied for all the experiments. The
students have to assist in designing and building
the experiments, decide a priori what data they want
to collect, perform the experiments, analyze the data,
and submit a report. Some of the experiments are
designed to permit flexibility in terms of simulating
various process configurations (first order, second
order, third order) or to demonstrate various pro-
cess control principles discussed in class. The im-





P.T. Vasudevan is a chemical engineering fac-
ulty member at the University of New Hamp-
shire. He obtained his PhD from Clarkson Uni-
versity. His research interests are in the area of
catalysis and biocatalysis.
Copyright ChE Division ofASEE 1993


portant features of each experiment will be high-
lighted in the following paragraphs.
The six experiments are divided into two phases.
The first three experiments (phase one) are per-
formed by all the students prior to the spring break,
while the remaining three experiments (phase two)
are performed after the spring recess.

EXPERIMENTS
Two of the three experiments in the first phase
deal with determination of time constants of simple
processes such as liquid level in a tank or pressure
in an air cylinder. The liquid-level process consists
of three Plexiglass tanks with interconnecting valves.
Water can be pumped to any of the three tanks, and
the feed water pressure is maintained constant to
avoid any fluctuations in the flow rate. The control
valve is located on the feed line. The students are at
liberty to select one, two, or all three tanks and set
the system up as either an interacting or a
noninteracting process. The level is monitored in
the third tank by means of a pressure transducer
mounted at a height of six inches above the tank
bottom, and the signal from the transducer is then
sent to a PC equipped with data-acquisition capa-
bilities (in this case, a Metrabyte DAS-8 card).
Labtech Notebook is used to set up the various in-
put and output channels. The PC is equipped with
additional Metrabyte boards for process control. The
students are thus exposed to various features of
data acquisition and control, and instrumentation
hardware very early in the semester.
In order to determine the time constant of the
process, the students have to use both a pulse- and
a step-forcing function. These forcing functions are
set up in an external file (in ASCII) and can be
accessed by Labtech Notebook when needed. Thus,
data pertaining to the magnitude of the step or pulse
and the type of pulse are stored in this external file.
Chemical Engineering Education









The controller (output channel) in open loop reads
the information from this file and changes the out-
put to the control valve accordingly. The duration of
the pulse or the exact moment at which the step
change is to be introduced is controlled by adjusting
the sampling rate in the output channel. The entire
operation is therefore carried out in a precise fash-
ion, with very little human intervention.
The data are recorded in a file and are also con-
tinuously monitored on the VGA monitor, so the
students can compare the experimentally obtained
value for time constant with the theoretical value
(knowing the valve resistance and area of the tank).
A linear valve is used in the experiment, and the
valve resistance can be easily determined experi-
mentally. The determination of time constant for a
step-forcing function is straightforward. For the case
of a pulse-forcing function, the following method is
used for determining the time constant for a first
order process (single tank).
We let
R = linear valve resistance
H = magnitude of the pulse
T = duration of the pulse
hd(s) = height in the tank in terms of deviation
variables
Then, in the Laplace domain
h ) RH (1--e-s (1)

In the time domain, for t > T,
hd(t) = RH(e-(t-T)/T e-t/I) (2)
We now define a function, f, equal to the product
of hd(t) and time, t. A plot of f versus time t will go
through a maximum. By differentiating Eq. (2) with
respect to time and equating it to zero, we can show
that the maximum occurs when t = T. Thus, this
method gives a simple procedure for estimating the
time constant for a first-order process. Alternately,
a plot of in hd(t) vs. time is a straight line with a
slope equal to the reciprocal of z. However, Eq. (2)
does not take into consideration the response of the
process for values of t < T. If the duration of the
pulse is sufficiently long, it is necessary to consider
the complete solution.
This problem is easily solved in the following man-
ner. It is possible to delay the storage of information
by specifying a time delay equal to the duration of
the pulse, T. By setting up a "calculated channel" it
is therefore possible to monitor and store time in an
external file as (t T), (referred to hereafter as
"adjusted time"). In the next channel, the data are
Summer 1993


stored or displayed as the product of height (in de-
viation variables, also easily set up in notebook
through the use of "calculated channels" once the
initial steady-state height is known), and adjusted
time. The product of height and adjusted time (func-
tion f) versus the adjusted time is continuously dis-
played on the screen (and also stored in an external
file) so that the information can be plotted later on.
Such a plot is shown in Figure 1. (Since the sam-
pling rate is 1 Hz, the data points are not shown.)
From this plot the time constant can be determined
as the value on the abscissa corresponding to the
value on the ordinate where the function f goes
through a maximum. Or, to obtain an accurate esti-
mate, a differential analysis of the data (function f
with respect to time) can be performed.
The value of the time constant from the plot is
about 210 seconds. This compares very well (within
5%) with the value of time constant obtained using
a step change. For this particular experiment, the
duration of the pulse was 100 seconds and the mag-
nitude of the pulse (change in flow rate) was 0.32
ft3/min. It is interesting to note that the valve resis-
tance can easily be determined once the time con-
stant for the process is known. Setting t = t is Eq.
(2), we get
hd(T)=f =0.368RH(eT/ 1)
The only unknown in this equation is R, and it can
be determined.
The same experimental setup is used to introduce
concepts such as transmitter gain and dead time.
For instance, since a pressure transducer is used to
measure the height in the tank or the pressure in
the cylinder (in the air-pressure process exper-
iment), the students are required to calculate the


Figure 1. Determination of time constant from
a pulse test. Sampling rate = 1 Hz









transmitter gain. This information is then entered
into the input channel. The students thus gain an
understanding of how a transducer works and the
range of the output signal for electrical and pneu-
matic transducers.
The first fifteen to twenty minutes of each labora-
tory period is spent in demonstrating the process-
control principles discussed in class the previous
week. For example, the liquid-level experiment is
used to demonstrate types of controller action and
how to set up the right action (reverse or direct) in
the output channel. The PC is equipped with a
Metrabyte DDA-06 controller card. The phenomenon
of reset windup and the concept of stability are also
demonstrated soon after the theoretical material is
presented in class. Since Labtech Notebook uses the
position form of the controller equation, reset windup
is demonstrated very effectively.
The students also develop a good understanding
of the dynamics of PID control and the effect each
element (P, I, D) has on the overall control process.
Important concepts such as offset, or how a simple
first-order process with PI control can behave in an
oscillatory manner, or how a second-order
overdamped process with simple proportional con-
trol can become underdamped, are demonstrated
with ease.
The pressure experiment is similar to the liquid-
level experiment. Students are required to deter-
mine both the experimental and the theoretical time
constants and to compare the two. They must deter-
mine the transmitter gain (scale factor) and offset
and set up various channels in the Notebook.
The third experiment deals with control-valve cali-
bration for both liquid and gas service. Here the
students gain a practical understanding of concepts
such as inherent and installed characteristics, valve
coefficient, and valve flow characteristics. Once
again, Notebook is used to set up various channels.
A Metrabyte DAC-02 card is used to change the
signal to the transducers located on the control valves
for both liquid and gas, in increments of 1V (range
is -5 to +5V). The students take data of flow rate,
valve stem position, current signal to transducer (4-
20 mA), upstream pressure, and downstream pres-
sure. They are required to calculate and report the
valve coefficients of the two valves as well as the
type of valve (linear, equal percentage, quick open-
ing) from suitable plots of the valve characteristics.
In their report, the students are required to com-
ment on the phenomenon of hysterisis observed in a
plot of valve coefficient versus valve stem position.
The second phase of the laboratory deals with con-
186


troller tuning based on Ziegler-Nichols closed-loop
settings, Cohen-Coon open-loop tuning setting, or
data from a pulse test. The liquid-level experimen-
tal setup or the air-pressure process can be used
again. In the case of the liquid-level experiment, the
students can choose any configuration they like.
From the process reaction curve generated (here
again, Labtech Notebook is used to set up the chan-
nels and store the information), the students use
Cohen-Coon setting to determine the controller set-
tings. They select a controller (P, PI, or PID) and
determine the response to both servo and load
changes. The controller settings obtained from the
process reaction curve serve as preliminary esti-
mates, and the students are required to obtain
the optimum settings using a dynamic criterion
such as IAE, ISE, or ITAE. This is easily done
through the use of various "calculated channels" of
Notebook, and IAE, ISE, and ITAE are set up in
different channels.
The display window for the monitor is divided into
four sections, and the students can observe the ac-
tual height in the tank, the error, IAE, ISE, or ITAE.
They are required to select one of the integral crite-
ria and try to obtain the optimum controller set-
tings. This is done by keeping the reset time con-
stant, for instance, and changing the proportional
gain and determining the response to a unit step
change in the set point (always from the same value).
The students then change the integral time (keep-
ing the gain constant) and observe the response. In
each case the integral value is reported.
The second experiment also deals with controller
tuning. This is done using the Ziegler-Nichols closed-
loop tuning method. The second half of this experi-
ment consists of using a pulse test to generate a
Bode plot. The objective of the experiment is to de-
termine the open-loop transfer function and calcu-
late the overall gain, time constant, and dead time,
if any. The students have to decide on a proper
pulse duration and magnitude.
The pulse is introduced by changing the position
of the control valve, and hence the flow rate to the
system, for a known duration. This is achieved by
setting the output channel in "open loop," which in
turn accesses an external file to obtain values of the
controller output. Care is taken to ensure that the
system returns to its original steady state. The in-
put and output data are then Fourier-transformed
and divided to give the system transfer function in
the frequency domain, G(iw). From the amplitude
ratio and phase angle, Bode plots are constructed
and the various parameters determined. The calcu-
Chemical Engineering Education























0.0001 0.001 0.01 0.1
Frequency, rad/sec
Figure 2. Bode plot generated from a pulse
test: magnitude ratio versus frequency.


lation of G(iw) from the pulse data is achieved by representing
the transfer function as

Jy(t)cos(wt)dt- ify(t)sin(wt)dt
G(iw) = o 0
Jx(t) cos(wt) dt i x(t)sin(wt)dt
0 0
where x(t) and y(t) are the input and output functions.
Then
G(iw) (AC + BD)+ i(AD BC)
C2 + D2
where


A= jy(t)cos(wt)dt
0
Tx
C = Jx(t)cos(wt)dt
0


Ty
B= Jy(t)sin(wt)dt
0
TX
D = fx(t)sin(wt)dt
0


0.0001 0.001 0.01 0.1
Frequency, rad/sec

Figure 3. Bode plot generated from a pulse
test: phase angle versus frequency.


0 .







0.1
Frequency, rad/sec

Figure 4. Bode plot generated from a pulse
test: second-order system.
Summer 1993


The duration of the input pulse and the time it takes the
response to return to the original steady state, are T, and Ty,
respectively. The integrals are evaluated numerically by pick-
ing different values for the frequency, w. The students do this
on a mainframe computer after up-loading the data from the PC
to the mainframe. The experiment yields reasonably accurate
frequency response curves. Numerical integration becomes a
problem because of the oscillatory behavior of the sine and
cosine terms at high values of frequency.
Since there is practically no human input necessary while
performing this experiment, and because of the resolution
and sampling rate used, data noise is not a problem. The
Bode plots generated for a first-order liquid-level process are
shown in Figures 2 and 3. The magnitude ratio and phase angle
at higher values of frequency are not shown because of the
problems associated with integration. Figure 3 indicates that
the phase angle reaches an asymptotic value around -90,
which is indicative of a first-order system without dead time. It
is also evident from Figure 2 that the transfer function of the
system is exactly first order.
These observations are not surprising considering the fact
that the process is first order and there is no measurement lag.
The time constant for the process can be easily determined from
Figure 2 once the corner frequency is known. The time constant
is found to be about 200 seconds and is within 5% of the value
previously reported.
The magnitude ratio for a second-order process (two interact-
ing tanks) is shown in Figure 4. It is clear from the slope of the
high-frequency asymptote that the system is exactly second
order. Labtech Notebook also has a Fast Fourier Transform
(FFT) capability that can be used to generate a power spectrum.
In the above experiments, the students are also required to
study the effect of sampling rate on data acquisition and on the
control characteristics.
Continued on page 193.


Cc










r classroom


INTEGRATING COMMUNICATION

TRAINING INTO

LABORATORY AND DESIGN COURSES


KAREN R. PETTIT, RICHARD C. ALKIRE
University ofIllinois at Urbana-Champaign
Urbana, IL 61801

recently, an increased awareness that good
communication skills are essential in the
engineering professions has led many chemi-
cal engineering departments to stress technical com-
munication in the undergraduate curricula. In fact,
a recent informal Chemical Engineering Progress
surveym1 of the communication requirements in 156
U.S. and Canadian chemical engineering depart-
ments found that "with few exceptions, most of the
departments that responded [to the survey] were
placing greater emphasis on communication skills."
As this survey indicates, chemical engineering de-
partments take a variety of approaches to incorpo-
rating communication into the curriculum: some de-
partments require a course in technical communica-
tion from another division of the university; others
have developed courses which specifically empha-
size technical communication within chemical engi-
neering; while a third common approach is to inte-
grate communication training into existing courses.
We initiated our communication program at the
University of Illinois in 1989 by offering a junior-
level communication-intensive course which empha-
sized the interrelationship between technical prob-
lem solving and communication of the results. Over
the semester, the students completed several short
technical projects and one longer one, each requir-
ing some technical writing, revision, and oral work.
Each project specified a particular audience and goal
to be reached so that the students learned to struc-
ture their problem solving with the ultimate com-
munication goal in mind. Limited to thirty students,
the course provided individual attention and feed-
back, opportunities for discussion among students,
writing workshops to help build composition skills,
peer editing of both oral and written work, mock
meetings and interviews to simulate professional
188


Karen R. Pettit has worked as the Communi-
cation Instructor in chemical engineering at the
University of Illinois since 1990. She received
her BA from Swarthmore College and her MA
in English from the University of Illinois. She is
currently a PhD candidate in English at the
University of Illinois



Richard C. Alkire, professor and head of chemi-
cal engineering, has been a faculty member at
the University of Illinois since 1969. His research
interests in the field of electrochemical engi-
neering have focused on chemical synthesis,
energy conversion, corrosion, and deposition/
etching of surfaces.

experience, and an opportunity to view videotaped
presentations for self-evaluation.
Encouraged by the success of this elective course,
the department decided to extend practical writing
and speaking experience to all students by stressing
communication in the required senior-level labora-
tory and design courses, and we have continued to
concentrate on this integrated program the past three
years. Although students in the unit operations labo-
ratory course and the design course always have
been required to present their work through either
written or oral reports, simply requiring communi-
cation work does not necessarily help students com-
municate more effectively. Therefore, to supply fo-
cused instruction and feedback, we have employed a
communication instructor (CI) (someone from the
English Department, hired for a two-thirds time po-
sition) to help integrate communication training into
these senior-level courses. As this article will ex-
plain, through their experience in these two courses
the students not only practice writing and speaking
through a series of assignments evaluated for com-
municative ability and technical content, but they
also receive instruction on technical communication,
they learn to revise and edit their work as well as
Copyright ChE Division ofASEE 1993
Chemical Engineering Education










the work of their peers, and they gain experience in
collaborative writing and speaking.
We have found that this integrative approach pro-
vides an opportunity for students to practice techni-
cal communication despite a tight curriculum which
otherwise limited their communication work to a
freshman rhetoric course. More importantly, though,
it offers our students experience in writing and
speaking within their own discipline.[21 Working on
communication within a discipline provides students
with professional experience and promotes learning
through writing and speaking. Unlike technical com-
munication courses where assignments may be arti-
ficially created for students to practice communicat-
ing, emphasizing communication in existing courses
enables the students to use writing and speaking as
tools for discovery. Research in the teaching of writ-
ing has also shown that students can learn material
more fully through writing.t[3 As C. W. Griffin points
out, "We are beginning to realize that writing is not
just the end product of learning; it is a process by
which learning takes place."[41 Similarly, oral pre-
sentations can be approached as a learning tool. At
Illinois, we are working to create opportunities for
students to investigate and assimilate technical in-
formation through writing and speaking. When com-
munication is approached as an integral part of the
learning process, students start viewing it as an


essential part of their work in engineering rather
than as a chore which takes times away from learn-
ing technical material.

DESCRIPTION: COMMUNICATION COMPONENT
Unit Operations Laboratory Course
The communication component of the laboratory
course includes instruction on writing and present-
ing experimental reports, a series of individual writ-
ing assignments, a revision exercise, and two oral
presentations.
Course Description Table 1 outlines the course
requirements. The class is divided into lab sections
of, at most, fifteen students. Each lab section is
supervised by a graduate chemical engineering teach-
ing assistant (TA) and meets in the lab for five
hours per week. The oral presentations are held
during the first hour of the lab period in a separate
classroom, preferably equipped with an overhead
projector. When the lab section is large (twelve
to fifteen students), we divide the oral presen-
tations into two rooms to save time. The pro-
fessor and the TA then evaluate one set of talks
each, and the CI rotates through the two rooms.
Table 2 further outlines the required resources for
the communication component in terms of teaching,
space, and equipment.


TABLE 1
Course Descriptions
Course Enrollment Group Size Assignments Written Reports* Oral Presentations*
Unit Operations 20-55 students 3-4 students 6 labs 4 minor (3-5 pp.) 2 (15 minutes each)
2 major (8-10 pp.)
1 revision of 1st major
Design 20-55 students 2-3 students 1 design project 2 preliminary (5 pp.) 2 (10 minutes each)
1 final (20 pp.) 1 (20 minutes)
*In the unit operations course the written and oral reports are prepared individually,
whereas in the design course they are prepared in groups.

TABLE 2
Required Resources for Addition of Communication Instruction

Course Teaching Resources Classroom Space Equipment Salaries
Unit Operations One professor One laboratory and One overhead $6,000/academic year
(50 students) One communication two seminar rooms projector per for 33% time
instructor for oral seminar room communication
SOne 25% time TA presentations instructor
per 15 students

Design One professor One classroom and Two overhead $6,000/academic year
(50 students) One communication one seminar room projectors for 33% time
instructor for oral Two videocameras communication
SOne 25% time TA presentations One VCR for instructor
per 15 students viewing videos

Summer 1993


All of the reports, in-
cluding the revision,
are graded for techni-
cal content by the TA
or the professor, and
the two major reports
and the revision are
graded by the CI as
well. The oral presen-
tations also are graded
for both technical and
communicative quality.
In assessing the com-
municative quality for
both written and oral
reports, the CI empha-
sizes that successful
expression of technical
ideas requires more
than good grammar
skills or stylistic
choices. Thus, the com-
munication grades on
the major reports and
the revision are based
primarily on the orga-
189









nization and use of the technical format; the writing
style, grammar, spelling, and punctuation are
checked but are not emphasized. The communica-
tion grade for the oral reports assesses the level of
organization, the degree of preparation, and the pre-
sentation skills displayed. Altogether, the commu-
nication component comprises thirty percent of the
grade for the course.
Throughout the semester the CI holds office hours,
and students are encouraged to consult with the
instructor individually, especially before revising
their first major report. The CI also attends class
and discusses how to prepare and organize written
and oral reports, provides details on the technical
format for written reports, and reviews stylistic
concerns for technical writing. To provide further
instruction on the technical format, the CI dis-
tributes a manual that details information about
technical laboratory report writing. (This manual
is written in the form of a technical report, so
it provides an example of the technical-report
format while also providing information about writ-
ing technical reports.)
Analysis of the Communication Component Sev-
eral attributes have contributed to the success of
the communication component in the laboratory
course. First, effective communication is approached
as an integral and important part of the course. For
some students, simply knowing that they will re-
ceive a communication grade encourages them to
spend more time and effort in writing their reports
and preparing their presentation.
Second, the course stresses that writing is a pro-
cess rather than just a finished product to be evalu-
ated. This approach is particularly emphasized
through the revision exercise. After receiving sub-
stantial commentary from both the CI and the TA,
the students are given two weeks to revise the re-
port and resubmit it for technical and communica-
tion grades. The revision exercise gives the students
a chance to apply feedback from written comments
on their first draft, to recognize positive changes in
their writing, and to learn how to improve their
writing. In fact, in evaluating the revision process
after the spring semester in 1992, ninety percent of
the students surveyed noticed improvement in their
second draft, half of them felt that their second draft
showed "a lot" of improvement, and eighty percent
indicated that the revision exercise helped them un-
derstand how to strengthen their writing.
An interesting student response to the revisions
has been improved original drafts of the first major
report. Many students are spending more time edit-
190


ing their first draft in an attempt to avoid signifi-
cant revision. Ironically, these students are, in ef-
fect, completing multiple revision exercises. We find
it encouraging to see students approach writing as a
sequence of composing and editing stages since the
editing and revision skills they are developing will
undoubtedly prove useful to them in the future.
To incorporate further revision opportunities and
to encourage individual contact with the CI during
the writing process, we intend to hold office hours
during the lab in a newly renovated Instructional
Computer Lab (ICL) adjacent to the Unit Opera-
tions Laboratory. With lab sections scheduled for
five hours and with, at most, twelve students per
section, we can allot a portion of class time to writ-
ing conferences in the ICL, while the remainder of
class time is spent taking data in the Unit Opera-
tions Lab.

Design Course
The communication component of the design course
complements the communication work in the labo-
ratory course. In particular, the design course places
more emphasis on oral work and provides experi-
ence in collaborative writing.
Course Description Table 1 also describes the
requirements for the design course. The students
collaborate on the written and oral reports and re-
ceive group grades on all work completed. The writ-
ten and oral reports receive both a technical and a
communication grade, with the communication grade
comprising thirty percent of the group grade for
each report and presentation. The CI also provides
instruction on writing and presentation techniques,
distributes a manual on the technical format for
design reports, and shows a video of an oral presen-
tation from a previous semester. The first set of oral
presentations are videotaped, and the students are
required to view and evaluate their own presenta-
tions. When the class is large (fifty students), we
divide the students into two rooms for the oral pre-
sentations, to save class time. The professor then
grades one group for technical quality and commu-
nication, and the CI and the TA evaluate the second
group. In addition, the students evaluate each other
during the oral presentations, using a peer-evalua-
tion form. Table 2 further outlines the required re-
sources for the design course.
Analysis of the Communication Component Sev-
eral aspects of the communication work in the de-
sign course have proven successful. First, we have
found it useful to provide samples of both written
and oral assignments at the beginning of the
Chemical Engineering Education










course. The sample-report manuals, similar to the
ones used in the lab course, are useful in helping
students organize their reports. Over half of the
students who seek individual help from the CI ask
questions about organizing their report, and (espe-
cially in the design course) students often are un-
sure of what information to include and where to
put it. Therefore, providing a standard to follow as a
guideline has proven beneficial. Providing examples
of professional design reports would also be useful,
although we have not distributed such samples in
the course thus far.
Similarly, the video shown at the beginning of the
semester offers a more tangible guideline than can
be explained in a lecture. This sample is not a
"perfect" presentation (if one exists), but it demon-
strates some good techniques to follow and some
blunders to avoid when speaking. The video is shown
after a lecture on preparing and organizing talks,
and a discussion of the videotape follows its presen-
tation. The video illustrates points introduced in the
lecture (thereby setting a standard for class presen-
tations), and the discussion gives students a lesson
in peer evaluation. Throughout the remainder of
the semester the students evaluate their classmates'
oral presentations, using the form illustrated in
Table 3. We purposely designed this form with

TABLE 3
Oral Presentation Evaluation Form
Speaker
Place an X in the blank that represents your assessment of the categi
comments to explain your assessment. Then circle a number to repres
for the speaker. Write general comments at the bottom.

Weak
1. Technical Content-- --- --
Relevance, clarity,
technical competence
COMMENTS
2. Planning ------- ---
Organization,
transitions, continuity
COMMENTS
3. Speaker's Manner -- --- -
Voice, eye contact,
gestures, confidence
COMMENTS
4. Visual Aids ---- -----
Visibility, simplicity
appropriateness
COMMENTS

Overall Rating (Circle) 1 2 3 4 5 6
Comments:


Summer 1993


only a few questions in order to give students more
time to write comments. Even though peer evalua-
tion is not a component of the grade, students have
taken it seriously and have offered each other
many helpful suggestions.
Videotaping the talks and requiring students to
view and evaluate the videos has also been a suc-
cessful exercise. For most students, especially those
who have never seen themselves speak before, it is
quite an eye-opener. Actually seeing their own diffi-
culties and successes in speaking helps them iden-
tify areas which need improvement and gives them
more confidence in their abilities. The students gen-
erally dislike watching themselves, but in the end
admit that it was useful to them. Certainly, self-
evaluation helps students prepare for their upcom-
ing presentations; during the semesters when talks
were videotaped and evaluated, the subsequent pre-
sentations showed great improvement.
Finally, the design course provides good experi-
ence in collaborative communication. Collaborative
work, whether written or oral, has become increas-
ingly common in the engineering professions. In the
workplace, collaboration may involve working with
others on research, or actually composing with oth-
ers, or having others review and edit an already
composed work, and the extent of collaboration var-
ies according to the job and
the specific group of people
working together.15s Thus, re-
quiring students to collabo-
ory listed. Add some rate on several stages of a
ent an overall rating large project-from planning
to analysis and presenta-
Strong tion-helps to develop essen-
------- --- tial organization, relational,
and communication skills. A
recent article on small-group
interaction during writing
projects noted that a well-
written report "represents the
team's successful working
through of both small group
and writing problems."[6]
Furthermore, collaboration
teaches peer review. To aid
the students, the CI discusses
what to look for when editing
others' writing and distrib-
7 8 9 10 utes a "Checklist for Collabo-
rators."t[] (See Table 4.) Ac-
cording to a questionnaire dis-
tributed after the spring se-
191










mester of 1992, when working in groups the stu-
dents either split up the writing and then edited
and proofread each other's sections, or they com-
posed and revised together. Two-thirds of the re-
spondents mentioned that they "revised," "edited,"
or "corrected" each other's work, indicating that they
participated in peer review.

DISCUSSION OF THE INTEGRATIVE APPROACH
Through our experiences over the past three years
we have discovered other successes and difficulties
of integrating communication work into existing en-
gineering courses. Certainly, one distinct attribute
of our program is the opportunity to work with stu-


dents in two separate courses. Since
students take the lab and design
courses in the fall and spring of their
senior year, they have two sequential
semesters of intense communicative
work. By the end of the second semes-
ter the CI has worked with each stu-
dent on at least five written and two
oral reports. To help facilitate connec-
tions between the courses, the CI also
keeps a log of individual difficulties
and progress to help students identify
their specific strengths and weak-
nesses in writing and speaking.
Although the CI is integral to our
program, we realize that hiring a per-
son without a chemical engineering
background creates too sharp a dis-
tinction between the technical and
communicative elements of written
and oral work. In reality, a well-writ-
ten report or speech must be both tech-
nically correct and well composed; the
two aspects cannot be separated. To a
large extent, the existing division is
lessened by the interactions between
the CI, the professor, and the TA. To
achieve successful results, the com-
munication work must be approached
as an integral part of the course ma-
terial. The professor must emphasize
the importance of the writing and
speaking assignments, not only when
designing the course but also when
addressing the students. Likewise,
since the TAs grade the reports, they
also must keep in close contact with
the CI to help maintain consistency.
When the communication component
is given adequate value and acknowl-
192


most of the


edgment, we have found that the artificial division
between communicative and technical elements can
actually help the students recognize that outstand-
ing technical knowledge means little if they cannot
effectively communicate their knowledge.
A second difficulty in hiring a non-technical in-
structor is that some understanding of the material
is necessary for a complete reading. Fortunately,
since basically the same material is covered each
semester, the CI can become familiar with it over
the course of several semesters.
One benefit of involving a non-technical instructor
is the opportunity it provides for students to com-
municate with someone who does not share their


TABLE 4
Checklist for Collaborators
Collaborative writing requires that you read and edit your peers' writing.
Therefore, the following checklist is provided to help you identify areas which
could be improved and revised in others' writing and in your own. Remember
that revising takes time, and plan accordingly.
1. Check the overall organization of the draft.
D Is the content presented in appropriate places? (e.g., A discussion of results does not
belong in the introduction.)
D Are the points sequenced logically?
[ Is enough information included for complete comprehension? (Could the writer
delete some information?)
D Does the report live up to its promises (from the abstract/introduction)?
D Does the writer avoid unnecessary repetitions?
O Are there logical transitions between the paragraphs/sections?
D Do the headings/subheadings help articulate the structure of the text? (Could the
report use some subheadings?)
2. Check the paragraphs.
O Do the paragraphs keep to one central idea?
E Do the paragraphs reflect a continuity of logic?
O Does the writer avoid contradictions within a paragraph?
D Are the paragraphs an appropriate length?
3. Check for style. Revise to make the language clear and direct.
[ Does the writer follow these principles for clear writing? (These are principles, not
rules; apply them judiciously.)
Keep sentences short and to the point
Vary the sentence length
Use simple words
Avoid indirect expressions
Use familiar words Avoid jargon Define terms
Avoid unnecessary words
Write to express, not to impress
D Does the writer follow these guidelines for using vigorous verbs?
Use as many active verbs as possible
Avoid nominalized verbs, or verbs trapped inside a noun
Look for words ending in -ion, -ment, -ing, -al which could be made into an
active verb
Try to change sentences which use wordy verb constructions, such as there,
this, it, these, combined with forms of the verb "to be."
Ask if the verb should be past or present. Generally, describe work done in
the past tense, and state principles and conclusions in the present tense.
4. Check grammar, spelling, and punctuation.
D Are the grammar, spelling, and punctuation correct to the best of your knowledge?

Chemical Engineering Education









technical background. Although the oral and writ-
ten reports are addressed to a technical audience,
when working individually with the CI the students
must express technical ideas to a non-technical au-
dience. This actually helps to develop a better un-
derstanding of the material and is a challenging
communicative exercise in itself.
Finally, we recognize that integrating communi-
cation training into existing courses does not allow
for as much instruction as could be offered in a
separate communication course. There is not enough
time to require helpful reading materials on speak-
ing and writing, or to evaluate and discuss pub-
lished articles, or to offer workshops on writing and
speaking. Many students would benefit from more
intense instruction-particularly on technical writ-
ing. But, acknowledging that good communication
skills are never "learned" once and for all, we feel
that by providing some limited instruction and sig-
nificant practice and evaluation, we are at least help-
ing students to improve their skills. As one student
remarked, his writing improved partly "because [he
was] actually writing for a change." An integrative
approach is certainly a step in the right direction.
We also still encourage students to take communi-
cation courses outside the department and to use
campus resources such as the "Writer's Workshop,"
a writing tutorial center sponsored by the Center
for Writing Studies.
As we work to provide our students with better
communication skills, we must remember that de-
veloping expertise in writing and speaking is a life-
long process. Integrating communication training
into existing chemical engineering courses may not
be extensive enough for some students, but it does
provide a significant amount of practice in both
speaking and writing, leaving students with some
professional experience and, hopefully, with an
awareness of the value of communication.
ACKNOWLEDGMENT
The communication work reported in this article
was developed by a number of individuals in addi-
tion to the authors: Dr. Charles A. Eckert, who initi-
ated our emphasis on communication in collabora-
tion with Marsha Bryant and Wayne Howell; Drs.
Edward W. Funk, Thomas J. Hanratty, Douglas A.
Lauffenburger, Richard I. Masel, Mark A. Stadtherr,
K. Dane Wittrup, and Charles F. Zukoski, who all
taught the courses; and Dr. Ruth Yontz, former Com-
munication Instructor.
REFERENCES
1. Caruana, Claudia M., "More ChE Departments Stress Com-
munications Skills," Chem. Eng. Prog., 87, 10 (1991)
Summer 1993


2. Gieselman, Robert D., and Ruth A. Yontz, in "Writing Across
the Curriculum: The Case of Engineering," ASEE Spring
Conf. Papers, Illinois/Indiana Section (1991)
3. Emig, Janet, "Writing as a Mode of Learning," The Web of
Meaning: Essays on Writing, Teaching, and Thinking,
Baynton/Cook, Portsmouth, NH (1983)
4. Griffin, C.W., "Using Writing to Teach Many Disciplines,"
Improving Coll. and Univ. Teaching, 31, 121 (1983)
5. Debs, Mary Beth, "Collaborative Writing in Industry," Tech-
nical Writing: Theory and Practice, Bertie E. Fearing and
W. Keats Sparrow, eds., MLA, New York, 33 (1989)
6. Forman, Janis, and Patricia Katsky, "The Group Report: A
Problem in Small Group or Writing Processes?" J. of Bus.
Comm., 23, 23 (1986)
7. Some of the ideas for this checklist were adapted from
Clark, David G., and Donald E. Zimmerman, The Random
House Guide to Technical and Scientific Communication,
Random House, New York, Chap. 10 (1987) 0


Process Control Lab Course
Continued from page 187.
The last experiment in the second phase is called
"Hardware." In it, the students are required to study
the features of Metrabyte cards such as DAS-8, DAC-
02, DDA-06, PIO-12, and to hard-wire a data acqui-
sition system for monitoring temperature in six poly-
mer reactors with different initiators or different
initiator concentrations. A multiplexer board
(Metrabyte EXP-16) is used to connect the different
thermocouples. The students thus learn about mul-
tiplexers, thermocouples (how the cold junction is
set up on the EXP-16), A/D converters, D/A convert-
ers, electro-pneumatic transducers, and other im-
portant features in data acquisition and digital con-
trol. The reaction is then started, and the students
monitor the temperature change in each reactor si-
multaneously. The students study the effect of chang-
ing sampling rate on data acquisition since six dif-
ferent temperatures are monitored simultaneously.

CONCLUSIONS
These six laboratory experiments are an effective
supplement to classroom lectures. Students gain
hands-on experience in controller tuning, data ac-
quisition, and control. Various process control con-
cepts are emphasized, and the students develop a
thorough understanding of the practical meaning of
the concepts. The laboratory sessions cover almost
all the topics discussed in class except certain ad-
vanced control strategies such as feedforward con-
trol or cascade control. Some of the available com-
puter simulation packages are used to illustrate a
few of these advanced control strategies. Interested
readers may obtain complete information on the
equipment or writeups of the experiment by con-
tacting the author. 0
193










laboratory


EXPERIENCE WITH

A PROCESS SIMULATOR IN A SENIOR

PROCESS CONTROL LABORATORY


SURESH MUNAGALA, DANIEL H. CHEN,
JACK R. HOPPER
Lamar University
Beaumont, TX 77710

When we developed the senior process con-
trol laboratory at Lamar University, we
acquired and installed a process simula-
tor with full-fledged control and instrumentation.
The process control laboratory has been a special-
topic course in the undergraduate chemical engi-
neering curriculum since the spring of 1992, and
future plans call for a regular lab course to be taught
along with it. A brief review of the development of
the laboratory is the subject of this paper.
The process control laboratory was originally
combined with the unit operations laboratory and
included three analog units for level, flow, and
temperature control in which the students were
exposed to the tuning of 3-mode PID controllers
by the Ziegler-Nichols method. These control units
were designed and installed by Scallon Control,
Inc., and the hardware was donated by Fisher
Controls Company.
The need for developing computer-assisted labs
became clear during the late 1980s.111 As a first
step, we developed a microcomputer-based pH con-
trol experiment in which the pH (process variable)
of a given sample of water is controlled by adaptive


Figure 1. Main Menu

control actions (variable gain). The process control
laboratory currently also includes a PID tutorial
program[21 and a video session for control valve se-
lection and sizing."3'

PROCESS SIMULATOR
Industry needs engineering graduates who have
a good understanding of plant practices in addition
to their command of engineering fundamentals. Pro-
cess simulators help students gain those insights by
giving them hands-on experience with plant-wide
process control. The Atlantic simulator was chosen
primarily because it is PC based and we wanted
to be able to use existing personal computers as


Chemical Engineering Education


_ __Jic
DY-Tlt
A nlyl
r. .c 8


Daniel H. Chen is Associ- Suresh Munagala is a Jack R. Hopper is Profes-
ate Professor of Chemical graduate assistant in the sor and Chairman of the
Engineering at Lamar Uni- chemical engineering de- Chemical Engineering De-
versity, where his duties in- apartment at Lamar Univer- apartment at Lamar Univer-
clude teaching and research sity. He received his BS in sity. He has had over thirty
in the areas of sulfur recov- chemical engineering from years of industrial and
ery, air pollution, process Andhra University and be- teaching experience, holds
control, and thermodynam- fore coming to Lamar in four US patents, and has
ics. He received his PhD in June of 1991 he worked as over fifty publications in a
chemical engineering from a project engineer in India variety of fields. Current ar-
Oklahoma State University for four years and with S.K. eas of interest are waste
in 1981 and is a registered Murthy Consulting Engi- management and min-
professional engineer. neers for three years. imization.
Copyright ChE Division ofASEE 1993


tlantlc simula.uon


GRAPHIC: M)enu. N)ext. L)ast., ). 9)










The process control laboratory has been a special-topic course in the
undergraduate chemical engineering curriculum since the spring of 1992, and future
plans call for a regular lab course to be taught along with it. A brief review of the development of
the laboratory is the subject of this paper.

Atlantic Simulation 0:18 workstations (for obvious eco-
2.7, MSCFH nomic reasons).
CON00 CO SER0 The newly acquired Atlantic
process simulator simulates a
280. MSCFH 55.0 GPM typical distillation (depentanizer)
12-- 02 GPM CO process with the equipment and
--- .14.o control structure similar to that
148.0 GPM REFLUX DRUM \ in industrial plants. The column
-.--- P D--120 50 (seven trays) separates a binary
7.38 PSI 80.3 F liquid feed mixture containing
FEED P 2.318 PSI 3 60% pentane and 40% hexane by
P-100,P-100S REFLUX PUP fractional distillation. Figure 1
SS88.1 GPM shows the main menu from which
-- 154.0 F ORHEADUCT the desired graphic (schematic
50.0 [326.76 PSIG diagram) can be selected by en-
REBOILER 9.63 MLBIH
E-- I0E-.6 M tering the appropriate "graphic"
I STEAM number. The simulator graphics
7.8 WT 57.8 G BOTTS CONDENSATE 1 through 4 are schematic flow
BOTTOMS thu
PRODUCT diagrams, and 5 through 8 are
BOTTOMS PUMP WATER PUMP process and instrumentation dia-
P-11o.0P-los P-O3.P-13os0 grams (see Figures 2 and 3).
GRAPHIC: M)enu, N)ext, L)ast, 1)..9) Simulator graphic 9 is the Dy-
Simulator graphic 9 is the Dy-
Figure 2. Overall schematic flow diagram namic Profitability Analysis
(shown in Figure 4) which sum-
Atlantic Simulation 0:18 marizes the cost aspects and
keeps updating the profit/loss be-
OVERVIEW DISTILLATION OVERHEAD VE ing made at that particular time.
OVEVE COLUMN CONDENSER
C-100 E_-120 ___ This helps the students gain in-
-- r0 -- sight into how process distur-
S'bances can affect plant profitabil-
ATER- -- ity and highlights the necessity
S- -- of bringing the process back to
":= -- REDLUXRUM 1 normal efficiency.
S- Figure 5 shows the instrument
FEED -- --- group screen. The set point of any
P100.100S REFLUX PUMP instrument can be changed from
L OI _- --- P2PO:S-S o this screen by selecting the re-
U--- L C TO O
.... ...... .g quired loop, and the control valve
I1 1 9 mode can be changed from auto-
E-- RBOILER matic to manual in order to alter
S--10 STEAM the manipulated variable. The
CONDENSATE trend of any instrument can be
PR0DUT seen from the group trend
screen-the simulator plots the
BOTTOMS PUMP WATER PUMP
P -no,-nos P-130,P-130S trends of any four instruments at
GRAPHIC: M)enu, N)ext, L)ast, 1)..9) a time, taking either five samples
Figure 3. Process and instrumentation diagram per minute or one sample per
Summer 1993 195










minute on a time scale from -12 minutes to 0 min-
utes. The trends are plotted vertically instead of
horizontally. Figure 6 shows the trend of the follow-
ing instruments when the tower feed pump fails
and actions are taken to rectify the disturbance:
1. FI-122 Top product flow indicator
2. FIC-100 Feed flow control
valve Atlantic Simulati
3. FI-110 Bottom product flow
indicator D Y N A M
4. FI-101 Reboiler steam flow
indicator DEBITS

The simulator runs on two IBM COLUMN FEED
VENTED MATERIAL
386 workstations. The hardware REBOILER STEAM
for each workstation consists of COOLING WATER
TIME
OFF-SPEC BOTTOMS
* A fully configured IBM 386 with OFF-SPEC OVERHE
Microsoft DOS 3.2
Floating point 80387 coprocessor CREDITS
640K RAM BOTTOMS PRODUCT
One serial port OVERHEAD PRODUCT
OFF-SPEC BOTTOMS
One parallel printer port with cable OFF-SPEC OVERHEA
BOT COMP (7.8 WTI
Monochrome card and monitor TOP COMP (93.4 Wi
with cable
EGA card (256K RAM) and moni- TOTALS
tor with cable
Hard disk GRAPHIC: M)enu,
Two floppy diskette drives
One Epson dot matrix printer
Operator TDC 3000 keyboard Atlantic Simula
The software consists of
Control System Emulation soft- GRP-
ware
Instructor Station software -
Pecan Power System Operative En- 60 -
vironment S
Process Model
20 -

COST
The software, including the pv MV
operator's keyboard with cable FIC-100 FI
(per unit) costs approximately GPM GP
$25,000 (the listed price as of Sep-
tember 1991), but considerable sPY
discount can usually be obtained
by educational institutions. The OUT % 5.4
above cost also includes install-
ing the software and training staff SELECT A LOOP
to operate the system. All other
hardware listed above is extra.
196


PROCESS MODEL
The Atlantic distillation model is based on first
principles of physical phenomena (unsteady state
heat and mass balance with thermodynamic
properties). It is capable of providing proper
dynamic response under normal operations, cold


on 0:23

IC PROFITABILITY ANALYSIS


COST USAGE TOTAL USAGE S SPENT TOTAL SPENT

1.000 $/GAL 148.0 GPM 3915 GAL 8880 $/HR 3912 $
67.0 SMSCF 2.77 MSCFH 1.22 MSCF 185 S/HR 82 $
8.00 $/MLB 9.63 MLB/H 4.25 MLB 77.1 S/HR 34 S
0.050 SMGAL 525.0 GPM 13866 GAL 1.575 $/HR 0.694 S
75 $/MIN 45 MIN 26 MIN 0 $/HR 0 $
0.040 $S/DV N/A N/A 0 S/HR 0 $
D 0.030 $/%DV N/A N/A 0 $/HR 0 $

PRICE PRODUCTS TOTAL PRODS S EARNED TOTAL EARNED

1.500 $/GAL 57.8 GPM 1528 GAL 5200 $/HR 2291 8
1.250 S/GAL 88.1 GPM 2331 GAL 6611 $/HR 2912 $
0.020 8/%DV N/A N/A 3 $/HR 1 8
.D 0.035 /DV N/A N/A NA 11 /HR 9 8
%) N/A 7.8 WT % 7.8 WT % N/A N/A
%) N/A 93.5 WT X 93.5 WT % N/A N/A

] N/A N/A N/A 2680 $/HR 1196 $

N)ext, L)ast, 1)..9)

Figure 4. Dynamic profitability analysis


tion


DISTILLATION TOWER


I I


II


PIC-120
PSIG
OVERHEAD


-110 Al-110 FI-122 AI-120 TIC-1DO FI-101
M WT % GPM WT % DEG. F MLB/H
M PROD BTM PROD TOP PROD TOP PROD BOTTOMS STEAM


154.0
7.8 7.5 88.1 93.5 154.0


14.90
9.63 14.90


Figure 5. Instrumenrt group screen
Chemical Engineering Education


0:22










starts, emergency shutdown, normal shutdown, and
plant upsets.
The distillation tower is modeled as approximately
eight equilibrium stages. Each stage has few trays,
and the vapor and liquid leaving each stage are
considered in equilibrium. The dynamic component
balance, heat balance, and total mass balance are
maintained on each stage of the tower simulation.
This assures heat and mass balance at all times in
all modes of operation. Based on the tray data, va-
por pressure curves are generated for all the compo-
nents at various temperatures. The vapor pressures
of components at startup conditions are also en-
tered into the database along with vapor pressures
obtained at design tray temperatures. The stage equi-
librium calculations are performed using Raoult's
law. The activity coefficient calculations are required
if nonideality needs to be considered. The time con-
stants for heat and mass balance equations are based
on the mass hold up on the trays and heat capaci-
ties. The metal heat capacity can be included but is
often ignored to speed up the simulation response
time so that a startup can be exercised within a
reasonable training session.
All the differential equations for the distillation
simulation are solved using the Euler integration
method. Because of the steepness of vapor pressures
at various temperatures, Atlantic has developed a
proprietary subroutine (tray) to solve all the stage
differential equations simultaneously using numeri-
cal methods.


INSTALLATION
The company recommendation was that the com-
puters be totally dedicated to the simulator, but
since the simulator lab is offered only during the
spring semester, we did not feel that exclusive dedi-
cation of two of our computers to the simulator would
be desirable. In order to test the idea of a non-
dedicated system, one computer was loaded with
the simulator only and the other was loaded with
the simulator and additional software. We then
checked the performance of both simulators and
found that both worked the same. Extra precau-
tions should be taken, however, to ensure that stu-
dents using other software check their disks for any
viruses before inserting them into the computer.

LABORATORY SCHEDULE AND EXERCISES
After allotting time for other experiments and as-
signments in the process control lab, we were left
with only five weeks for the simulator training. With
this limited amount of time we had to select a few
"typical" exercises from the manual. After we had
reviewed the manual, practiced, and trained our-
selves, the following program outline was decided
upon:
Week 1 Familiarization with the process, control
philosophy, and keyboard operation
Week 2 Correction of known disturbances
Week 3 Identification and correc-
tion of unknown distur-
bances


Atlantic Simulation
FI-122
GPM
TOP PROD
0.0 GPM 300.0

MIN



5
SAMPLES
PER
MINUTE





-12
MIN


FIC-100 FI-110
GPM GPM
TWR FEED BTM PROD
0.0 GPM 300.0 0.0 GPM 300.0


FI--101
MLB/H
STEAM
0.00 MLB/H


20.00


F)ast, S)low, N)ew trend

Figure 6. Instrument group trend (feed pump failure)
Summer 1993


Week 4 Cold start-up
Week 5 Buffer for any incomplete
work and report submis-
sion
The process control lab was
scheduled once a week. In order to
accommodate fourteen students, we
split them into six groups and each
group was allotted 1 hour and 50
minutes per session.
In the exercise for correction of
known disturbances, the students
practiced the corrective actions to
be taken when some particular dis-
turbance occurs (e.g., when the feed
pump stops, the cooling water block
valve closes, the reboiler tube fouls,
pentane concentration in feed in-
creases, the ambient temperature
changes, etc.).
In the exercise for identification
197









and correction of unknown disturbances, a
disburbance was introduced through the instructor's
console and the students had to identify and correct
the problem. The time taken by the students to
identify the disturbance and to subsequently correct
the process was observed.
After the students became familiar with the "nor-
mal" operations, they were asked to work on the
cold start-up exercise, which gives a step-by-step
procedure to start/commission the column. Finally,
the students had to submit a report on the whole
simulator program.

STUDENT RESPONSE AND PERFORMANCE
Once the students became familiar with the simu-
lator, they showed more interest in the system. In
addition to the exercises assigned to them in the
lab, some of the students worked on almost all of
the equipment failure exercises given in the manual.
On the whole, the students' performance in
identifying and correcting equipment failures was
more than satisfactory. Atlantic Simulation, Inc.,
recommends certain time limits for identifying and
correcting equipment failures: 82% of the students
could identify, and 100% of the students could cor-
rect, the equipment failures within the stipulated
time. Most of the students also successfully com-
pleted the cold start-up of the plant to the normal
operating conditions.
Most of the student's perception of the overall pro-
cess was very good, and they learned a number of
fundamental aspects of plant operation. For example,
when a feed pump fails, after identifying the prob-
lem the student would normally start the spare feed
pump without realizing that the flow control valve
on the pump discharge was wide open as a result of
no feed flow-immediately starting the spare feed
pump could cause an excess flow of feed into the
column, creating more problems. They learned that
the flow control valve must be switched over to
manual mode and closed to about 20% before start-
ing the spare pump-the valve must be manually
opened to obtain the required design flow and then
switched back to the automatic mode. Knowing such
operational aspects definitely helps young engineers
do a better job.
STUDENT FEEDBACK
Since this was the first time the Atlantic Simula-
tor was included in the process control laboratory,
we needed feedback from the students to assess how
useful the simulator had been to them from an
engineer's viewpoint. A questionnaire was prepared
for this purpose. We felt the feedback would also be
198


1 2 3 4 5 6 7 8 9 10
Points


Cold Start-up
Equip. Failures
DPA


Figure 7. Distribution of student evaluations of the
simulator
valuable to us in making further improvements in
the training program.
From an engineer's standpoint, 75% of the stu-
dents felt that the simulator training was "useful,"
and 25% considered it "very useful." Figure 7 pre-
sents the students' point-evaluation of the equip-
ment failure exercises, the cold start-up exercise,
and the dynamic profitability analysis (DPA). It
shows that about 67% of the students gave points
ranging from 6 to 8 (out of a maximum of 10) for the
cold start-up and equipment failure exercises. The
students' opinion of the DPA, however, has not been
as consistent as in the other two cases.
There is only one cascade control loop in the pro-
cess, and all of the students said that the simulator
did not help them to better understand the concept
of cascade control. This could be due to the fact that
the program assumes that the user already has some
knowledge of control concepts and does not include
an explanation. Another reason could be the timing,
i.e., the simulator lab was scheduled right after our
students have gone through the advanced control
scheme lectures.14'
About 60% of the students wanted more time al-
lotted for this program so they could do additional
exercises, including emergency shutdown exercises.
To the question of whether other simulators for pro-
cesses involving reactors, absorption columns, fur-
naces, etc., would help their understanding of the
operation of plants and process control concepts, al-
most all of the students responded "yes."

CONCLUSION
A process simulator was installed and integrated
into the process control laboratory at Lamar Uni-
versity. The simulated process is the distillation of a
C5/C6 feed (depentanizer). Based on student perfor-
Chemical Engineering Education









mance and feedback, the simulator training is
deemed to have been successful. In addition to learn-
ing certain fundamental aspects of plant operations
and plant-wide process control, the simulator was
also useful in emphasizing safety aspects such as
emergency shutdown procedures. For new engineers,
knowledge of operational and safety aspects could
be a real asset when they begin work.
To summarize, the simulator was well received by
the students and was regarded by the instructors as
an effective teaching tool.
ACKNOWLEDGMENT
Financial support from the Amoco Foundation for
the development of the process control laboratory is
gratefully acknowledged. We also acknowledge
Chetan R. Amin, Mike Kroll, and Joe Siebem for
providing us with the details of the process model.
REFERENCES
1. Edgar, T.F., "Process Control Education in the Year 2000,"
Chem. Eng. Ed., 24 (1990)
2. PID Control Tutorial, Instrument Society of America, Re-
search Triangle Park, NC (1986)
3. Control Valves and Actuators: Design, Selection, and Siz-
ing, Instrument Society of America, Research Triangle Park,
NC (1989)
4. Luyben, W.L., Process Modeling, Simulation, and Control
for Chemical Engineers, 2nd ed., McGraw-Hill, New York,
Chap. 8 (1990) 0

REVIEW: Chemical Engineering, Vol. 2
Continued from page 183.
Experience and judgment are evident in the ex-
planations and discussions of the basic science and
industrial usage. A level of comparative knowledge
is offered that is often omitted from other texts in
favor of physics and mathematics. The reasons why
one process is chosen over another in industrial ap-
plications are explained. In a section on membrane
separations of biological materials, a philosophy is
suggested for selecting a process: follow the way in
which nature has solved the problem. For example,
even though dialysis is a slow process unsuited for
large-scale industrial separations, its gentle treat-
ment of blood is appropriate for hemodialysis. The
discussion of ion exchange delves into the polymer
chemistry of the cationic and anionic resins that
facilitate the range of applications of this important
unit operation. Motivation is provided for the un-
derstanding of drying as a process following evapo-
ration, filtration, or crystallization, to improve han-
dling and reduce transportation costs. A brief de-
scription of fluidized-bed catalytic cracking explains
the essential features of this outstanding achieve-
ment of chemical engineering. Insightful explana-
Summer 1993


tions such as these are one reason why this re-
viewer will open this book before some other engi-
neering handbook when seeking background infor-
mation on a separation technique.
The topics covered include chapters on: Particu-
late Solids; Size Reduction of Solids; Motion of Par-
ticles in a Fluid; Flow of Fluids through Granular
Beds and Packed Columns; Sedimentation; Fluidi-
zation; Filtration; Gas Cleaning; Centrifugal Sepa-
rations; Leaching; Distillation; Absorption of Gases;
Liquid-Liquid Extraction; Evaporation; Crystalliza-
tion; Drying; Adsorption; Ion Exchange; Chromato-
graphic Separations; Membrane Separation Pro-
cesses.
To illustrate the depth of treatment, consider the
chapter on sedimentation. Sections fully describe
topics on terminal velocity, height of suspension,
shape and diameter of vessel, effects of suspension
concentration, Kynch theory, flocculation, settling
of coarse particles, and analysis of a continuous thick-
ener. A separate chapter deals with centrifugal sepa-
rations, including centrifugal pressure and shape of
the liquid surface, separation of immiscible liquids,
sedimentation, filtration, mechanical design, and
equipment descriptions. The chapter on adsorption
treats the nature and structure of adsorbents, ad-
sorption equilibria (including mathematics of
Langmuir, BET, Gibbs isotherms, and Polanyi po-
tential theory), kinetics, equipment, and regenera-
tion (including thermal and pressure swing, para-
metric pumping, and cycling-zone adsorption). The
exposition of these topics is clear and balanced.
To summarize: this book is a useful and usable
contribution to the chemical engineering literature,
welcome as an introductory text or as a general
reference on separation and particle processes. 0

ei= book review

PLASTICS RECYCLING: PRODUCTS
AND PROCESSES
Edited by R.J. Ehrig
Oxford University Press, 200 Madison Ave., New
York, NY 10016; $64 (cloth), (1992)
Reviewed by
Charles Beatty
University of Florida

This is an excellent primer on the products and
processes used in the early phase of plastics recy-
cling. It covers the commodity plastics that are avail-
able for recycling in reasonable volumes. For this
Continued on page 219.
199










15 classroom


GRAND WORDS,

BUT SO HARD TO READ!

Diction and Structure in Student Writing


ALOKE PHATAK, ROBERT R. HUDGINS
University of Waterloo
Waterloo, Ontario, Canada N2L 3G1

What makes student reports so hard to read?
After all, students are a pretty competent
lot, are they not? Can't we assume, there-
fore, that they write well too-that technical compe-
tence and writing ability go hand-in-hand?
Not at all.
The skill with which most students manipulate
differential equations or design PID controllers is
rarely reflected in the way they write about these
things. Although few of them commit the worst faults
(ungrammatical sentences, dangling modifiers, and
heavy reliance on the passive voice), even the best
students often produce muddled prose. Why?
The answer, we think, lies in two chief faults of
student writing: sloppy, imprecise use of words and
phrases and a disregard for the natural sequence of
ideas that the reader expects. In this paper we will
show, using examples of student writing, how even
the most straightforward technical material can be-
come confused, obscure prose when not enough at-
tention is paid to choosing just the right word and to
arranging ideas in a coherent manner.
How can we improve our students' technical
writing? Merely pleading with them to choose and
arrange their words carefully is not enough. We
must first convince them that learning to write
well is not only essential in communicating their
ideas to others, but that it is also fundamental to
the act of learning itself. Morton Denn, former edi-
tor of the AIChE Journal, said,'l "Skill in communi-
cation is closely tied to the way in which an indi-
vidual formulates and approaches problems, and the
failure of schools to emphasize writing has had a
major impact on technical education and profess-


Bob Hudglns holds degrees in chemical engi-
neering from the University of Toronto and
Princeton University. He teaches about stoichi-
ometry, unit operations and reaction engineering,
and studies the periodic operation of catalytic re-
actors.



Aloke Phatak obtained his BASc and MASc in
chemical engineering from the University of Wa-
terloo and is presently working on his ChE doctor-
ate. His thesis topic is applications of multivariate
statistics in chemical engineering, and he also
has a strong interest in technical writing.

ional practice" (our italics). His observation, though
a little disheartening, suggests a tantalizing ques-
tion that is certainly worth mulling over-by mak-
ing our students better writers, can we also make
them better engineers?

THE PURPOSE OF TECHNICAL WRITING

How do Diction and Structure Fit In?
Now, what I want, is Facts .... Facts alone are what are wanted
in life .... Stick to the Facts, sir!"
Thomas Gradgrind
in Charles Dickens' Hard Times
Students often forget that the purpose of technical
writing is not merely to present facts and informa-
tion, but also to communicate them. In other words,
as Gopen and Swan put it,[21 "[it] does not matter
how pleased an author might be to have converted
all the data into sentences and paragraphs; it
matters only whether a large majority of the read-
ing audience accurately perceives what the author
had in mind" (our italics). To communicate clearly,
effectively, and persuasively without misleading the
reader, therefore, the writer must choose words care-
fully and structure ideas logically so that the reader
knows precisely what is meant.


@ Copyright ChE Division ofASEE 1993


Chemical Engineering Education










The skill with which most
students manipulate differential
equations or design PID controllers is rarely
reflected in the way they write about these things
... even the best students often produce
muddled prose.

Our choice of words and phrases, or diction in the
language of grammarians, determines the accuracy
and clarity of our writing. Proper diction might not
seem, at first sight, to present much of a problem in
scientific writing. After all, engineers and scientists
write about concrete things-models, simulations,
controllers, packed-columns, reactors- and how they
work. Finding the right word or term should be
easy, especially in the straightforward writing we
demand of our students. But in technical writing we
also describe, analyze, recommend, argue, and dis-
criminate. To do these thing well and without ambi-
guity, we must be mindful of selecting exactly the
right words. Yet, either consciously, to hide their
ignorance, or unconsciously from sheer sloppiness,
students often choose their words poorly, with the
result that sometimes we don't know what they are
trying to say, or indeed, whether they really under-
stand what they are trying to write about!
Good diction, however, is only one ingredient of
clear prose. We must also strive for coherence when
presenting information or arguing a point. In stu-
dent writing, ideas within a paragraph are often
presented in a haphazard fashion, with no thread to
bind them together. The result, as with poor diction,
is confusion and frustration in the mind of the reader.
No matter how carefully the sentences in a para-
graph are crafted, the ensemble will mean little if
there is no logical connection among its constituent
units. To avoid confusion and to present ideas as
smoothly as possible, the writer must be careful
about where he places information within a single
sentence and within groups of sentences. The ar-
rangement of this material is what is meant by the
structure of prose. In a well-structured paragraph,
the beginning of a sentence looks back to what was
just said, while the information at the end of the
sentence represents new material that the author
wants to introduce. In this way the reader always
knows where he is in the exposition or discussion.
Like Theseus, he always has his hand on the thread
(here, the thread of the argument) and will have
little trouble finding his way about even the most
labyrinthine discourse.
In the following two sections we will look at some


examples of student writing that illustrate what
happens when words are poorly chosen and when
the expectations of the reader about where informa-
tion should appear are not fulfilled. The result is
that even simple, straightforward technical mate-
rial becomes very difficult to follow.

DICTION
Alice had no idea what Latitude was, or Longitude either, but
thought they were nice grand words to say.
Lewis Carroll
Alice's Adventures in Wonderland
Student reports are often full of big, "scientific-
sounding" words, some of which are chosen solely to
impress the reader. Here is just one example:
Step tests were run to determine the boundaries of the prob-
lem statement (i.e., the valve positions at 750C and 850C).
Grand words indeed, but what does "boundaries of
the problem statement" really mean? Were the stu-
dents really interested in characterizing the "prob-
lem statement" itself? Or, if we carry their words to
the extreme, can valve positions have boundaries?
Using pompous expressions or words can lead to
absurd statements like the one above and can con-
fuse the reader. After reading the sentence a couple
of times, we still do not know why the students
carried out step tests. Unfortunately, such inflated,
imprecise prose is common in student writing. Con-
sider the following:
Due to the stochastic nature of the conclusions of this study,
no real difference between the two statistical methods could
be ascertained.
It is clear that the author found no difference (the
word "real" is superfluous here) between the two
methods, but how and why he came to such a con-
clusion remain a mystery. Indeed, if we are to take
the author literally, rational enquiry is of little use
to either him or to us-our conclusions themselves
are subject to the laws of probability!
In the two examples just presented, the students
used pretentious expressions such as "boundaries of
the problem statement," "stochastic nature of the
conclusions," and "ascertained" to project an air of
unassailable authority. In the second example, how-
ever, there is another objective: to hide what the
author thinks is a result for which he will be penal-
ized. Instead of saying the obvious (something like
"the two methods yield the same result"), he feels
compelled to embellish such a simple, straightfor-
ward statement; in the end the effect is more comic
than convincing.
Yet another group of students writes:


Summer 1993










The same tuning constants [that we used in the simulation]
were used with PID control on the actual process. The
response was found to be less than excellent. This
indicates that the simulation is lacking in the heat rejection
department.
The faults in this example are too numerous to
list. We note, however, that here the students have
managed to combine obfuscation with pomposity by
using terms such as "less than excellent" and "heat
rejection department." In addition, the second sen-
tence is particularly confusing. Was the response of
the process (to a step input, set-point change, etc.)
poor, or was the agreement with the simulation poor?
As we stated earlier, poor diction confuses the
reader and leaves him doubting the writer's grasp
of the subject. For example, here is how some stu-
dents described a computer simulation of a stirred-
tank heater:
The system was modeled using two different simulations.
One simulation was based on the Euler equation, while the
second simulation was based on the Runge-Kutta #4 equa-
tion.
Nonsense. First, there may have been two simula-
tions, but there was only one model, one set of dif-
ferential equations. Second, numerical methods of
integration do not form the basis of any simulation.
Third, Euler, Runge, and Kutta wrote many equa-
tions; which ones do the authors mean? Here, the
students are unsure of the meaning of the words
model and simulation; hence, they "model" a system
using a "simulation," and they base their simula-
tions on numerical methods of integration!
In our last example we find the following:
The model is only as good as the system parameters as
identified by the experimental tests. It is assumed that the
process parameters found in the experimental tests are an
accurate representation of the process.
Confusing? What if we replace the word model in
the first sentence with simulation? Although the
sentence is still faulty, we can now begin to under-
stand its general meaning-something like "If the
parameter estimates are unreliable, the simulation
will be too!" In the second sentence, excess verbiage
("an accurate representation of the process") camou-
flages what we think is the authors' real intent: to
say that the parameter estimates they obtained were
indeed reliable. Here, as in the previous example,
poorly chosen words and phrases leave us wonder-
ing how well the students have grasped certain fun-
damental notions such as the distinction between a
model and a simulation, what parameters and pa-
rameter estimates are, and how engineers "repre-
sent" processes.
In most technical writing we would like to choose
202


the right word to be as precise as possible, not to
satisfy requirements of nuance, balance, rhythm,
and subtlety. Technical terms usually have a single,
precise meaning and cannot always be inter-
changed-if we mean "parameters" we should not
say "parameter estimates," and if we are describing
a "model" we should not use the word "simulation"
in its place. At the same time, however, scientific
prose is more than a mere list of technical words-it
also requires verbs, adjectives, and adverbs. It is
here that students are tempted to use vague, impre-
cise phrases, either out of a desire to obscure the
real meaning or simply out of sloppiness. If students
find that two different methods give the same re-
sults, they are not likely to choose the simplest words
to say so but will write instead that "the two meth-
ods could not be differentiated" or "the two methods
yield approximately the same conclusions." Choos-
ing just the right word is hard work, but it is essen-
tial to do so to say exactly what is meant. Good
diction enforces clarity, accuracy, and honesty in
writing-essential components too of scientific in-
vestigation.

STRUCTURE
[Writers] should, whenever possible, prepare their readers for
new information by beginning their sentences with a "topic," ideas
that are familiar to the audience or that have already been
referred to, and then moving to ... newer, less predictable, less
familiar information
By consistently choosing to arrange information in this way,
writers ... enhance the coherence of their documents ....
J. M. Williamst41
Most technical reports are divided into logical units
called sections. For example, a typical document
might be structured in the following manner: Intro-
duction, Experimental Method, Results and Discus-
sion, and finally, Conclusions and Recommendations.
Not only do most readers expect this structure, but
it also provides a framework in which the writer can
logically present an analysis or argument. Imagine,
for example, trying to read a discussion of results
before any results have been shown!
Just as a report is arranged into logical units,
so too can a sentence be divided, although its
functional divisions are not explicitly labeled.[21 This
way of looking at the structure of prose has been
formalized in a linguistic principle known as func-
tional sentence perspective. In brief, it states that a
sentence should begin with a "topic" idea, infor-
mation that is familiar to the reader, and then move
on to the "stress position," an idea or information
that is less familiar, more complex, and more impor-
tant[4, p. 93]. Organizing a sentence in this way not
only makes the flow of ideas more coherent and less
Chemical Engineering Education










choppy, but it also ensures that the reader under-
stands exactly what the writer is trying to empha-
size.
The expected structure of a sentence can be por-
trayed very simply as

Expected Sentence Structure
Topic position -> verb stress position
(refers to old information) (new information)

The first part of the sentence (the topic) refers to a
particular subject and looks backward to ideas that
have already been presented, usually in the pre-
ceding sentence. New information is then located
toward the end of the sentence (the stress position).
This repeated overlapping of the new informa-
tion in one sentence by the topic of the next s
suggests, we think, a particularly apt metaphor-
the laying of shingles. By "shingling" his sentences
in this way the writer can lead the reader from
start to finish, from premise to conclusion, in a me-
thodical manner.
What happens when we violate this principle? Take
a look at the following example of student writing in
which the authors paid little attention to the smooth
flow of ideas.
The chemical engineer is often faced with the problem of
analyzing the relationship between two large sets ofprocess
data. In the past, multivariate statistical methods have pri-
marily been applied in the social sciences. Due to the large
amount of data generated by industrial processes, chemical
engineers need these types of statistical tools. The purpose
of this report is to investigate two different methods...
Each sentence above makes sense when read by
itself. Strung together in the way they are, however,
means that we have to read the passage several
times before we can understand what the authors
are trying to say: that for certain types of statistical
analyses, chemical engineers need tools that, until
now, have been used mainly in the social sciences.
Why is the passage difficult to follow and to un-
derstand? The first sentence sets up certain expec-
tations in the reader's mind about what is being
discussed-the analysis of large sets of process data.
In the second sentence, however, we are confronted
with new information ("multivariate methods in the
social sciences") that has no connection to what we
have just read. At the end of the second sentence we
move on to the third with the term "social sciences"
fresh in our minds-but again, we encounter a topic
("data generated by industrial processes") that has
nothing to do with what we have just read. By vio-
lating the reader's expectations of what he expects
to read at each step and by beginning each sentence
Summer 1993


with a topic that does not refer to old information,
the authors have written a passage that has no
focus.
In the second example, another group of students
writes:
Over a period of time, weak acids, sodium mercaptides and
sodium sulphides accumulate in the prewash caustic, requir-
ing the spent caustic to be replaced periodically. A strong
odour in the spent prewash caustic indicates that the process
is running inefficiently. In the #1 plant, when the caustic
needs to be changed, the column is completely drained of
spent caustic and replaced with fresh caustic. When the
caustic is dumped it is sent to a spent caustic storage tank.
Here, as in the first example, we can understand
the individual sentences, but we have no sense
that anything ties them together. The reader is
confronted in the topic position of each sentence
with completely new information. Thus, going from
start to finish occurs in a series of jerky move-
ments, and we are not sure just what the writers
are trying to emphasize.
The above discussion of expected sentence struc-
ture presents a much simplified picture. For ex-
ample, the topic may refer to an idea farther back
than the preceding sentence. Furthermore, the size
of the stress position can vary quite a bit. In some
sentences it may be as short as a single word, while
in others it may extend over several lines.[21 Never-
theless, if the writer follows the simple paradigm
pictured above within single sentences and within
groups of sentences, the reader will be able to fol-
low, with little effort, the flow of the argument. The
following example illustrates this point:
In suspension polymerization the conversion ofmonomer to
polymer takes place in the aqueous phase. At the end of the
reaction, the slurry contains not only polymer and monomer
but also emulsifier and other water-soluble impurities. Be-
cause these impurities affect the quality of the final product,
they must be removed from the polymer. Thus, the method of
drying the polymer is ofprime importance.
The first sentence introduces the general subject
(suspension polymerization) which, as the author
informs us, takes place in the aqueous phase. The
emphasis on water is followed by putting the word
"slurry" in the topic position of the second sentence.
The focus then shifts, in the stress position, to wa-
ter-soluble impurities, and these reappear in the
topic position of the third sentence. As the third
sentence unfolds with a dependent clause ("Because
these impurities ."), we begin to sense that some-
thing important is coming up, and by structuring
the sentence in this manner the author makes it
clear to the reader that it is important to remove
Continued on page 209.
203










rem. classroom
- II--------------


SAFETY AND WRITING

Do They Mix?


ROBERT M. YBARRA
University ofMissouri-Rolla
Rolla, MO 65401-0249


he chemical engineering profession has long
voiced a concern that engineers often gradu-
ate with inadequate training in chemical
safety and with less-than-desirable writing skills.
Some educators have reacted strongly to these
concerns and, in response, have developed entire
courses on chemical safety and technical commu-
nications.[21 Pitt"3 has argued the futility of teaching
laboratory safety and suggested the "benefit of
'safety awareness' teaching must be to increase
people's motivation." Educational research has found
writing "a unique mode of learning-not merely
valuable, not merely special, but unique .. higher
cognitive functions, such as analysis and syn-
thesis, seem to develop most fully only with the
support of verbal language-particularly it seems,
of written language."4'" If we can, therefore, uniquely
blend safety and writing into our curriculum, we
create a possible mechanism to motivate our stu-
dents' safety awareness.
This paper highlights how I have blended safety
and writing into my laboratory instruction to im-
prove both safety awareness and written communi-
cation. This experience should offer creative ways
for other engineering educators to effectively and
efficiently integrate safety and written communica-
tion into their own curriculum.

MOTIVATION
We offer a two-course unit operations laboratory
sequence which our majors take in their sixth and
seventh semesters. The laboratory projects in one
course (one credit hour) emphasize momentum and
heat transfer principles, while the second course
(two credit hours) emphasizes mass transfer opera-
tions. Using lectures and laboratory demonstrations
sprinkled throughout these two courses, we intro-
duce the students to the elements of statistical analy-
sis of data, experimental design, and model build-
Copyright ChE Division ofASEE 1993


ing. With these tools, they can then undertake "an
appropriate laboratory experience" that satisfies the
Accreditation Board for Engineering and Technol-
ogy (ABET) curricular content criteria.'51
With regard to safety and written communication,
ABET's criteria clearly state that the engineering
professional must have (the bold-face emphasis is
mine)
* .an understanding of the engineer's responsibility
to protect both occupational and public health and
safety...
The engineering design component must... include a
variety of realistic constraints such as... safety...
Instruction in safety procedures must be an inte-
gral component of the students' laboratory experi-
ence.
Competence in written communication in the En-
glish language is essential ... the development and
enhancement of writing skills must be demonstrated
through student work in engineering courses.
The unit operations laboratory serves as a natural
environment to meet those criteria. If we add to the
laboratory other ABET criteria, such as design con-
tent, open-ended problems, and oral communication,
we either dilute the "hands-on" experience of our
three-credit hour laboratory sequence or transform
it into a course deserving of six or more credit hours.
With the Environmental Protection Agency (EPA)
and Occupational Safety and Health Administra-
tion (OSHA) starting to demand that university labo-
ratories comply with federal regulations regarding
chemical storage, waste disposal, and chemical hy-
giene, the task of meeting all the ABET, EPA, and
Chemical Engineering Education


Robert M. Ybarra is a lecturer in chemical
engineering at the University of Missouri-Rolla.
He received his BSChE from the University of
Califomia, Santa Barbara, and his PhD from
Purdue. His teaching and research interests
are in polymer processing, computer-aided ex-
perimentation, and technical communication.










OSHA requirements becomes nearly intractable. For-
tunately, our department offers a three-week, one-
semester-hour course on chemical laboratory safety
that all students enrolled in Freshman Chemistry
must successfully complete by passing a written ex-
amination. This passive method of safety instruc-
tion, however, does not insure compliance with the
federally imposed regulations. Therefore, we sought
to devise a laboratory environment which actively
promotes safety as well as meeting the ABET, EPA,
and OSHA requirements.

APPROACH
The problem of cramming more material into a
limited curriculum needed addressing. My solution
was to merge two seemingly unrelated topics-safety
and technical writing. I patterned this integration
of safety and written communication after a similar
structure at Dupont's Seaford Nylon Plant, where I
had previously worked. At Seaford, the safety pro-
gram actively involved everyone from the technical
superintendent right down to the clerk typists.
Stressing personal responsibility for a safe work-
place seemed to instill a strong sense of safety aware-
ness in the participants. Since safety audits served
as a good means to actively involve the people at
Dupont, I decided such an activity could work equally
well with my unit operations laboratory students.
All successful businesses require frequent and con-
cise communication between their operating units.
The company memorandum is the principal mode of
written communication because it promotes a rapid
exchange of clearly and concisely written informa-
tion. Similarly, I adopted the memo as the principal
way for our students to communicate within the
laboratory. It allows me to quickly cut to the es-
sence of the students' work without spending hours
reading comprehensive reports. Others have used
memos and other short written communication tech-
niques for similar reasons.12''81

COURSE SETTING
Our blending of safety and technical writing oc-
curs in the mass transfer operations course, which
meets weekly for a one-hour common lecture and a
five-hour laboratory session for the individual
sections. Before the first laboratory session, I ran-
domly assign the students to groups and projects.
An individual student group has responsibility for
only one project during the entire semester, and the
group periodically issues written and oral reports to
summarize its progress. To expose our students to
the other laboratory projects, the groups having pri-
mary responsibility for each project plan short ex-
Summer 1993


periments for the other groups to perform during
scheduled visitations to their projects. Such a labo-
ratory structure openly promotes active communi-
cation between groups.

LECTURE AND LABORATORY ACTIVITIES
In this section I outline the specific activities I
have successfully used to teach safety with written
communication in our undergraduate laboratories.

Safety Audit Team Reports
During the second laboratory session I form
two-person safety-audit teams, issue a semester
schedule for the teams, briefly discuss what safety
items the teams should look for, and send the first
team off to inspect the laboratory (see Table 1). Some
of the items to look for include properly operating
safety shower and fume hood, clear access to exits,
properly labeled chemical containers, frayed elec-
trical cords, and water spills. They are also asked
to correct any unsafe situation and report their
findings in a memo.
The team must complete their inspection before
the other laboratory groups can begin working, a

TABLE 1
Safety Audit Team Checklist
Student-designed document used as a checklist for the safety audit
teams. The results of the audit are then summarized in a memo and
filed in a "red notebook." The procedure insures compliance with
EPA and OSHA regulations.

Safety Audit Team Check List
Unit Operations Lab
Room 110
OK Unsafe Item Item Description
1 Safety shower operates properly
2 Safety shower is easily accessible
3 Fire alarms and extinguisher are accessible
4 Chemical containers properly labeled
5 Walkways are free of clutter
6 Floors are clear of water puddles
7 Lab is neat and in good working order
Comments:


Date:


Auditors:


I -_ _










process that takes about fifteen minutes. The
very first audit uncovered a particularly dan-
gerous situation-a safety shower that only a
person over six feet tall could reach!
The students file these safety audit reports in
a "red notebook" which I periodically review
but do not grade. Should any safety item re-
quire immediate attention that the team can-
not handle, they are instructed to personally
contact the proper university personnel and
to document the conversation in the audit
report by also issuing a memo to the person
they contacted.
After the first round of audits, I noticed that
one student had composed a series of audit
checklists for the five laboratory areas the teams
had to check. Since these checklists greatly im-
proved the team's efficiency, they were used in
all subsequent audits. (Table 1 gives the check-
list for one of the areas.)
This activity offers an excellent way to prac-
tice writing, to comply with EPA and OSHA
regulations, and to encourage student owner-
ship in creating a safe workplace.

Equipment Safety Analysis
After some brief introductory comments about
the semester project, I discuss the laboratory
section that concerns performing an equipment
safety analysis. Bethea's NIOSH Instruction
Module Units V and IX19` serve as a guideline
for the discussion. In their analysis, I ask the
groups to include a sketch of the floor plan of
the laboratory area in which they work, to list
chemicals used and the proper disposal of waste
chemicals, to review MSDS's for toxicity, flam-
mability and incompatibility, and to identify all
electrical, mechanical, and tripping hazards.
Each group then summarizes their analysis in
a memo. In the following laboratory period, the
graduate teaching assistant and I orally review
this ungraded memo with the groups.

Writing Workshops
During the common lecture period I run a
series of workshops on technical writing in
weeks two through four.
Agents of Wordiness Handout The ma-
terial I present on technical writing draws
heavily from a short course I took in 1981 when
I worked for Dupont, called the "Burger Course
in Effective Writing."[10 Burger identified thirty-
nine agents that contribute to wordiness, with
the rankings indicating how frequently the
206


agents occur (see Table 2). I distribute a handout that
defines and gives examples of these agents, which I briefly
review in the first lecture.
To warm students up to Burger's method, I start with a
discussion of the number-one agent, "verb mutilation" and
ask them to find the key verb thought in the first sentence
in Table 3. This sentence mutilates the verb thought "to
recommend" by turning it into a noun. The second sentence

TABLE 2
Burger's Agents of Wordiness

The following list Burger's compilation of agents that contribute to wordiness. Our
discussion is limited to the first 29 agents since they occur more frequently than the
last ten. We also suggest ways to eliminate them from the students' written commu-
nications.


Overpoweringly Important
1. Verb mutilation
2. Saying what goes without saying
3. Disregard of common elements
4. Overuse of the passive
5. The zero word

Very Important
6. Prepositionitis
7. The irrelevance
8. The wrong point of view
9. Failure to use second-time words
10. The trivium
11. Fractional anticipation
12. Zigzagging
13. The pointless modifying clause
14. The pointless third-level modifier
15. The impersonal introduction
16. The wrong number
17. The unnecessarily difficult verb
18. The club-member phrase
19. Pointless repetition
20. The long-winded negative


21. Modifier mutilation
22. Pointless attribution
23. Repetition plus

Important
24. The Misattached modifier
25. The bangbang paraphrase
26. The name substitute
27. Noun mutilation
28. The wrong "each"-type word
29. Failure to use prepositions

Unimportant
30. Failure to use indirect objects
31. "If" first
32. Name first
33. Preposition first
34. "The" first
35. Failure to use summary words
36. The long-winded affirmative
37. Failure to use the possessive
38. Overuse of the possessive
39. Failure to use the passive


Chemical Engineering Education


TABLE 3
Examples of Verb Mutilation and Other Agents of Wordiness Used
in Writing Workshop

The boldface-type words designate the problem areas in the sentence.
Eliminating these agents leads to a clear and concise sentence about half
as long as the original.

1. My recommendation for the new system is that we replace the fouled heat
exchanger tubes.
Agents of Wordiness: Verb mutilation
2. The replacement of the fouled heat exchanger tube is recommended.
Agents of Wordiness: Verb mutilation; Overuse of the passive
3. It is recommended that we replace ...
Agents of Wordiness: Impersonal Introduction; Pointless third-level modifier
4. I would recommend we replace ...
Agents of Wordiness: Unnecessarily difficult verb (conditional)
5. I recommend we replace ...










TABLE 4
Original Safety Rules Handout

SAFETY
Safety is of the ultimate importance. The key to safety is your awareness
of potentially dangerous situations. In this lab dangers include hazard-
ous and flammable chemicals, moving equipment, and high-pressure
steam.

SAFETY REGULATIONS
1. Goggles will be worn when corrosive chemicals are mixed from
bulk, or dangerous chemical reactions are in progress.
2. Safety glasses are to be worn around moving machinery.
3. Loose ties, shirt cuffs, trouser cuffs, or other floppy cloth pieces
are prohibited around moving machinery parts. Leather shoes and
socks, or approved equal, are required at all times.
4. Cylinders of gas under pressure should be treated with respect. A
dangerous situation is created if the valve portion is cracked from
the cylinder. Gas cylinders should be locked to a solid structure
when in use and when in storage. They should be locked to a cart
when in transit. When not in use, safety valve-cap should be kept
on the cylinder.
5. There will be no horseplay in the laboratory. The possibility of
accident and serious injury is ever present.
6. Each member of the lab is responsible for knowing the location of
1) all fire extinguishers, 2) all safety showers, 3) all exits, and 4)
all first aid supplies.
7. No cola bottles, food of any sort, paper cups, paper towels, or
scratch paper is to be brought into or consumed within the labora-
tory.
8. There will be no smoking within the laboratory.
9. All containers of liquid must contain a label with the following
information: name of material contained, strength or purity if
known, date placed in container, name of person doing the plac-
ing. Any container not labeled is to be emptied, washed and
returned to the storeroom.
10. Keep your work area neater than you found it.


TABLE 5
Safety Rules Rewritten to Reduce Wordiness

SAFETY
Safety is everyone's concern. Awareness of potential dangers is the key
to safety. Laboratory dangers include hazardous chemicals, rotating
equipment, and high-pressure gases.

SAFETY PRACTICES
1. Goggles and protective footwear must be worn in the laboratory.
2. Loose clothing or jewelry are prohibited near rotating equipment.
3. Pressurized gas cylinders must be: securely anchored when in use,
securely anchored and capped when stored, and strapped to a cylin-
der cart and capped when moved.
4. Horseplay is prohibited in the laboratory.
5. Everyone must know the location of all fire extinguishers, safety
showers, exits, and first aid supplies.
6. Smoking, food and drink are prohibited in the laboratory.
7. All liquid containers must be labeled with the following information:
contents, concentration, date, and experimenter.
8. Keep your work area clean.
9. Properly dispose of all chemical waste.

Summer 1993


seems to improve the situation but it actually re-
sults in another mutilated verb as well as use of the
passive voice. I offer the third sentence, but this
choice results in "the impersonal introduction"
(Burger's #15) and "the pointless third-level modi-
fier" (Burger's #14), while the fourth sentence suf-
fers from an "unnecessarily difficult verb" (Burger's
#17). We finally settle on the fifth sentence as an
acceptable choice.
Safety Rules Review When I first came to the
University of Missouri-Rolla, I inherited the set of
Safety Rules for the unit operations laboratory (see
Table 4). As I examined the document, I found it
lush with Burger's Agents of Wordiness and decided
it would provide an excellent platform from which
to discuss safety and technical writing.
Table 5 represents a major revision of the Safety
Rules which resulted in more than a sixty percent
reduction in the number of words. The review pro-
cess uncovered many less frequently occurring
agents, such as "the name substitute" and "noun
mutilation." It also revealed the need to add a rule
about waste disposal. Inspecting these "rules" in
more depth, we find they actually represent safe
"practices" rather than rules. This switch builds a
more proactive attitude about safety.
E-Prime Bourland11' introduced a writing sys-
tem called E-Prime, a name he derived from the
following equation:
E'=E-e
In this equation, E represents standard English and
e represents all forms of the verb "to be." Therefore,
E-Prime English eliminates the verb "to be" from
use. This practice eliminates most of the passive
voice, much of the subjunctive mood, and some par-
ticipial uses. As a further revision to the Safety
Rules, I ask the class to consider rewriting them in
E-Prime. Table 6 gives some examples of the Safety
Rules written exclusively in E-Prime.
Readability Results To quantify the effect of

TABLE 6
Examples of Safety Rules Written in E-Prime

1. Always wear goggles and protective footwear in the laboratory.
2. Do not wear loose clothing or jewelry near rotating equipment.



5. Know the location of the nearest: fire extinguisher, safety shower,
exit, and first-aid supplies.


7. Label all containers with the following information: contents,
concentration, date, and experimenter.










editing the Safety Rules, I assessed the three ver-
sions for readability using Writing Tools Group's
Correct GrammarTM for the Macintosh,[12 O a software
package that checks spelling, style, and grammar.
Correct Grammar and other grammar-checking soft-
ware, such as Reference Software's Gram mat' ik
and Que Software's RightWriter also run in the
DOS environment and check for readability.
Table 7 gives the results of the readability analy-
sis of the Safety Rules. It clearly shows that by
eliminating the major contributors to the wordiness
of the Safety Rules we significantly reduced the num-
ber of both sentences and words and the percent use
of the passive voice. The reduction in the total num-
ber of sentences corresponds to a simple elimination

To answer the question posed in
this paper's title-writing mixes very
well with safety. Our unique blending of the
two has definitely enhanced our students'
safety awareness.

of irrelevant sentences. Two measures of readabil-
ity, Flesch-Kincaidt"3 and Gunning Fog Index,1141
show mixed results between the original and the
revised documents. When we write the Rules strictly
in E-Prime, however, two very interesting results
occur: the passive verb tense disappears and there
is a significant reduction in the educational level
required to read the Safety Rules. The second result
offers the true promise of E-Prime and shows it to
be an economical and understandable mode of writ-
ten communication that reduces fogginess. The
reader cannot afford to misinterpret the intent of
any technical communication that deals with criti-
cal issues such as safety procedures.
Final Examination At semester's end, students
take a comprehensive final examination. I include a
section on writing to assess how well they can iden-
tify and suggest improvements to sentences taken
from published scientific literature. Recent exam re-
sults showed that over 75% of the students could
adequately identify the "agent of wordiness" and
suggest significant improvements.
CONCLUSIONS
We have created a laboratory environment where
students take an active role in safety. Audit teams
foster a sense of laboratory ownership because the
students assume responsibility for ensuring compli-
ance with EPA and OSHA regulations.
We have also significantly improved our students'
writing skills, as witnessed by a marked improve-
ment in their memos and reports. Using ungraded
208


TABLE 7
Results of Readability Analysis from Correct Grammar"
Safety Document Versions
Quantity Evaluated Original Revised E-Prime
Sentences 22 16 15
Words 298 110 108
Passive Sentences (%) 54 37 0
Flesch-Kincaid 9.0 10.0 8.8
Gunning Fog Index 7.1 6.6 5.3

memos has proved to be an effective and efficient
way to check the students' progress, and they pro-
vide meaningful and timely feedback. The memos
also give students an opportunity to practice writ-
ing by forcing them to continually "distill out" the
important aspects of their work and present the
product in a coherent form.
To answer the question posed in this paper's title-
writing mixes very well with safety. Our unique
blending of the two has definitely enhanced our stu-
dents' safety awareness. In addition, the safety and
writing activities presented in this paper could be
beneficial to any engineering discipline with a large
laboratory safety component, especially if chemicals
are involved.

REFERENCES
1. Carpenter, S.R., R.A. Kolodny, and H.E. Harris, "A Novel
Approach to Chemical Safety Instruction," J. of Chem. Ed.,
68(6), 498 (1991)
2. Amyotte, P., "A Communication Course for Engineers," Eng.
Ed., 81(4), 436 (1991)
3. Pitt, M.J., "Can Laboratory Safety Be Taught?" J. of Chem.
Ed., 65(12), A312 (1988)
4. Emig, J., "Writing as a Mode of Learning," College Comp.
and Commun., 68(2), 122 (1977)
5. Criteria for Accrediting Engineering Programs, 1990-91,
1991 ABET Accreditation Yearbook, Accreditation Board
for Engineering and Technology, New York (1991)
6. Yoxtheimer, T.L., "Utilizing Short Writing Assignments in
Technical Courses," Proc., 1986 Front. in Ed. Conf., ASEE/
IEEE, Arlington, TX, New York (1986)
7. Swanson, L.W., and H.M.R. Aboutorabi, "The Technical
Memorandum: An Effective Way of Developing Technical
Writing Skills," Eng. Ed., 80(4), 479 (1990)
8. Snell, L.M., "Teaching Memo and Letter Writing Techniques
in the Classroom," Eng. Ed., 80(4), 481 (1990)
9. Bethea, R.M., Incorporation of Occupational Safety and
Health into Unit Operations Laboratory Courses, NIOSH
Instructional Module, U.S. Dept. of Health and Human
Services, Cincinnati, OH (1991)
10. Burger, R.S., How to Write So People Can Understand You,
Tinicum Press, West Chester, PA (1970)
11. Bourland, D.D., "A Linguistic Note: Writing in E-Prime,"
General Semantics Bulletin, 32-33, 111 (1965-66)
12. Correct Grammar 3.0-For the Macintosh Manual, Writing
Tools Group, Sausalito, CA (1992)
13. DOD MIL-M-38784B
14. Gunning, R., The Technique of Clear Writing, McGraw-Hill,
New York, NY (1968) J
Chemical Engineering Education










Grand Words
Continued from page 203.
the impurities from the final product. The final sen-
tence, then, which states the subject of the report,
appears as a logical consequence of what has come
before. Indeed, we might say that it occupies the
stress position of the entire paragraph itself.
Although the previous example fits quite neatly
the paradigm outlined above, mere mechanical ap-
plication of the "topic -- verb -> stress" pattern in
each sentence of a paragraph cannot guarantee co-
herence. The author must decide, for example,
whether the reader is capable of making a connec-
tion between the material in the stress position of a
sentence and the topic position of the next. The link
may be obvious to the writer, but if it is not clear to
the reader the thread of the argument may be lost.
In the example cited above, the author is writing for
an audience of chemical engineers; it is taken for
granted that the word "slurry" appearing in the topic
position of the second sentence will be recognized by
most readers as referring back to the term "aqueous
phase" which appeared in the first sentence.
Up to this point we have emphasized only the
author's responsibility to write coherent prose. How-
ever, just as searching for the right word forces a
writer to think hard about what he really wants to
say, so too can thinking about how best to structure
an argument compel the writer to re-examine the
logic, coherence, and clarity of what he is trying to
communicate to the reader. This link between clear
writing and clear thinking is a theme that we take
up in the final section.

CLEAR WRITING AND CLEAR THINKING
Ce sont les mots qui conservent les idges et qui les transmettent, il
en rdsulte qu'on ne peut perfectionner le language sans
perfectionner la science ni la science sans le language. [emphasis
added]
A Lavoisier
Traite dIementaire de chimie
(It is words that preserve and transmit ideas. As a result, we
cannot perfect language without advancing science, neither can
we advance science without perfecting language.)
For most students, writing clear, precise, logical
prose is never an easy task. They look upon report
writing as a loose end to be tied up after the real
(meaning technical) work is done. Consequently, they
give little thought to technical communication. Small
wonder then that student reports are frustrating to
read, that they contain poorly chosen words and
phrases, that ideas are haphazardly thrown down
on paper. Yet, as we stated at the beginning, stu-
dents are technically competent. They can develop
Summer 1993


mathematical models and simulations without know-
ing the distinction between the two terms. Are we
being merely pedantic, therefore, by insisting upon
good diction and coherently written paragraphs? Is
writing ability simply a desirable, but not an essen-
tial, element of the engineer's art? We don't think
so, for two reasons.
First, although student prose may be confusing,
instructors can usually decipher it-but only because
they supplied the topic or problem in the first place
and are probably familiar with it. When students
become practising engineers, however, their audi-
ence may not be so well-acquainted with the sub-
ject. Thus, not only does sloppy writing automati-
cally place their ideas and arguments out of reach,
but it can also jeopardize their careers. An engineer
who writes incomprehensible prose is in danger of
being passed over for promotion in favour of some-
one who can write clearly, logically, and precisely.
Second, clear thinking begets clear writing. In other
words, the better we understand our subject, the
better we will be able to write about it. However,
careful writing can also help us clarify and under-
stand the ideas that we grapple with. Except in
those rare instances when we come to a visceral
understanding of something almost immediately,
most ideas and notions circulate about in our heads
in a vague, half-baked form. Only when we are
obliged to write them down, to explain them, and to
justify them do we really force ourselves to think
deeply and logically about them.
For the engineer or scientist, therefore, writing
serves two purposes: to communicate our ideas to
others, and, perhaps more important, to help us get
those ideas straight in our own minds. Thus, we
must emphasize to our students that learning how
to write clear, precise prose is just as much a part of
their technical education as learning how to solve
differential equations. The ability to write well is an
essential ingredient in developing a logical scientific
argument and, as the epigraph to this section makes
clear, it is therefore fundamental to our craft.

REFERENCES
1. Denn, M., quoted in Rosenzweig, M., "Are ChE Students as
Good as We Were?" Chem. Eng. Prog., 87(2), 7 (1991)
2. Gopen, G.D., and J.A. Swan, "The Science of Scientific Writ-
ing," Amer. Scientist, 78, 550 (1991)
3. Quoted in Woodford, F.P., Scientific Writing for Graduate
Students: A Manual on the Teaching of Scientific Writing,
Rockefeller University Press, New York (1968)
4. Williams, J.M., "How Can Functional Sentence Perspective
Help Technical Writers Compose Readable Documents?" in
Solving Problems in Technical Writing, L. Beene and P.
White, Eds., p. 91, Oxford University Press, New York (1988)
209










outreach


A UNIT ON ACID RAIN

IN A

HIGH SCHOOL OUTREACH PROGRAM


JOHN A. MARSH, MICHAEL A. MATTHEWS
University of Wyoming
Laramie, WY82071-3295

Declining enrollment in engineering pro-
gramst1i has been a cause for concern in the
educational community nationwide. To
counter this downswing, engineering faculty may
have to place greater emphasis on outreach pro-
grams in the future. (Bayles and AguirreE2' described
one such effort in the chemical engineering depart-
ment at the University of Nevada, Reno.)
The Engineering College at the University of
Wyoming has an outreach program aimed at high
school science and math teachers and at high school
seniors-to-be. Since many high school students are
interested in environmental engineering as a ca-
reer, in 1992 the Chemical Engineering Department
contributed a unit on acid rain, with emphasis on
chemical engineering analysis and solutions to this
environmental problem. We will briefly describe the
college outreach program in this article and will
include details on our class dealing with acid rain.
ESP AND HISTEP
College Outreach Programs
The College of Engineering conducted outreach
programs in June and July of 1992, with the help of
financial support from the National Science Foun-
dation. The June session was the Engineering Sum-
mer Program (ESP) for high school seniors-to-be,
and the July session was the High School Teachers
Engineering Program (HISTEP) for teachers of math,
life and earth sciences, and physics. The program
involved faculty from four undergraduate depart-
ments (chemical, mechanical, electrical, and civil and
architectural engineering).
The goal of the ESP program was to introduce
students to several of the engineering disciplines
Copyright ChE Division ofASEE 1993


John A. Marsh has BS degrees in molecular biol-
ogy and chemical engineering from the University
of Wyoming and is presently a master's candidate
in chemical engineering there. His research inter-
est is in developing alternative processes for recy-
cling spent plastics.




Michael A. Matthews received his PhD from
Texas A&M University in 1986. His research and
teaching interests are phase equilibrium thermo-
dynamics, supercritical fluid science and tech-
nology, and separations.

available to them on campus and to allow them to
explore various career paths in engineering. They
were exposed to several real-world issues through
laboratory activities, computational work, and field
trips, as well as through lectures and discussion.
There are many reasons why so few students are
interested in engineering; the answer is not simply
a lack of technical preparation in the secondary
schools.[31 We felt that many school counselors and
teachers do not effectively describe engineering as a
career, so HISTEP was designed to show science
and math teachers what engineers actually do. By
giving them exposure and hands-on experience, we
hoped to equip them to act as effective advocates for
careers in engineering.
The three main interest areas in the ESP and
HISTEP programs were
Environmental Engineering
Computer-Aided Engineering
Materials Engineering
and each interest area consisted of three related
subtopics. For example, in the environmental engi-
neering area, faculty from electrical, civil, and chemi-
cal engineering discussed solar power, biological
treatment, and acid rain. (The various topics for the
Chemical Engineering Education










1992 programs are given in Table 1.) Students chose
two of the three interest areas, thus covering six
topics during the three-week program. A sample stu-
dent schedule is given in Figure 1.
ESP participants were selected from applications
which included high school transcripts, two letters
of recommendation, and a 300-word statement ex-
plaining the student's interest in science and engi-
neering. Thirty students were selected from approxi-
mately sixty-five applications from four states.
A total of twenty-three high school teachers par-
ticipated in HISTEP, and they received continuing
education credit for the program as well as a sti-
pend and room and board. In addition to laboratory
and discussion sessions led by the engineering
faculty, the teachers participated in teaching- and
learning-workshops and developed curriculum units
based on their experiences in HISTEP.

THE ACID RAIN UNIT
Acid rain is normally associated with the north-
eastern United States, or northern Europe, or other
areas with high industrial density, particularly those
areas with power plants that burn high-sulfur coal.
The Rocky Mountain West is not known for acid
rain,t41 although cities such as Denver and Phoenix
have some acid rain and "brown cloud" problems

TABLE 1
Research Topics in the Engineering Summer Program


Interest Area and Topics E
Environmental Engineering
Acid Rain
Solar Power
Biological Perspectives
Computer-Aided Engineering
NMR Image Processing
Digital Electronics
Electronic Materials/Manufacturing
Materials Engineering
Sports Dynamics
Composite Materials
Structural Engineering


engineering Discipline

Chemical
Electrical
Civil

Chemical
Electrical
Electrical

Mechanical
Mechanical
Civil


Time Monday Tuhesdan Wdnesda) Thursda [ rridnay
7:30-8:00 Breakfast
8:30-11:30 LAB LAB Plant Tour LAB LAB
11:30-1:00 Lunch
1:00-4:00 LAB LAB LAB LAB
Plant Tour
4:00-5:30 Recreation Recreation
5:30-6:30 Picnic Dinner
-------- Picnic ---- | -------------
7:30-10:00 Entleainnment Entertainment

Figure 1. Weekly schedule for the 1992 Engineering
Summer Program
Summer 1993


Activities included laboratory
demonstrations and research using a
water-sulfur dioxide scrubber, computer process
simulations of the water-sulfur dioxide scribber,
class discussions, and homework dealing
with cost/risk/benefit analysis of
electrical power production.

familiar to many students. Nevertheless, the topic
is pertinent for students living in this area of the
country. Pedagogically, studying acid rain allowed
us to explore a wide range of integrated industrial
activities (such as mining, transportation, combus-
tion, and flue-gas cleanup) associated with a very
familiar and acceptable product-electricity. Wyo-
ming is the nation's leading coal producer; most of
the coal is low-sulfur and is used by electrical utili-
ties. Some of the participants in the course had
parents or spouses working in coal mining opera-
tions or in power plants, and thus they were already
acquainted with the subject of air quality. Most of
the students were also aware of the 1990 Clean Air
Act which focused attention on electrical utilities
that burn coal. In addition, the Rio de Janeiro "Earth
Summit" was much in the news during the summer
of 1992, and newspaper and other reports were plen-
tiful, stimulating student interest and awareness.
Thus, the participants in our program were highly
motivated to explore some of the technological and
societal issues associated with acid rain.
We used several methods to introduce students to
the history and technology of power production and
to methods of dealing with acid rain. Activities
included laboratory demonstrations and research
using a water-sulfur dioxide scrubber, computer
process simulations of the water-sulfur dioxide
scrubber, class discussions, and homework deal-
ing with cost/risk/benefit analysis of electrical
power production.
Since we wanted our unit to be more than just a
technological treatment of acid rain, we also empha-
sized the many possibilities for work in areas re-
lated to clean-up and preservation of the environ-
ment. Additional information on chemical engineer-
ing and environmental issues was disseminated
through videotapes, plant tours, department tours,
newspaper clippings, and handouts. Students were
asked to consider the benefits of power production
as well as the societal costs and risks, and a good
deal of time was devoted to roundtable discussions
of these issues and the technology involved.
As can be seen from Figure 1, the unit was con-
ducted in four three-hour days (not counting off-
211









campus plant tours). A typical day began with forty-
five to sixty minutes of class discussion, videotapes,
numerical solutions to homework, and question-and-
answer sessions. On the first day of class, each stu-
dent was given a folder of handouts (over fifty pages)
which included details of all laboratory demonstra-
tions, blank data sheets, and calculation details. Also
included were notes on the history of air pollution,
several tables and charts dealing with energy pro-
duction and usage, and questions designed to foster
thought about cost/benefit analysis. A few news-
paper clippings and magazine articles dealing with
environmental issues in general were also included
in this handout.

LABORATORY ACTIVITIES
To illustrate the meaning of acidity, we introduced
and discussed the pH scale on the first day of class,
and to quantify the concept, we asked the students
to measure the pH of several common fluids-tap
and distilled water, soft drinks, vinegar, solutions of
sodium bicarbonate and sodium hydroxide, and soap
solutions. We used three analytical methods to mea-
sure the pH: indicator solution (Fisher Brand Uni-
versal Indicator, pH range of 4 to 12); pH papers;
and a digital pH meter.
We divided the students into smaller groups to
make the pH measurements. As we expected, while
pH measurements using the various techniques were
sometimes in good agreement, at other times they
were not. Also, comparison of measurements made
by using the same analytical method, but by differ-
ent classes at different times, showed some discrep-
ancies. This gave us an opportunity to talk about
concepts such as precision, accuracy, reproducibil-
ity, personal technique, and use of different analyti-
cal methods. This in turn led to consideration of the
all-too-familiar situation where opposing groups in-
volved in an environmental discussion present seem-
ingly conflicting data, analysis, and interpretation.
We took this occasion to emphasize the role of the
engineer in objectively gathering and reporting data.
The pH studies taught students that living things
can tolerate a wide range of pH, but that chemistry,
concentration, dosage, duration, and other factors
are important. The discussion on experimental er-
ror and technique illustrated the difficulties involved
in quantifying the level of acidification at a given
location. In some cases there are little or no histori-
cal pH data, making it difficult to estimate the rate
of acidification. We also cite the possibility of con-
flicting and erroneous measurements, and point out
that all of these factors have contributed to the con-
troversy over the rate, extent, and even the exist-
212


ence of damage due to acid rain.E51
To show how sulfur oxides contribute to acid pre-
cipitation, the students used a Bunsen burner to
ignite a small amount of pure sulfur held on a
spatula. The burning sulfur was then inserted into
the headspace of a flask partially filled with water
and the Fisher indicator. As the sulfur dioxide was
absorbed, there was a rapid color change from green
to bright red, vividly demonstrating the acidifica-
tion of water by sulfur dioxide. Repeating the ex-
periment using a pH meter allowed a more precise
measurement of pH.
We spent most of the laboratory time in operating
a water/SO2 scrubber (see Figure 2). This gas/liquid
absorption column is constructed of plexiglas and is
four feet high, three inches in diameter, and is packed
with hollow glass cylinders. A compressor feeds air
to the column, while SO2 is supplied from a cylinder.
Electronic mass flow meters and needle valves on
each line allow measurement and control of the in-
let gas composition. In our work, a nominal compo-
sition of two mole percent SO2 in the inlet air stream
was used. Tap water is fed to the top of the column
with a pump, and water flow rates are measured
and controlled with a rotameter and valve.
Several visual demonstrations can be made with
this apparatus. Since the column is made of plexiglas,
students can observe the flow of liquid over the pack-
ing. Varying gas and liquid flow rates allows them
to observe both gas flooding and liquid flooding,
which leads to a discussion of design and operating
variables, capacity, and design for flexibility. Add-
ing a small amount of sodium hydroxide and phe-
nolphthalein indicator to the water in the tank makes


IThe 50, Water Scrubber


Figure 2. The SO, Water Scrubber
Chemical Engineering Education









the inlet water bright pink. As the water flows down
the column and SO2 is absorbed and reacted, the
pink disappears. Varying the gas and liquid flow
rates causes the location of the color front to move.
Students get a strong visual indication of the
progress of the absorption process and see the effect
of operating variables on breakthrough.
The bulk of the research, however, was done using
plain tap water for scrubbing. In this work, stu-
dents measured inlet gas and liquid flow rates, and
titrated the outlet sample using a standard method.[6'
The ideal gas law was invoked to calculate gas mo-
lar flow rates, and water volumetric flow rates were
converted to molar rates using the density of water.
It was then a matter of simple material balances to
calculate the percentage SO2 removal and subse-
quently to observe the effect of flow rate and inlet
gas composition on the removal efficiency.

COMPUTER ACTIVITIES
To complement the laboratory work, students were
introduced to chemical process simulation using a
commercial package, PRO/II (Simulation Sciences,
Inc., Fullerton, CA), running on 486-based PC ma-
chines. We pointed out that our own undergradu-
ates learn this and other simulation packages in
their senior year, and that many of them subse-
quently use the same software in industry.
Since setting up a realistic simulation requires
extensive background, we had to provide assistance
to the students. PRO/II allows the user to specify
column and packing type, thermodynamic model
for SO2 solubility, and many other options with
which the students were unfamiliar, so we instead
used detailed handouts to try to give them a feel for
the simulator.
The purpose of the simulation was to re-create
as closely as possible the experimental conditions
used in the laboratory SO2 scrubber. Output from
the simulation included flow rate and composition
of all streams, so students could calculate percent-
age SO2 removal. It is a simple matter to adjust flow
rates and re-compute stream compositions, and stu-
dents had a chance to perform numerical experi-
ments similar to the physical experiments carried
out in the laboratory.
We discovered that our simulator did not agree
well with laboratory results. Experimentally we ob-
served seventy to eighty percent removal of SO2 from
the inlet gas, while the simulator consistently pre-
dicted removal in excess of ninety-nine percent. While
this was not the desired result, it did prompt class
discussion on the potential errors present in both
Summer 1993


simulation and laboratory work. Possible experimen-
tal error, inconsistent technique, and other difficul-
ties in laboratory work became apparent to them.
We were also able to point out that setting up a
simulation involves many menu selections and as-
sumptions which must be consistent with the ex-
perimental setup. Nevertheless, the students could
see clearly the iterative process that many practic-
ing chemical engineers must use in the design pro-
cess: laboratory experimentation over a specified set
of operating variables, followed by numerical simu-
lation, followed by careful error analysis and com-
parison of experiment and simulation, followed by
an improved set of experiments.

OTHER ACTIVITIES
We spent most of the time either in the classroom
or the laboratory, but for broader exposure we also
included several other activities (see Figure 1). ESP
students took several field trips, including an all-
day trip to a near-by power plant and a coal mine
where they were exposed to many of the activities
and unit operations associated with electrical power
production: open-pit strip mining, land reclamation,
rail transportation, solids handling, combustion,
steam generation, electrical turbines, cooling tow-
ers, scrubbers, electrostatic precipitators, and ancil-
lary control and monitoring equipment.
HISTEP participants were geared more toward
curriculum development, so they spent at least an
hour each day on teaching methodologies and on
developing course plans which incorporated the
HISTEP subjects into their classroom instruction.
These curriculum activities were coordinated by the
Wyoming Center for Teaching and Learning. The
teachers were evaluated on the basis of their course
plans and received continuing education credit for
their participation in the program.
We made extensive use of videotapes during the
four-day session. They were not exclusively on acid
rain, but were chosen to given the students a broader
exposure to issues and possible career paths in
chemical engineering."71 We also showed a depart-
ment videotape on chemical engineering careers
which we use in high school recruiting efforts and
gave each of the participants a copy of the tape for
later use at their respective schools.
Much of the acid rain unit dealt with experimen-
tal chemistry and engineering. While most of the
high school teachers found this interesting from a
personal standpoint, those whose teaching specialty
was mathematics had difficulty finding material to
take back to their classrooms. All HISTEP partici-
213










pants had been presented with a student version of
TKSolver (Universal Technical systems, Rockford,
IL). Therefore we presented the math teachers with
an extra problem dealing with the environmental
effects of paper and plastic grocery sacks, using a
discussion recently presented by Allen and
Bakshani."s This problem was explored using
TKSolver in List Solve mode.

DISCUSSION
Both ESP and the HISTEP classes were run in a
very informal atmosphere, and we had ample oppor-
tunity to ask the students what they were learning,
what they enjoyed, and what they did not like. In
addition, a brief, anonymous class evaluation form
was filled out by each student at the conclusion of
the unit. In this section, we will describe student
reactions to the unit and will pass on our impres-
sions of the successes and failures of the program.
Regens and Rycroft1t5 present some interesting his-
tory of air pollution (see Table 2). Our students were
interested to learn, for instance, that air pollution
was a problem in imperial Rome, that smoke from
wood and coal has been a problem for centuries in
Great Britain, and that King Henry once issued an
edict calling for decapitation of any who were found
"guilty" of burning coal.
An early example of completely misguided govern-
ment legislation in this area came in 1834, when
the British Parliament enacted laws requiring that
locomotives must consume their own smoke. (Our
students immediately sensed something amiss with
this law!) This particular bit of history gave us a
good opening for a discussion on the principle of
mass conservation and led to the first law of ther-
modynamics as well.
Because environmental issues are in the news
daily, we deliberately tried to provoke class discus-
sion about acid rain and other topics such as recy-
cling, nuclear waste storage, land reclamation, and
economic impact of environmental regulation. The
"Earth Summit" of 1992 led to a call for holding CO,
emissions at 1992 levels, so we asked the students
what conveniences they would be willing to give up
to help meet this goal. Not surprisingly, the "sacri-
fices" they volunteered were minimal (toaster
ovens, curling irons, etc.). The follow-up question,
however, did provoke considerable interest and dis-
cussion. The question was, "When would you be will-
ing to give up the items?" A few altruistic people
argued for voluntary conservation, while others said,
in effect, "We will conserve when the government
forces us to." A third group seemed content to wait


for Armageddon. The high school teachers pointed
out that when the price of electricity became too
high, people would voluntarily reduce consumption
(the law of supply and demand, and pricing). We
considered this sort of discussion to be important
because it put some laudable but abstract environ-
mental goals (conserve, reduce) on a very personal
basis for the students.
In reviewing the written comments, most students
made no mention, pro or con, regarding the class
discussions. We estimate that ninety percent of the
students actively participated in the discussions,
however, and they seemed to enjoy it. Indeed, among
the high school teachers it usually required an ef-
fort to terminate class discussion and move them on
to the laboratory.
All students enjoyed the hands-on demonstrations
and operation of the scrubber, and most of them
listed laboratory activities as their favorite part of
the course. It was usually a matter of showing them
the basics and then getting out of their way. The
mathematics teachers were the exception: while they


TABLE 2
Some History of Air Pollution

A.D. 61: Senaca (Roman philosopher) wrote of Rome's polluted
vistas:
As soon as I had gotten out of the heavy air of Rome and from
the stink of the smoke chimneys thereof which, being stirred,
poured forth whatever pestilential vapors and soot they had
enclosed in them, I felt an alteration of my disposition.
A.D. 1060: Eleanor of Aquitaine, wife of King Henry I of England,
moved from Tutbury Castle in Nottingham because of the pollution
of wood smoke.
A.D. 1273: English royalty issued decrees barring the burning of
coal in London. The effort was futile, because with the depletion of
forests in England (and lack of firewood) people increasingly turned
to coal.
Be it known to all within the sound of my voice, whosoever shall
be found guilty of burning coal shall suffer the loss of his head.
(King Edward I, ca. 1300)
A.D. 1578: Elizabeth I is annoyed by coal smoke and complains to
Parliament. Coal burning is banned while Parliament is in session.
A.D. 1661: John Evelyn wrote, "Fumifugium, or the Inconvenience
of Aer and Smoak of London Dissipated (together with some
Remedies Humbly Proposed)."
1772: Second edition of Evelyn's book is published.
1819 and afterward: British Parliament issued pollution abatement
decrees. Scrubber technology was developed in 19th century. In
1845, Parliament passed a law requiring locomotives to consume
their own smoke.
1952: A dense fog blanketed London from December 5-8. Fog,
mixed with polluted air, caused an estimated four thousand deaths
from emphysema, bronchitis, and cardiovascular problems. This led
to Britain's first modern clean air legislation.

Chemical Engineering Education









enjoyed the experiments personally, they expressed
concern about the practicality of transferring the
hands-on work to their own classrooms.
The booklet of supporting information seemed to
be of special value to the high school teachers. Con-
verting the laboratory readings into data for mass
balance calculations involved simple algebra, some
reaction stoichiometry, ideal gas laws, and knowl-
edge of fluid properties. In addition, we provided a
few word problems related to fuel, ash handling,
and shipping requirements for a coal-fired power
plant. The paper versus plastic bags problem of Allen
and BakshaniE81 was also very popular. The teachers
appreciated this real-world data and felt that their
students would enjoy working on such problems. We
believe we were successful in providing some math-
ematical problems that could be used in high school
mathematics, physics, or chemistry classes.
An area that needed more time, according to the
students, was computer modeling of the SO, absorp-
tion process. They would have preferred an opportu-
nity to try the various menu options and run more
cases. In retrospect, we see the need for more time
to experiment with the computer and to answer ques-
tions about its operation.
Another feature of the course that did not go over
as well as anticipated was the videotapes. Both high
school students and teachers indicated that there
were too many videos and that some of them ran too
long. We expected that the adults might grow rest-
less watching videotapes, but we were surprised (and
rather pleased) to find that high school kids, too,
preferred hands-on work to passive viewing.
Our department videotape did generate interest,
however-particularly among the teachers. We pro-
vided each of them with a copy of the tape to take
home, and they indicated they would show it in
their classes as a way to introduce students to the
field of chemical engineering in general and to our
department in particular.

CONCLUSIONS
Judging from the students' verbal and written com-
ments, we believe the unit was a success. Linking
chemical engineering principles and practice to an
environmental problem proved to be very effective
in capturing their interest. The operation of the
scrubber, analytical wet chemistry, and lab demon-
strations gave a hands-on experience that all of them
enjoyed. We were able to bring a strong element of
personal and societal values into the discussion, as
well as a discussion of technical issues. This helped
students appreciate the potential that chemical en-
Summer 1993


gineers have to design and develop useful solutions
to problems of public interest and concern. The value
of the engineer as a literate and articulate advocate
for the profession was stressed.
We entered the project with some trepidation. We
are accustomed to dealing with students whose chem-
istry, physics, mathematics, and engineering skills
are more developed. In addition, college-age students
well into their major are typically motivated by other
factors, such as a desire for good grades, fear of
failure, desire for a good job, and protecting their
large investment of money and time. Our only hold
on the participants in this program was to make the
course genuinely interesting and challenging.
While we feel that this course was a success, only
time will tell if we have achieved our goal of in-
creasing the number of students interested in chemi-
cal engineering as a career. We hope to monitor
incoming freshmen and transfer students in the
future to see if any graduates of ESP join our under-
graduate program. Similarly, we hope to find out if
high school teachers have been helped in describing
chemical engineering careers to their students. In
the meantime, we will continue with this and other
outreach efforts.

ACKNOWLEDGMENTS
This program was supported by NSF Grant #TPE-
9055487 and #RCD-9154885. Sally Steadman, De-
partment of Mechanical Engineering, was coordina-
tor of the ESP and HISTEP programs. David Cooney,
Chemical Engineering, provided assistance in the
SO2 scrubber exercises.

REFERENCES
1. "Diversity Increasing, Enrollments Declining in Engineer-
ing Schools," Chem. Eng. Prog., 16 (July 1992)
2. Bayles, T.M., and F.J. Aguirre, "Introducing High School
Students and Science Teachers to Chemical Engineering,"
Chem. Eng. Ed., 26(1), 24 (1992)
3. Felder, R.M., "There's Nothing Wrong With the Raw Mate-
rial," Chem. Eng. Ed., 26(2), 76 (1992)
4. Turk, J.T., and N.E. Spahr, "Chemistry of Rocky Mountain
Lakes," in Acidic Precipitation, Vol I. Case Studies, Adriano
and Havas, eds., Springer-Verlag (1984)
5. Regens, J.L., and R.W. Rycroft, The Acid Rain Controversy,
University of Pittsburgh Press (1988)
6. Standard Methods for the Examination of Water and Waste-
water, Procedure 429. APHA-AWWA-WPCF, 14th ed. (1975)
7. "Chemistry and the Environment," World of Chemistry #25,
Annenberg/CPB, South Burlington, VT; "Opportunities for
Chemical Engineers in Environmental Protection," AIChE;
"Opportunities for Chemical Engineers in Biotechnology,"
AIChE; "Opportunities for Chemical Engineers in Advanced
Materials," AIChE
8. Allen, D.T., and N. Bakshani, "Environmental Impact of
Paper and Plastic Grocery Sacks," Chem. Eng. Ed., 26(2),
82 (1992) C
215










, classroom


COMPUTING TEACHING WITH

FORTRAN 90


IAN FURZER
University of Sydney
Sydney, New South Wales, Australia 2006

Fortran 77 is taught in chemical engineering
departments throughout the world and is the
center of all scientific and engineering com-
puting. Fortran was developed by John Backus of
IBM and has passed through a number of stages
including Fortran 66 and in 1978 to the then-new
standard Fortran 77. ANSI and the International
Standards Organisation (ISO) began work on a new
standard for completion in 1982 and in 1991 intro-
duced ISO/IEC 1539: 1991. This new standard is
Fortran 90.
What teaching changes will be involved for com-
puting with Fortran 90? This question will undoubt-
edly be of primary concern to chemical engineering
departments during the next few years. The first
element to be considered is that everything written
in Fortran 77 will be fully compatible with Fortran
90. It will be possible to make no teaching changes
and to simply call on a new compiler, f90. But a
department that follows this policy will miss out on
all of the advantages developed over a decade of
work by experts in Fortran compilers. The advan-
tages make use of the best features of other lan-
guages so that a robust and reliable software code
can be written.
The intent of this paper is not to list all of the
Fortran 90 features but instead to simply introduce
it and give the reader an idea of the nature of the
new programs. Most of these new programs will not
look like Fortran 77 programs. The selection of the
programs presented in this paper will demonstrate
what Fortran 90 is all about.
One of the introductory topics discussed in the
teaching of Fortran 77 is the requirement that For-
tran statements lie between columns 7 and 72. Com-
ments start with a C in column 1 and continuation
lines with a character such as + in column 6, with


statement numbers in columns 1 to 5. This is called
fixed format Fortran and can be compiled with one
option of the Fortran 90 compiler. But this fixed
format is now considered obsolete and instead, pro-
grams can now be written in free format. This makes
teaching much easier.
With free format, Fortran 90 programs can start
in column 1 or any column through 132. There is no
need for indentation at the start of a Fortran line,
although this is often done for readability. Com-
ments begin with an exclamation mark (!) which
can be in any column. Comments can be added after
a Fortran statement by !, followed by the comments.
Continuation of a long Fortran 90 statement is per-
formed by adding an ampersand (&) to start the
next continuation line, and then an & to start the
next continuation line. This change to a free form
will be the first teaching change for Fortran 90.
Obviously, a free format form of Fortran will not
compile on a f77 compiler, so this begins the use of
added Fortran 90 features that require the f90
compiler. It should be noted that the word obsolete
was used above-a number of Fortran 77 statements
are not recommended as they are considered
obsolete. This recommendation leads to robust and
reliable code.
Teaching Fortran 77 led to difficult statements
like COMMON and BLOCK COMMON which were
useful in transferring information from a main pro-
gram to subroutines, or from subroutine to subrou-
tine. But neither of the COMMON forms are recom-
mended in Fortran 90. How can a Fortran 90 pro-
r1


Ian Furzer has been a faculty member in the
Department of Chemical Engineering at the Uni-
versity of Sydney for over twenty-five years. He
has extensive teaching and research interests
that include computing, process simulation, and
chemical engineering plant design. He is the
author of over eighty research publications and
the textbook, Distillation for University Students.


Chemical Engineering Education


Copyright ChE Division ofASEE 1993











gram operate without a COMMON statement? It
has a new feature, called the module. It removes
any doubts on program reliability that originated in
the COMMON statement.
Teaching changes may be progressive from giving
the full COMMON details in a Fortran 77 course
to giving no details at all in a Fortran 90 course.
The object of this teaching change concerns think-
ing about robust and reliable code while writing
the code.
Other Fortran 77 statements that have become
obsolete include DOUBLE PRECISION, computed
GO TO, and arithmetic IF statements, to mention a
few. Teachers of computing in chemical engineering
who are aware of these statements will need to know
the new replacements statements in Fortran 90.
A better approach to teaching Fortran 90 is to
present it as a new language with a wide range of
new features and statements. It covers a wider range
of features than Fortran 77, introducing structures,
pointers, arrays, and procedures, and can process
character strings and bits of information. Its
vocabulary is well defined and includes what at
first glance looks like unusual expressions, such as
"structure constructor." (Further details on For-
tran 90 can be found in Metcalf and Reid.") The
following are a few simple free format Fortran 90
programs, presented to provide a flavor of Fortran
90 to the reader.

FORTRAN 90 EXAMPLES

Example 1

Example 1 is shown below and includes the pro-
gram and end program statements. Note how the
program name, example 1, assists in locating the
full extent of the main program.
program example_l
print *, 'This is the output from example_l'
stop
end program example_l

The output from this program is given by
This is the output from example_l

Example 2
Example 2 is a Fortran 90 program that no longer
looks like a Fortran 77 program. Its function is to
demonstrate some features, including high preci-
sion calculations and derived data types. Real vari-
ables can be calculated with at least 10 decimal
places by the definition of an integer parameter r10,
shown on line 2. Real variables such as sum, c, and
d in the program carry the "r10" notation, giving
Summer 1993


What teaching changes will be
involvedfor computing with Fortran 90?...
The first element to be considered is that
everything written in Fortran 77 will
be fully compatible with Fortran 90.

these variables 10-figure precision. It is possible to
define other data types that have components. These
structures can be complex, but a simple example is
given of a type called university. It has three
components with data types: character, real, and
integer. It is bounded by the end type university
statement. The variables nsw and queens are de-
fined to be of type university and each will have
three components.
program example_2
integer, parameter : rl0=selectedreal_kind(10)
real (kind=rlO) sum, total, c, d, e
integer difference
type university
character (len=30) name
real engineering_depts
integer academic_numbers
end type university
type (university) nsw
type (university) queens
Simple examples follow
sum= 1.23456789 rl0
total=sum**2
c=1.0 rl0
d=3.0 rl0
e=c/d

write (*,*)'Kind=R10 Variables SUM=', sum
write (*,*) 'TOTAL=', total
write (*,*)' E=', e
nsw = university ('University of NSW', 8, 150)
queens = university ('University of Queensland', 6, 120)
difference = nsw%academicnumbers-queens%academic_numbers
write (*,*)'Difference in Academic Numbers=', difference
stop
end program example_2

The output from Example 2 is given by
Kind=R10 Variables SUM= 1.2345678899999999
TOTAL= 1.5241578750190519
E= 0.3333333333333333
Difference in Academic Numbers= 30

The precision of the output should be carefully
noted. Example 2 continues with the structure con-
structors for the type university variables nsw and
queens. This input of information includes the
"name" of the university, the number of
"engineering_depts," and the "academic_numbers"
in all departments. The difference between the
"academic_numbers" at nsw and queens is given by
the variable difference. The output shown above is
of course only as accurate as the information in the
structure constructors.











Example 3
Example 3 is similar to Example 2 but demon-
strates the use of procedures. The program consists
of three parts: the main program "example_3," a
module called type maker, and a subroutine called
calculation. Each part may be separately compiled
and the object code linked for execution. The main
program contains the statements
use type_maker
external calculation
call calculation
call writeresult

The main program listing is
program example_3
use type_maker
external calculation
integer, parameter :: rl0=selected_real_kind(10)
real (kind=rl0) sum, total, c, d, e
integer difference
type (university) nsw
type (university) queens

!simple examples follow

sum=1.23456789 r10
c=1.0 r10
d=3.0 r10
call calculation ( sum, total, c,d,e)

write(*,*)'Kind=R10 Variables SUM=', sum
write(*,*) 'TOTAL=', total
write(*,*) E=', e
nsw= university('University of NSW', 8, 150)
queens=university('University of Queensland', 6, 120)
difference=nsw%academic_numbers-queens%acadedmicnumbers
call write_result (difference)
stop
end program example_3

Modules are a very important feature of Fortran
90. They can be used by the main program and
subroutines to access information (such as the defi-
nition of the type university) that more than one of
them needs.
MODULE TYPE_MAKER
type university
character (len=30) name
real engineering_depts
integer academic_numbers
end type university
contains
subroutine write_result (number)
integer, intent (in) :: number
write (*,*)'Difference in Academic Numbers=', number
end subroutine write result
END MODULE TYPE MAKER

Modules are procedures that can also contain sub-
programs such as the subroutine write result. Note
that the module is named type maker and ends with
end module type maker. An example of an external
subroutine is given by the subroutine calculation.

subroutine calculation (a, b, c, d, e)
integer parameter :: rl0=selected_real_kind(10)
218


real ( kind=rl0), intent (in) :: a,c,d
real ( kind=rl0), intent (out):: b,e
b=a**2
e=c/d
end subroutine calculation

The subroutine arguments a, b, c, d, and e are
of kind, r10, that is of at least 10-figure pre-
cision. Fortran 90 also uses the attributes, intent
(in) and intent (out), to be used to specify the input
and output arguments to the subroutine. The out-
put from Example 3 is identical with the output
from Example 2.

Example 4
Example 4 shows some of the powerful features of
Fortran 90: pointers, targets, automatic arrays, do
loops, and matrix multiplication.

program example_4
real, pointer :: finger
real, target :: a,b
real, dimension ( : : ), allocatable :: matrixa, matrix_b
real, dimension ( : : ), allocatable :: matrix_c
integer n
a=1.0 ; b=2.0 ; n=5 !n could be entered by read (*,*)n
allocate ( matrix_a(n n), matrixb(n n) matrix_c(n n)
j_loop : do j=l, n
k_loop : do k=l, n
matrix_a(j,k) =real(j)*real(k)
matrixb(j,k) =matrix a(j,k)
if(matrix_b(j,k) <= 10.0 ) cycle
matrixb(j,k)matrixb(j,k))trib(j + sqrt(0.5)
end do k_loop
end do j_loop
matrix_c=matmul(matrix_a, matrix_b)
finger=>b
if(n==l) finger => a
write(*,*) 'Example_4 Finger=', finger
stop
end program example_4

A real variable, "finger," is given pointer attributes
by its definition. A pointer points to a target and
real variables, "a,b," are given target attributes. They
are a new and powerful feature of Fortran 90, par-
ticularly useful in operating on linked lists such as
an adapted refined grid. Their use in engineering
may not immediately appear to be obvious, but a
simple example is shown as part of Example 4.

Arrays in Fortran 90 can have fixed bounds such
as a(10), but only a section of the array can be used
with the colon ( : ) notation, such as "a( 5 : 10 )." A
two-dimensional array with initially unspecified
lower and upper bounds is given by matrix a ( :. : ).
These bounds can be allocated during execution,
making for greater flexibility. Example 4 shows three
arrays: matrix_a, matrix_b, and matrix_c, with the
attribute allocatable. An integer n that is given the
value 5 in Example 4 could have been read in and
could take on a wide range of integer values. The
allocate statement then allocates the required

Chemical Engineering Education









amount of memory for these arrays.
Do loops are considerably different in Fortran 90
and may include no labels. They start with a do
statement and end with an end do statement.
Each loop may be given a name, such as "j_loop,"
which assists in locating the limits of a parti-
cular loop. There are no statement numbers in the
recommended form of the do loop, the final state-
ment being the end do statement. The cycle in-
struction permits a direct jump to the end do state-
ment and the exit statement permits a direct exit
from the loop.
Fortran 90 permits direct matrix multiplication
through the "matmul" statement. Other matrix op-
erations are standard in Fortran 90.
The statement
finger => b
shows the pointer finger is pointing at the target b.
The next statement contains the logical equal com-
parator, and if it is satisfied
finger => a
The output from Example 4 is given by


Example_4


Finger= 2.0000000


SUPERCOMPUTERS
Fortran 90 could well be the new world standard
in computing until the year 2000. Fortran 90 com-
pilers can exploit the advanced architecture of par-
allel processors or supercomputers. It might be ex-
pected that desktop supercomputers will lead to con-
siderable advances in engineering, particularly in
finite element methods and computational fluid me-
chanics. It might also be expected that the obsolete
and archaic features of some parts of Fortran 77
(such as the arithemetic IF statements, some DO
statements, and the H edit descriptor) will be re-
moved in later versions of Fortran 90. Fortran 90
statements can begin in column 1, thus removing
the obsolete card image concept in Fortran 77. One
of the important advances of Fortran 90 is the avail-
ability of instructions that give a good methodology
in program design. This can lead to both robustness
and an error-free code, which can be fully exploited
on supercomputers.

CONCLUSIONS
Engineers will need to spend some time learning
the new features of Fortran 90 if they wish to un-
dergo the conversion from Fortran 77. A program
written in Fortran 90 may not look like a Fortran 77
program because of the many new features of
Fortran 90, and many obsolete features of Fortran
Summer 1993


77 should no longer be used. The only Fortran
90 compiler available from NAG provides reason-
able error messages during compiling, which is an
improvement over Fortran 77. In some cases the
error messages even identify a line number and
print a part of the statement that contains the er-
ror. Also, the compiler lists undefined variables.
Error messages during execution are good and,
for example, provide the dimensions of matrices if
they do not compute.
One of the most important advantages of Fortran
90 will be the portability of code. A large number of
software products such as mathematical subroutines
and graphical packages will be rewritten in Fortran
90 to provide good interfaces.
This article makes no attempt to list all the state-
ments and features of Fortran 90, as it is an exten-
sive and powerful new language. There can be no
doubt that it will have an important impact on the
engineering profession for a number of years to come.

ACKNOWLEDGMENT
Comments by John Reid of the Rutherford Lab,
England, were most welcome.

REFERENCES
1. Metcalf, M., and J. Reid, "Fortran 90 Explained," Oxford
University Press (1990) O

REVIEW: Plastics Recycling
Continued from page 199.
rapidly changing field, however, some of the infor-
mation in this book is already dated. Advances in
recycling have produced better processes which
allow recycled plastics with specifications (much
like virgin plastics) to be produced, particularly
by companies like Union Carbide, Dow, Mobil, Quan-
tum, and Waste Alternatives. These improvements
in processes and products have inevitably led to
the demise of several small companies-especially
in the plastic lumber area. Many large companies,
however, (such as Mobil and Amoco) have entered
the field with more efficient processes and better
quality control.
A more recent book, published by the American
Chemical Society, ACS Symposium Series 513,
Emerging Technologies in Plastics Recycling, (1992),
is also becoming dated, but has significantly more
scientific data. Clearly, plastics recycling is a dy-
namic area of research and business, and continued
developments are in progress. This book provides
an excellent starting point for those who are inter-
ested in plastics recycling. O
219










B classroom


SIMULATION IN THE

CHEMICAL ENGINEERING CLASSROOM


WALLACE B. WHITING
West Virginia University
Morgantown, WV26506-6101

any educators have found that active learn-
ing strategies can help students develop
the higher level thinking skills of analysis,
synthesis, and evaluation that are the essence of
engineering.[1'21 Simulation is an effective technique
for creating a classroom environment that is condu-
cive to such learning. Whether we simulate a pro-
cess on a computer or the work of an engineering
team in a design project, the simulation experience
brings a sense of reality to an assignment.3'4" Stu-
dents become more active, more interested. Few fac-
ulty who have engaged their students in simulation
doubt its effectiveness.
Can we use simulation in all our courses? Should
we? Through the following examples, I hope to show
that the answer to both of these questions is "yes."

ROOT-FINDING ALGORITHMS
On the first day of a course in numerical methods
for chemical engineers, I break the class into small
groups of three to five students, and each group is
told to find the largest positive real root of a func-
tion. The students don't know, however, that each
group is given the same function, but in a different
form. One group must find the solution by querying
a computer program set up on a PC that accepts
their guess and gives a value of the function, while
another group is given an algebraic form of the func-
tion, which is a cubic. A third group is given a differ-
ent form of the function, and yet another group is
given the problem behind the assignment: to solve
the van der Waals equation for the vapor molar

Can we use simulation in all our courses?
Should we? Through thefollowing examples, I
hope to show that the answer to both of these
questions is "yes."

Copyright ChE Division ofASEE 1993


Wallace B. Whiting, P.E., is Associate Profes-
sor of chemical engineering at West Virginia
University, where he has taught for the past
decade. He is active in ASEE and AIChE, and
his research and teaching interests range from
thermodynamics to process safety and process
design. He welcomes dialogue on this and all of
his articles.


volume given the constants a and b, the tempera-
ture, and the pressure. Each group is told to find
the solution to the problem and to carefully keep
track of how they solved it so that they can explain
their technique to the rest of the class.
The results of this exercise are amazing. My sopho-
more students (who have had no previous courses in
numerical methods) independently develop all the
root-finding algorithms in the text within thirty min-
utes, in addition to some other more sophisticated
algorithms. (It should be noted that while the stu-
dents have all brought the text to class, usually
none of them has opened it yet!) As the groups go to
the front of the class to explain their techniques, I
tell them that this is called the bisection method, or
Newton's method, or resubstitution with accelera-
tion, or brute-force, or whatever. We discuss such
topics as error propagation, accuracy criteria, ad-
vantages of solving the function analytically (for the
groups that have the analytic form of the function),
strategies for developing a good initial guess, and
the risks and benefits of using a method that has
already been programmed. Our discussion invari-
ably broadens to include the roles of textbooks, com-
puters, their own physical understanding, and their
colleagues in solving numerical problems. We talk
about why they are taking the course and what its
relationship is to the rest of the curriculum and to
their future careers.
Simulation? Yes, we simulated an engineer's at-
tempt to solve a numerical (in this case, thermody-
namics) problem. The students had a wide variety
Chemical Engineering Education










of experiences (guaranteed by the dif-
ferent versions of the problem) which
they shared with each other in a struc-
tured way.
Through simulation the students de-
velop new strategies for solving nu-
merical problems. They learn quite a
bit about specific numerical techniques
(more than they could have learned in
one traditional lecture), and they gain
confidence in their abilities to solve
new classes of problems. Perhaps most
important-they learn the connection
between numerical methods and the
rest of chemical engineering. And, of
course, they learn some thermo.
For larger classes, I use more groups
rather than larger ones, and I choose
one group of each type to lead the class
discussion. If discussion lags (which is
rare), another group is asked how its
approach differed.
I have used the same kind of exer-
cise with an optimization problem, with
similar results.

THE LEVEE PROBLEM
I have used the "levee problem" sev-
eral times-in senior design courses
at two different universities and at a
conference where most of the partici-
pants were chemical engineering fac-
ulty.[5E The kernel of the simulation
comes from a homework problem, the
source of which has been lost over the
years. As originally stated, the prob-
lem appears to be a single-answer eco-
nomics problem in which the student
compares building a flood-control levee
to not building it. Students are given a
scenario in an expanded version of the
problem (see Figure 1). They must
write a memo, detailing their recom-
mendations, to a specific person who
has a specific technical background.
Responses to the assignment are
wide ranging. Some students focus
solely on the economics, concluding
that the levee is a terrible alternative.
Others ignore the costs and make what
they feel to be the ethically correct rec-
ommendation (to build the levee at any
cost). Many other solutions are pre-
Summer 1993


But, what have they learned? ... They knew what to do next;
they knew how to learn on their own; they gave themselves the next
assignment; they were good problem solvers. Heady stuff.
And they learned that statistics can be useful-
all within the confines of one class period.

sented, but few students attempt to tie the various aspects of the
problem together.

PART 1
Memo and Feedback
In class, the students are paired and asked to play the role of
Chris E. Smyth as they read each other's memos. Chris is a very
busy executive who has only three minutes to read the memo and
to answer the following three feedback questions: (1) What is the
recommendation? (2) Do you trust (or believe) the recommenda-
tion? (3) What will you do next? The students give immediate



Technocats Inc.
403 Engineering Sciences Bldg.
P.O. Box 6101
Morgantown, WV 26506-6101
(304) 293-2111

TO: Workshop Participants
FROM: W.B. Whiting, Vice President of Engineering U/&
DATE: 18 October 1990
SUBJECT: Economics Revisited
CC:
Please submit your answer to the following problem by noon, Wednesday, Octo-
ber 24, 1990. Someone will be chosen to present the problem to the group at 1:30
P.M., Thursday, October 25, in Room 449.
In your position as Director of Investment Planning at Technocats, you must
make a specific recommendation for or against the capital expenditure described
below. Write a memo to your boss, Chris E. Smyth, Executive Vice President,
recommending and justifying appropriate action.
After a chemical plant was built on an island in a river, our geology/hydrology
department discovered that the island was occasionally under water. U.S. Army
Corps of Engineers data indicate that the chances of a flood are about one in seven
each year. The likely flood damage would be $250,000, and we would need to lay
off ten employees for three months during the repair, which would be a boon for a
local construction firm.
A protective levee would cost $600,000 (total fixed capital investment), and our
civil engineering department estimates its useful life as 28 years, with no salvage
value. Our economics group has worked up an opportunity cost of 10% per annum,
based on after-tax cash flows and no inflation. Other economic data are given below.
Remember, your boss is a former geologist who doesn't know much about time
value of money, discounting, annuities, etc.
Fixed costs other than maintenance: negligible
Annual maintenance cost: 1% of fixed capital investment
Effective income tax rate: 48%
Depreciation: AMACRS, 15-year class life (ignore mid-year convention)
Working capital: none
Use end-of-year annualized costs.

Figure 1. Levee problem assignment memorandum










feedback to each other by answering these
questions.

We then discuss the responses. With some
classes I have already brought in engineering
ethics, but with others I have not. Regardless
of their background, however, the intensity of
the discussion is always high.
PART
Engineering Presentation
The students are formed into small groups,
and each group is given a role to play in the
ensuing simulation (see Figure 2). The stu-
dents are given fifteen to twenty minutes for
preparation before the simulation begins. One
member of a "Director of Investment Planning"
group is chosen to present the results and rec-
ommendations to the vice president, chosen
from a "Chris E. Smyth" group. Some groups
prepare extensive overhead transparencies or
"chalk talks," while others develop note cards
or lists of questions. As before, more than one
group may be given the task of preparing the
Chris E. Smyth role, but only one member
from one group is chosen for the simulation in
front of the class. Thus, all students actively
prepare for the simulation and all have an
interest in the outcome, even though only one
student will ultimately play the role. The other
group members are anxious to see how their
strategy works, and the groups that are not
chosen have an opportunity later to compare
their strategies to the ones demonstrated.
During the preparation time, no group knows
what the other groups have been told. In par-
ticular, Chris E. Smyth and the Director of
Investment Planning have no idea what will
happen next. As a reporter enters the scene,
with tape recorder running, the simulation
heats up, and what you fear will happen usu-
ally does. The engineer talks to the reporter as
if the reporter were an engineer, or an idiot.
Outrageous one-liners (sound bites) are uttered.
The reporter has a field day. Sometimes Chris
E. Smyth brings things back under control,
and sometimes not. You never know what is
going to happen.
Next an observer group leads the class dis-
cussion. I initiate the first topic: Why are we
doing this? Although they come up with vari-
ous responses, none of the students has ever
indicated any concern that time was wasted in
this simulation exercise. They all know that it
222


Chris E. Smyth
You are the Executive Vice President of Technocats, Inc. As
indicated on the assignment memorandum, you are a former
geologist (not a chemical engineer) who doesn't know much
about time value of money, discounting, annuities, etc. How-
ever, you are a good manager and have risen through the ranks
at Technocats because of your ability to deal with people and to
get things done. The Director of Investment Planning reports to
you, and you really need to know what to do about this levee/
flooding situation. You will make the final decision, but you
will need to justify it to the President, to the CEO, and, perhaps,
to the Board of Directors.
We have simulated your office, complete with desk, phone,
etc. The Director of Investment Planning will come to your
office to make a presentation.


Director of Investment Planning
You are a chemical engineer. As indicated in the assignment
memorandum, you must make a specific recommendation for
or against the capital expenditure described. You will go to
Chris Smyth's office and give a presentation (you are expected).
Chris will make the final decision, but you and your group have
done all the analysis. Remember the background of Chris Smyth
while preparing your presentation. You will have only about 5
minutes for the presentation.


Reporter
Neither Chris Smyth nor the Director of Investment Planning
knows that you will show up. You are an investigative reporter
for the Washington Post. You have heard a rumor that Technocats
has a chemical plant that could be subject to a flood. As with
most journalists, you have no scientific or engineering training,
but you are bright and ambitious. You go to see Chris Smyth,
Executive Vice President, without an appointment. You have a
tape recorder to record everything. During the simulation, you
will be announced by the secretary, but, before Chris has a
chance to say no, you enter the office and begin asking ques-
tions.

Observer
Your job is to observe what is going on. The Director of
Investment Planning will give a presentation to Chris Smyth
(Executive Vice President). During the presentation, an investi-
gative reporter from the Washington Post will show up and
begin asking questions. Neither Chris nor the Director of In-
vestment Planning has any idea that the reporter exists.
After the simulation, you will be called upon to give an
analysis of the simulation and to lead the postmortem.

Figure 2. Roles for the levee problem simulation
Chemical Engineering Education










can happen to them. They realize that the engineer's
inability to say the "right things" to the reporter
stems from the engineer's overly simplified analysis
of the situation and a lack of understanding of the
difference between engineers and the rest of the
Distribution of Floods


4- V /
3- 11y 4 y
2- H

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Number of Floods in 28 Years

Figure 3. Distribution offloods
Distribution of NPV
30


0-60 110160 -260 310-360 410460 510-560
-NPV (thousonds of dollars)

Figure 4. Distribution of net present value
Cumulative Distribution of NPV
Cumulative Distribution of NPV


200 400
-NPV (thousands of dollars)


Figure 5. Cumulative distribution of net present value
Summer 1993


population which they serve. The ensuing discus-
sion ranges from topics such as ethics, to net present
value, to statistics.

PART
Monte Carlo Simulation
Yes, statistics. Students typically assume, for their
calculations, that the interval between floods is ex-
actly seven years. Some students assume that the
first flood is in the first year while others assume
that it occurs in the seventh year. The first time I
used this simulation, my students asked how to quan-
tify the chances that there would be a given number
of floods in a given number of years and how the
timing of such floods would affect the economics.
They realized that they couldn't integrate the soci-
etal, technical, and financial aspects of the problem
without such data. The students developed the next
assignment: to do a Monte Carlo simulation to de-
termine the frequency distribution of floods and the
frequency distribution of the net-present-value com-
parison (essentially a probabilistic benefit-cost analy-
sis"61). Most of these students had taken no courses
in probability or statistics, and none of them had
heard of Monte Carlo simulation when I told them
that that was what they were describing. Most stu-
dents chose to use LOTUS 1-2-3 for the Monte Carlo
simulation, but some wrote BASIC or FORTRAN
programs. The results of their simulations are shown
in Figures 3, 4, and 5. The same thing has happened
with succeeding classes.
But, what have they learned? After years of dili-
gently studying for what they hoped would be a
rewarding career, they now know it's real. What
engineers do is important (and difficult and scary).
They are preparing for their career, and they know
it. Realizing that their calculations can affect real
people in serious ways is exciting. They knew what
to do next; they knew how to learn on their own;
they gave themselves the next assignment; they were
good problem solvers. Heady stuff. And they learned
that statistics can be useful-all within the confines
of one class period.
But who was simulating what? The students were
simulating a common engineering situation wherein
after analysis, synthesis, and evaluation the engi-
neer presents and defends a recommendation to tech-
nical and non-technical audiences, receives criticism,
and decides what to do next.
For ease of use in different classroom settings, I
have broken this levee problem simulation into three
parts. The memo and feedback portion of the prob-
lem (Part 1) can be used to introduce students to
223










simulation. The engineering presenta-
tion (Part 2) can be used with or without
the reporter, or the full simulation, in-
cluding the Monte Carlo assignment (Part
3) can be used. In each case, a postmor-
tem discussion of the simulation, bring-
ing in as many student viewpoints as pos-
sible, is essential.
TO:
OTHER SIMULATIONS
FROM:
There are many other types of simula- DATE:
tions, ranging from intricate ones to simple SUBJE4
ones. At West Virginia University we have cc:
used an extended and involved simula-
tion for the year-long senior design project Pleas
for over fifty years.[4] An AIChE Student OSHA r
Contest Problem can also be adapted for engineer
individual or group simulation. A very and We
simple simulation which I have used to present
introduce a guest lecturer in a course is reactor c
shown in Figure 6. At first glance, it ap- Our r
pears that nothing has changed. I have Progres.
merely let the students know the topic for
two lectures while I am out of town. The
subtle simulation here introduces the stu-
dents to the concept of professional devel-
opment activities and to the role of gov-
ernment regulation in their profession. Attack
Their attitudes toward the lectures are
modified, and their learning is enhanced. Lo.
Pe
I have sometimes videotaped more in- Ku
tricate simulations, with the class view- Ac
ing and analyzing the simulation during
the postmortem. Additional faculty, gradu-
ate students, and others (including our
outside seminar speakers) are sometimes
brought in to participate.

Design projects and laboratory experiments can
easily be developed as simulations, but a good home-
work problem from any engineering course can be
put into a real context. After preparation, students
role-play the situation. Finally, class discussion can
help students explore and understand the impor-
tant relationships both within the subject matter
and between it and the bigger picture.

CONCLUSIONS
Simulation is an ideal technique for improving
student learning. These activities help students rec-
ognize connections between courses, the curriculum,
and their profession. They exercise critical thinking
skills and develop self-guided learning strategies.
Simulation offers an opportunity to both broaden
224


Technocats Inc.
5532-F Boelter Hall
University of California, Los Angeles
Los Angeles, CA 90024-1592


Task Force 108A
W.B. Whiting, Vice President of Engineering ob"
26 February 1992
CT: Safety


e read the attached article from the Los Angeles Times concerning new
regulations. To prepare for these regulations, it is imperative that all Technocats
*s attend a special Professional Development Seminar on Monday, March 2,
dnesday, March 4, 8:00-9:50 A.M., in Room 2444 of our headquarters. We
anged for an outside expert in HAZOP (HAZard and OPerability study) to
these seminars. For Friday, March 6, you will perform a HAZOP on the
)f our subsidiary XXX's Acrylic Acid Plant.
research department has found the attached article from Plant Operations
s that should be of interest.







hments:
s Angeles Times, "OSHA Issues Safety Rules to Avert Explosions at
trochemical Plants," February 15, 1992, p. A5.
rland, J.J., and D.R. Bryant, "Shipboard Polymerization of Acrylic
id," Plant/Operations Progress, 6(4), 203 (1987).


Figure 6. Announcement of guest lecturer


and deepen coverage while enhancing student learn-
ing.

REFERENCES
1. Smith, K.A., "Cooperative Learning:An Active Learning
Strategy," Proc., Frontiers in Educ. Conf., Binghamton, NY,
pp. 188-193, October (1989)
2. Felder, R.M., "Creativity in Engineering Education," Chem.
Eng. Ed., 22, 120 (1988)
3. Woods, D.R., D.W. Lawson, C.A. Goodrow, and R.A. Romeo,
"Career Planning and Motivation Through an Imaginary
Company Format," Chem. Eng. Ed., 16 44 (1982)
4. Gardner, A.A., P.H. Whiting, and A.F. Galli, "From Raw
Materials to Profit: Career Role-Playing in a Senior Design
Project," AIChE Ann. Meet., Los Angeles, CA, November
(1982)
5. Whiting, W.B., "Simulation: The Key to Effective Educa-
tion," AIChE Ann. Meet., Los Angeles, CA, November (1991)
6. Morgan, M.G., and M. Henrion, Uncertainty: A Guide to
Dealing with Uncertainty in Quantitative Risk and Policy
Analysis, Cambridge University Press, Cambridge (1990)

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

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

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

ACKNOWLEDGMENT Include in acknowledgment only such credits as are essential.

LITERATURE CITED References should be numbered and listed on a separate sheet in the
order occurring in the text.

COPY REQUIREMENTS Send two legible copies of the typed (double-spaced) manuscript on
standard letter-size paper. Submit original drawings (or clear prints) of graphs and diagrams on
separate sheets of paper, and include clear glossy prints of any photographs that will be used. Choose
graph papers with blue cross-sectional lines; other colors interfere with good reproduction. Label
ordinates and abscissas of graphs along the axes and outside the graph proper. Figure captions and
legends will be set in type and need not be lettered on the drawings. Number all illustrations
consecutively. Supply all captions and legends typed on a separate page. State in cover letter if
drawings or photographs are to be returned. Authors should also include brief biographical sketches
and recent photographs with the manuscript.










MILLIKEN & COMPANY
College Relations Department, M-285
P.O. Box 1926
Spartanburg, SC 29304


ENTRY LEVEL OPPORTUNITIES FOR CHEMICAL ENGINEERS



GENERAL INFORMATION

Milliken is a major manufacturer of textile products for apparel, commercial, home,
and industrial markets, as well as specialty chemicals for a wide variety of applica-
tions. The company was founded in 1865 and has nearly 50 plants and 13,000 associ-
ates in the US. Milliken is a world leader in quality manufacturing and was the 1989
winner of the Malcolm Baldridge National Quality Award. Careers at Milliken involve
challenge, innovation, advanced technology, promotion from within based on indi-
vidual performance, and extensive education and training opportunities. Entry level
opportunities are available in South Carolina and Georgia.



CITIZENSHIP REQUIREMENTS: U.S. citizenship or Permanent Resident Visa
REGIONS WHERE BS/MS CAMPUS RECRUITING IS CONDUCTED: Southeastern United States
HOW TO APPLY IF UNABLE TO SCHEDULE A CAMPUS INTERVIEW: Send cover letter with
functional area interests and geographic preference statement, resume, and a copy of your transcript to
the above address.


BS/MS -

Process Engineering: Provides technical support in textile dyeing and finishing operations
and in Specialty Chemicals production. Responsibilities include manu-
facturing compliance with customer product quality specifications
and process efficiency/improvement project assignments.
Manufacturing Management: Responsible for the production resources of people and machinery.
The first line production manager may be promoted to either Ad-
vanced Production Manager or Process Engineer in the dual career
ladder.
Research: Develops new products and associated machinery or processes. Prefer
PhD, but will consider MS


An Equal Opportunity Employer




Full Text