Citation
Chemical engineering education

Material Information

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

Subjects

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

Notes

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

Record Information

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

UFDC Membership

Aggregations:
Chemical Engineering Documents

Downloads

This item has the following downloads:


Full Text





cemcal engieig e







We wish to
acknowledge and thank...


3M


FOUNDATION


...for supporting
CHEMICAL ENGINEERING EDUCATION


with a donation of funds.










EDITORIAL AND BUSINESS ADDRESS

Department of Chemical Engineering
University of Florida
Gainesville, Florida 32611
Editor: Ray Fahien (904) 392-0857
Consulting Editor: Mack Tyner
Managing Editor:
Carole C. Yocum (904) 392-0861
Publications Board and Regional
Advertising Representatives:
Chairman:
Gary Poehlein
Georgia Institute of Technology
Past Chairmen:
Klaus D. Timmerhaus
University of Colorado
Lee C. Eagleton
Pennsylvania State University
Members
SOUTH:
Richard Felder
North Carolina State University
Jack R. Hopper
Lamar University
Donald R. Paul
University of Texas
James Fair
University of Texas
CENTRAL:
J. S. Dranoff
Northwestern University
WEST:
Frederick H. Shair
California Institute of Technology
Alexis T. Bell
University of California, Berkeley
NORTHEAST:
Angelo J. Perna
New Jersey Institute of Technology
Stuart W. Churchill
University of Pennsylvania
Raymond Baddour
M.I.T.
NORTHWEST:
Charles Sleicher
University of Washington
CANADA:
Leslie W. Shemilt
McMaster University
LIBRARY REPRESENTATIVE
Thomas W. Weber
State University of New York


Chemical
VOLUME XXII


Engineering
NUMBER 2


Education
SPRING 1988


Educator
58 James E. Bailey of Caltech, Robert Finn

Department
62 Illinois Institute of Technology,
Satish J. Parulekar

Laboratory
68 A Pilot-Scale Heat Recovery System for
Computer Process Control Teaching and
Research, P. J. Callaghan, P. L. Lee,
R. B. Newell
74 Safety and Loss Prevention in the Under-
graduate Curriculum: A Dual Perspective,
Daniel A. Crowl, Joseph F. Louvar
80 Engineering Management: A Course to Minimize
Functional Handicaps of Graduates,
Rex T. Ellington
84 A Course on Presenting Technical Talks,
Richard M. Felder

Classroom
88 Flow Sheet is Process Language, Manfred Fehr
98 Discrete-Event Simulation in Chemical
Engineering, Daniel J. Schultheisz,
Jude T. Sommerfeld
104 Levels of Simplification: The Use of Assumptions,
Restrictions, and Constraints in Engineering
Analysis, Stephen Whitaker

Lecture
92 The Mystique of Entropy, B. G. Kyle

71 Letter to the Editor

72 Memoriam: James J. Christensen

73 Memoriam: Roland Andrew Ragatz

79, 86, 90, 91, 103 Book Reviews

108 Books Received


CHEMICAL ENGINEERING EDUCATION (ISSN 0009-2479) is published quarterly by 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 Engineerin Department, University of Florida, Gainesville, FL 32611. Advertising mate-
rial may be sent directlyE. O.Painter Printing Co., P. O. Box 877, DeLeon Springs, FL 32028. Copyright
1988 by the Chemical Engineering Division, American Society for Engineering Education. The statements
and opinions expressed in this periodical are those of the writers and not necessarily those of the ChE
Division, ASEE, which body assumes no responsibility for them. Defective copies replaced if notified with
120 days of publication. Write for information on subscription costs and for back copy cost and availability.
POSTMASTER: Send address changes to CEE, Chemical Engineering Department, University of Florida,
Gainesville, FL 32611.


SPRING 1988










Educator


JAMES E. BAILEY


of Caltech

ROBERT FINN
California Institute of Technology
Pasadena, CA 91125

HIS COLLEAGUES describe James E. (Jay) Bailey
as a man open to new ideas. Trained as a classi-
cal chemical engineer (Bailey did his doctoral work on
chemical reactor theory, optimization theory, and non-
linear mathematics) he soon developed an interest in
the relatively young field of biochemical engineering.
Lately, Jay has turned even more biological in his ap-
proach, pioneering a brand new field he calls metabolic
engineering. His graduate students, who have joined
in this odyssey, are well prepared to combine the tech-
niques of molecular biology with more traditional en-
gineering methods in the chemical engineering prob-
lem-solving armamentarium.
Jay ascribes many of the twists and turns of this
intellectual odyssey to fortunate encounters with
bright students and interesting colleagues. One of
these chance encounters took place at the University
of Houston in 1972 when Jay, then on the UH faculty,
happened to meet biochemical engineer David Ollis.
(Ollis was at Princeton University at that time; he is
now Distinguished Professor of Chemical Engineering
at North Carolina State University.) As Jay recalls it,
"I was really impressed by the breadth of Dave's
knowledge in biochemical engineering. I didn't know
what a protein was in those days, but after a 30-min-
ute conversation and a couple of phone calls, we de-
cided to write a textbook."
The result was Biochemical Engineering Funda-
mentals, which has by now become the required text
in almost every biochemical engineering course in the
United States and in much of the rest of the world. It
is now in its second edition. "Writing the first edition
of this text was a tremendous learning experience,"
Jay recalls. "I didn't know much biochemistry, micro-
biology, or biochemical engineering at that time. I
read texts and review articles, and I went to Elmer
Gaden's short course. All this convinced me that we
had to emphasize the background material in
biochemistry, physiology, and genetics that differen-

Copyright ChE Division ASEE 1988


tiates biochemical engineering from chemical en-
gineering. A lot of other texts assume that you know
what a protein is, what a bacterium is, what a gene
is. We decided to define those terms, and others, very
explicitly. I think that my entry into the field as a
beginner made the text suitable for other beginners.
The second edition, prepared after I had been heavily
involved in biochemical engineering research for more
than a decade, provided an opportunity to inject some
personal perspectives into the text, in addition to im-
proving its organization and updating its content."
Jay's own education as a chemical engineer was
typical enough. He did both his undergraduate and
graduate work at Rice University, earning his doctor-
ate working on reactor theory and optimization with
Fritz Horn. After a two-year stint with Shell Develop-
ment in Emeryville, California, Jay established his
own laboratory in the Department of Chemical En-
gineering at the University of Houston. It was there
that he first let himself be nudged in the direction of
biochemical engineering.
"In the late 1960s and the early 1970s, the National
Science Foundation sponsored a major effort in en-


CHEMICAL ENGINEERING EDUCATION










Trained as a classical chemical engineer, [Jay] soon developed an interest in the relatively young field
of biochemical engineering. Lately, he has turned even more biological in his approach, pioneering a brand
new field he calls metabolic engineering. His graduate students, who have joined in this odyssey .
combine the techniques of molecular bioloav with more traditional enaineerina methods ...


zyme engineering," Jay notes. "I was doing computa-
tional and theoretical work on immobilized enzyme
catalysis and on some diffusion-reaction problems with
unusual properties. Two of my graduate students at
UH had interesting backgrounds-Mike Cho had a
background in biochemistry, and Jila Fazel-Madjlessi
had a background in microbial fermentation. They
helped me start experimental programs in those
areas."
With Cho, for example, Bailey made major
theoretical and practical advances in the area of im-
mobilized enzymes. Their work involved the first ap-
plication of the carbodiimide-mediated covalent at-
tachment of enzymes to activated carbon. Using this
approach they constructed hybrid catalysts that com-
bined an enzyme's activity with the catalytic action of
the support surface. Specifically, they looked at the
oxidation of glucose by the enzyme glucose oxidase. A
byproduct of this reaction is hydrogen peroxide, which
tends to deactivate the enzyme. Bailey and Cho dis-
covered that activated carbon decomposes the
peroxide and also assists in the supply of oxygen to
the enzyme sites within the porous catalyst. They
also discovered, however, that loading too much en-
zyme onto the support compromises its efficiency in
deactivating peroxide, and they were able to charac-
terize the tradeoff between higher enzyme loading and
peroxide decomposition in some detail.
Jay's experimental work on immobilized enzymes
continues to this day. One of his most significant dis-
coveries was accomplished in collaboration with Ph.D.
student Doug Clark (now an assistant professor and
NSF Presidential Young Investigator at the Univer-
sity of California, Berkeley). Clark and Bailey com-
bined classical activity studies with electron
paramagnetic spin resonance measurements of spin-
labeled immobilized enzymes. Using this technique,
they were able to establish direct evidence of active
site modification caused by immobilization and to show
the relationship between the degree of modification
and the activity of the enzyme.
Bailey's collaboration with Clark characterizes
Jay's philosophy of graduate education. He firmly be-
lieves that graduate students should participate heav-
ily in their research projects and in formulating
strategies for solving their problems. He observes,
"Graduate students are here to learn how to do re-
search-not just how to do experiments or calcula-


Jay enjoys weekly group seminars to discuss progress
and problems in the lab.
tions, but how to identify a particular problem within
an important general area. I encourage them to con-
sider the method of accomplishing a solution and its
feasibility as well as the ultimate impact that the so-
lution will have. Besides, since I've been fortunate to
work with many excellent students, it would be foolish
of me to presume that I have all the necessary ideas
and knowledge. I depend on my students to contribute
substantially to all aspects of their projects. On many
occasions, including Doug Clark's work, my students
have transformed my vague suggestions into highly
original and productive research."
Jay's work on periodic processes and biochemical
engineering during nine years on the faculty of the
University of Houston was recognized in 1979 by the
Allan P. Colburn Award of the American Institute of
Chemical Engineers. He came to Caltech in 1980. Ac-
cording to John Seinfeld, a close friend of Bailey's and
Caltech's Louis E. Nohl Professor and Professor of
Chemical Engineering, "I first met Jay about twelve
years ago at an AIChE meeting and we became
friends. Around 1979 I realized that Jay might con-
sider moving. He was quickly establishing himself as
one of the leading people in biochemical engineering
in the United States."
Jay has been good for Caltech, and the intimacy
and informality of the Caltech environment have ap-
parently been very good for him as well. "The biggest
qualitative change in my work came after I arrived at
Caltech. Here my students and I can learn from and


SPRING 1988









interact with some of the top people in biochemistry,
molecular biology, and cell biology, not to mention
chemical engineering. Cross-disciplinary collaboration
happens almost automatically at Caltech."
It was at Caltech that the techniques of genetic
engineering began influencing Jay's work. The revolu-
tion in recombinant DNA technology that began in
the late 1970s has been widely seen as the biologist's
show. But Bailey realized that if these techniques
were to realize their commercial potential, chemical
engineers would have to get intimately involved.
He also realized that modeling concepts from trad-
itional chemical reaction engineering could be effec-
tively applied in a new context-viewing a single cell


A sophisticated flow cytometer-cell sorter, here operated
by Dane Wittrup and Elaine Meilhoc, has supported
many PhD projects in Bailey's laboratory.
as a complex chemical reactor and transport ap-
paratus. Working in Jay's group, Sun Bok Lee (now
at the Korea Advanced Institute for Science and
Technology) and Steve Peretti (currently an assistant
professor at North Carolina State University) formu-
lated powerful new mathematical models that included
genetic level regulation.
Another consequence of Jay's interest in genetic
engineering was his collaboration with Caltech's
Judith Campbell, associate professor of chemistry and
biology, who studies basic mechanisms of DNA repli-
cation in yeast. Genetic engineers often use the tech-
nique of inserting recombinant plasmidss" into yeast
cells to get them to produce desired proteins in quan-
tity. Plasmids are relatively small, circular segments
of DNA that contain regulatory genes and genes that
code for proteins. There is one important problem
with foreign plasmids, however: As the recombinant
yeast cells divide, they tend to lose their plasmids
generation by generation, gradually becoming less
and less productive.
Bailey and Campbell have made a concerted effort
to understand this process using flow cytometry mea-
surements to provide unprecedented detail on single-


cell plasmid content in recombinant populations. In
flow cytometry individual cells are sent past a laser
beam, one by one, but very rapidly. In a process that
takes about one millisecond per cell, flow cytometry
permits determinations of cell size, shape, internal
structure, and the concentration of specific compounds
in individual cells. Many graduate students and post-
doctoral fellows have contributed to a growing body
of flow cytometry research in Jay's lab, and several of
these collaborators, including Friedrich Srienc at Min-
nesota and Jin-Ho Seo at Purdue, are now extending
these methods in their research labs.
Some of Jay's work in progress involves sharpen-
ing the focus of the genetic engineering studies to
more subtle and interesting problems. For example,
he's gotten interested in the details of what exactly
happens to a protein after synthesis. There are essen-
tially three possibilities: The protein can be degraded,
it can be exported from the cell, or it can remain in
the cell in solution or in an aggregate. If genetic en-
gineering is to fulfil its economic promise, these alter-
natives must be understood and the factors that con-
trol protein targeting must be better characterized in
quantitative terms.
Typically, genetic engineers concern themselves
with attempts to make cells produce certain proteins
in quantity. But chemical engineers are well aware of
the fact that proteins are not the only compounds of
economic importance made by cells. Realizing this,
Jay's group is concentrating on a brand new field that
Jay calls metabolic engineering. "Metabolic engineer-
ing involves the application of genetic engineering
techniques to alter or enhance the metabolism of
cells," Jay explains. "The goal is to overproduce spe-
cific amino acids, acetone, vitamins, certain biopoly-
mers, or other important chemicals made in cells.
Metabolic engineering requires a lot better under-
standing of what a cell is doing.
"First, we must understand all the pathways avail-
able to an organism for producing the target com-
pound and determine what additional pathways we
could make possible by introducing new enzyme ac-
tivities into that organism. Then, it's important to
judge which regulatory points along the pathway are
most important, since these determine the rate and
the selectivity for the target compound. After deter-
mining the most promising target enzymes, we must
clone the gene for that enzyme, attach the gene to a
suitable vector, and insert it into the organism. Then,
of course, we must investigate the effects of our gene-
tic manipulations on the production of the target com-
pound and its intermediates."
Jay's group has made significant advances in all


CHEMICAL ENGINEERING EDUCATION









steps of this process. In terms of understanding
metabolic pathways in detail, Alex Seressiotis, a
former student of his who's now at Columbia Univer-
sity, developed a computer program called MPS
(Metabolic Pathway Synthesis). The program contains
a database system that stores enzyme and substance
descriptions. "MPS can be used on a qualitative basis
to examine the effects of adding or deleting enzyme
activities to or from the cellular environment, to clas-
sify pathways with respect to cellular objectives, and
to extract information about metabolic regulation,"
Jay says. In an illustration of the power of MPS,
Seressiotis and Bailey had it consider the conversion
of pyruvate to the amino acid L-alanine. MPS came
up with a route to L-alanine that does not incorporate
the enzyme alanine aminotransferase, which is com-
monly assumed to be a required step for alanine
biosynthesis.
Jay's lab has the distinction of being one of the
first chemical engineering labs in the world with full
cloning facilities. 'The traditional way of producing
strains of an organism with desired properties was to
use a mutagenesis program," notes Bailey. This in-
volves exposing the cells to radiation or other muta-
gens and assaying the resulting organisms for the
property in question. "But this is a very random, time
consuming, and sloppy technique. By using recombin-
ant DNA technology we can make the whole process
far more rational. Furthermore, students with cloning
experience are more equipped to consider genetic as
well as process solutions to engineering problems.
And they'll be far better prepared to interact produc-
tively with molecular biologists in the future."
To investigate the effects of genetic manipulations
made in his lab, Jay's group has recently developed
novel data analysis methods for use with his new, 300
megahertz, wide-bore NMR. These methods allow im-
portant measurements to be accomplished simultane-
ously from a group of living cells. In a single experi-
ment, Jay's students can estimate intracellular pH,
the concentration of several key sugar phosphate
metabolic intermediates, and the concentrations of
adenosine di- and triphosphate (ADP and ATP).
At present, Bailey's lab is working on three pro-
jects in metabolic engineering. In the first of these
projects, the lab is attempting to enhance the uptake
of the sugar hexose in the yeast Saccharomyces cere-
visiae. Hexose uptake is the rate-limiting step for
growth in this economically important organism,
which is used for ethanol production and in the genet-
ically engineered production of other compounds. The
second project involves engineering the bacterium Es-
cherichia coli to produce ethanol by inserting genes


Jay explored the crater of M. Bromo in Java, Indonesia,
in April 1987.

for pyruvate decarboxylase and alcohol dehydro-
genase. In a third project, the group is trying to im-
prove production of ATP (the cell's energy currency)
in E. coli.
Not all chemical engineers agree with the direction
Bailey's research has taken. According to John Sein-
feld, there are still a few diehards out there who ex-
press skepticism about chemical engineers who get
involved with essentially biological techniques. But,
Seinfeld says, there are fewer and fewer naysayers.
"I think biochemical engineering is going to mature.
People are recognizing that it will take its place as a
firm field in chemical engineering and, in a sense, set-
tle down. It is difficult for biochemical engineering to
be as rigorous as some other fields of chemical en-
gineering that are closer to physics. That level of rigor
is being supplied by people like Jay Bailey. There will
be more biology in the education of future chemical
engineers as a result of his work."
The education of chemical engineers is a subject
that's important to Bailey. He's particularly proud of
the fact that many of his former graduate students
have become faculty members at important univer-
sities around the world. Frances Arnold, assistant
professor of chemical engineering at Caltech and
Bailey's wife says, "Jay takes a personal interest in
each of his students. He's always willing to sit down
with them and tell them the 'facts of life' of the chem-
ical engineering profession. His students show a lot of
loyalty as a result of his interest in them."
Important institutions to Jay's group of students
and postdocs are Friday afternoon group seminars
and, following that, the traditional "Ho-Ho" in the
Rathskeller of the Athenaeum, Caltech's faculty club.
"Gathering outside the lab for conversation and a few
Continued on page 102.


SPRING 1988









department


I Chemical engineering offices and laboratories are located in Perlstein Hall.


ILLINOIS INSTITUTE OF TECHNOLOGY


SATISH J. PARULEKAR
Illinois Institute of Technology
Chicago, IL 60616

THE DEPARTMENT OF Chemical Engineering at
the Illinois Institute of Technology (IIT), formally
established in 1904, is one of the oldest chemical en-
gineering departments in the nation. The develop-
ment of the department in the pre-1971 period was
previously described in this journal [1]. Dramatic
changes have occurred in the department since that
time in terms of faculty, undergraduate and graduate
programs, and research. All current chemical en-
gineering faculty, with the exception of one member,
were added after 1974. Since then, there has been a
significant expansion in research activities and re-
search areas pursued, with the current faculty having
expertise in traditional as well as emerging areas in

Copyright ChE Division ASEE 1988


chemical engineering. This is also reflected in the un-
dergraduate and graduate curricula, which have
changed significantly to meet the changing needs of
the profession.
IIT's main campus, internationally recognized as
an architectural landmark and located three miles
south of Chicago's downtown area and a mile west of
Lake Michigan, was designed in the 1940s by Ludwig
Mies van der Rohe, one of this century's most influen-
tial architects. The current university population on
the main campus includes about 6500 undergraduate
and graduate students and about 500 faculty. The
other campuses of IIT are located in downtown
Chicago and in the western suburbs of Chicago in
DuPage County. Chicago, one of the largest cities in
the world and a national and international center of
business and industry, offers students an exception-
ally wide variety of professional and cultural re-
sources. The close proximity of several chemical, food,
and allied industry research centers and production


CHEMICAL ENGINEERING EDUCATION










Since engineering is largely a team effort, we believe that development of the individual's ability to
work effectively as part of a team is important. To accomplish this, the laboratory courses
and the design courses involve teams of students laboratory sections are small and
a high level of personal contact between students and instructor is maintained .


plants provides for the IIT community an invaluable
source of close interaction with industry.
THE DEPARTMENT
The chemical engineering offices and laboratories
are located in Perlstein Hall. Presently, about 150 full-
time undergraduate students and 60 full-time
graduate students are enrolled in the department. In
addition, there are about 40 part-time undergraduate
students and 50 part-time graduate students. Three
graduate degrees are offered by the Chemical En-
gineering Department: Master of Science (MS), Mas-
ter of Chemical Engineering (MChE), and Doctor of
Philosophy (PhD). The department annually awards
nearly 40 Bachelor of Science, 10 MS, 10 MChE, and
8 PhD degrees.
Over the years, the department has graduated
many competent engineers, a number of whom have
obtained significant national prominence in their pro-
fessional careers. Several of our alumni have held na-
tional office in the American Institute of Chemical En-
gineers. Although a majority of our alumni have pur-
sued professional careers in industry, we are proud
that, over the years, a significant number have joined
chemical engineering faculties at major institutions.
RESEARCH
The research areas pursued in the pre-1971 period
were chemical reaction engineering, separation pro-
cesses, thermodynamics, and transport and interfacial
phenomena [1]. In recent years, many additional areas
of research have been initiated and strengthened.
They include biochemical engineering, biomedical en-
gineering, combustion, energy technology, enhanced
gas and oil recovery, and process dynamics and con-
trol. Interdisciplinary research has been a strong
tradition at IIT and several joint initiatives with other
departments have evolved to address problems in the
critical emerging areas. Some of the faculty have
cooperative research activities with the Institute of
Gas Technology, IIT Research Institute, and Argonne
National Laboratory. All such interactions provide an
intellectually stimulating environment for the
graduate students. The development of IIT's very
strong research and educational center in food
technology at the Moffett Technical Center (located in
the Chicago area) will be completed this summer.


Some of the chemical engineering faculty will be
closely associated with this center.
It is interesting to look at the current faculty in
chronological order of joining the chemical engineer-
ing department at IIT in order to trace the growth of
our program in recent years.
Darsh Wasan, currently acting Dean of the Ar-
mour College of Engineering, is the oldest faculty
member in terms of service, having joined IIT in 1964.
He served as department chairman from 1971 to 1987.
His research deals with interfacial and colloidal phe-
nomena, enhanced oil recovery, and separation pro-
cesses. Darsh and his research team are developing
experimental techniques to accurately measure sur-
face and interfacial theological properties. This re-
search has significant utility in control of a variety of
dispersed phase systems and in study of the mecha-
nisms of oil bank formation and propagation in chem-
ical flooding processes involving surfactants and al-
kaline agents, foams, emulsions, and polymers for mo-
bility control. Research in separations is concentrated
on separation of fine particles from non-aqueous media
based on electrokinetic phenomena, emulsification/de-
mulsification processes and thin liquid film phenome-
na. Darsh is a recipient of the Western Electric Fund
Award of the ASEE, the Hausner Award of the Fine
Particle Society, and the Special Creativity Award of
the National Science Foundation.
The mid-1970s saw the addition of two of the cur-
rent faculty members, Rob Selman and Dimitri Gidas-
pow. Rob Selman, currently acting chairman of the
department, joined the faculty in 1975. His research
interests are electrochemistry and electrochemical en-
gineering with a special emphasis on high-tempera-
ture processes. Rob and his students are currently
investigating the formation, growth, and corrosion of
dendritic metal deposits, a problem common in the
charging of zinc-based aqueous batteries which are
being developed for electric vehicles and load leveling.
Other current research projects deal with fundamen-
tal aspects of molten carbonate fuel cells and develop-
ment of porous-electrode models for these fuel cells.
The characterization of micro-emulsions by AC impe-
dance is being investigated to assess the effectiveness
of electrochemical oxidation as a means of breaking
micro-emulsions. Other research projects under Rob's
direction deal with molten salt processes involving


SPRING 1988










carbon cathodes and amorphous metal deposition from
molten salts.
Dimitri Gidaspow joined the faculty in 1977.
Dimitri's research interests are in the areas of
mathematical modeling and analysis of various energy
conversion processes. His current research involves
studies of the hydrodynamic theories of fluidization
and gas-solids transport. Generalizations of Navier-
Stokes' equations are being solved using supercom-
puters to predict phenomena such as cluster forma-
tions in risers and bubble motion in fluidized beds. In
collaboration with Darsh Wasan, he also conducts re-
search in the areas of separation of colloidal particles
and electrostatic desulfurization of coal. Dimitri has
been recognized for his research through the Donald
Q. Kern Award of the AIChE and the Special Creativ-
ity Award of the National Science Foundation.
The bulk of the present faculty members arrived
on the scene in the 1980s. Richard Beissinger joined
the chemical engineering department in 1981.


I---
Taha Alkhamis and Richard Beissinger working with a
Weissenberg rheogoniometer for blood viscosity mea-
surement.

Richard's research is concerned with transport phe-
nomena in biological systems. His current research
activities are in blood-artificial surface interactions,
biorheology, and development of artificial red blood
cells. He also conducts research in the development of
implantable drug infusion systems and in phar-
macokinetics. Some of his current research projects
are: augmented mass transport in sheared suspen-
sions; macromolecular adsorption to solid surfaces;
liposome-encapsulated hemoglobin and hemoglobin-in-
oil-in-water multiple emulsion droplets as red blood
substitutes; and the effects of red blood cells on
platelet adhesion and aggregation in laminar shear


Jill Weldon and Selim Senkan discussing experiments
with a flat flame burner in the combustion research lab-
oratory.

flow. Rich is also directing development of the inter-
disciplinary Polymer Science and Engineering pro-
gram.
Ali Cinar, who arrived in 1982, conducts research
in process control and dynamics. Control strategies
for multivariable chemical processes are being de-
veloped and tested using experimental systems and
real-time microcomputers. One of his research pro-
jects deals with developing methods for selecting
robust operating configurations and improved control
strategies using a pilot plant consisting of two cataly-
tic tubular reactor beds with internal heat exchange
and a feed-effluent heat exchanger. Ali and some of
his graduate students are also investigating forced
periodic operation of chemical reactors using the vi-
brational control approach. Their work is focused on
stabilization of exothermic continuous stirred tank
reactors and tubular packed bed reactors as well as
on improvement in selectivity and yield of complex
reactions. Another of Al's research projects deals
with development of expert systems for fault-tolerant
computer control of complex processes, such as multi-
bed autothermal reactors.
Selim Senkan joined IIT in 1982. His research is
in the areas of combustion and high-temperature
chemical reaction engineering as applied to problems
in energy conversion, propulsion, and environmental
protection. One of his research projects deals with ex-
perimental and theoretical investigation of oxidation
and pyrolysis of chlorinated hydrocarbons (CHCs)
using model compounds. In the experimental pro-
gram, stationary flames of selected CHCs are probed
for the determination of species and species profiles


CHEMICAL ENGINEERING EDUCATION










using supersonic jet sampling coupled with on-line
molecular beam mass spectroscopy (MBMS). An out-
come of this research is the recently patented,
economic process for conversion of methane into
acetylene and ethylene. Because of low NO, pollutant
emissions, catalytic combustion is an attractive alter-
native to flame combustion when very high tempera-
tures are not required. Selim and his group are also
involved in the study of the impact of halogens on
catalyst development, reaction rates, and selectivity
in catalytic combustion processes.
William Weigand joined the IIT faculty in 1983.
His research interests are biochemical engineering
and process control. Bill and his research group are
involved in kinetic modeling of cell growth and prod-
uct formation for microorganisms which produce pri-
mary and secondary metabolites. The optimal operat-
ing procedures are then derived and examined with
the aid of computer simulations and experiments car-
ried out with highly-instrumented, computer-coupled
fermentors. Bill is also interested in development of
new sensors and the use of estimation techniques
which permit optimal operation in the absence of com-
plete variable measurement. In the area of process
control, techniques for control in the presence of mod-
eling error, for changing system dynamics, and for
nonlinear interacting systems are being developed.
Hamid Arastoopour and Satish Parulekar joined
the department in the fall of 1985. Hamid Aras-
toopour has research interests in the areas of multi-
phase flow, flow in porous media and unconventional
gas reserves, and fossil fuel conversion processes.
Current research projects being investigated by
Hamid and his graduate students are hydrodynamic
analysis of pneumatic conveying of solids, numerical
analysis of single and multiphase flow in unconven-
tional reserves, analysis and measurement of the
agglomeration of sticky particles and heat transfer in
fluidized beds, and analysis and measurement of tran-
sient gas and condensate flow in gas transmission and
distribution systems.
Satish Parulekar conducts research in biochemi-
cal engineering and chemical reaction engineering.
His research in biochemical engineering deals with
production of extracellular and intracellular enzymes/
proteins by recombinant and wild-type microor-
ganisms and production of biochemicals using im-
mobilized cell reactors. The research with recombin-
ant organisms is aimed at gaining a fundamental un-
derstanding of host-plasmid interactions in these
species and the effect of these interactions on plasmid
stability, cell growth and product gene expression.
The research with wild-type microorganisms is di-


rected towards understanding the mechanism of syn-
thesis of extracellular enzymes such as amylases and
proteases. The research dealing with immobilized cell
reactors is focused on study of alteration in cellular
metabolism due to immobilization and implications of
such alteration in design of these reactors. His re-
search in chemical reaction engineering is concerned
with identification of optimal reactor structures for


Prasad Davuluri and Hamid Arastoopour studying tran-
sient gas transport with a Laser Doppler velocimeter.

complex reaction networks and experimental and
theoretical investigation of forced periodic operation
of continuous flow (CSTR and tubular) reactors.
Henry Linden joined the faculty in the spring of
1987 as the F. W. Gunsaulus Distinguished Professor
of Chemical Engineering. Henry has had an excep-
tionally illustrious career in research in fuel technol-
ogy at both the Institute of Gas Technology and the
Gas Research Institute and was president of each of
these institutions for several years. His distinguished
career has led to membership in the National
Academy of Engineering, as well as to numerous hon-
ors and awards such as the Distinguished Service
Award from the American Gas Association. Having
Henry on board will give us the benefit of his wide-
ranging experience and insight in energy-related re-
search.


SPRING 1988









UNDERGRADUATE PROGRAM
The mission of the undergraduate program in
chemical engineering at IIT is to prepare our students
for their professional careers and to enable them to
develop the technology of the future. There is enough
flexibility in the program for each student to tailor an
individualized curriculum to satisfy his or her particu-
lar scientific or technical interests. The first two years
are devoted to the fundamental sciences, mathema-
tics, and engineering sciences and are particularly
concerned with the development of professional skills.
In addition to developing engineering competence, the
program examines the economic and societal implica-
tions of chemical engineering. The required courses
are material and energy balances, and unit opera-
tions-I in the sophomore year; thermodynamics, trans-
port phenomena, unit operations-II, and chemical en-
gineering laboratory-I in the junior year; and chemical
reaction engineering, process control, chemical pro-
cess design (two semesters), and chemical engineering
laboratory-II in the senior year.
In addition, there are elective courses in bioen-
gineering, computer applications in chemical en-
gineering, colloidal and interfacial phenomena, elec-
trochemistry, energy technology, food technology,
microelectronics fabrication, and polymer processing.
Students interested in gaining professional specializa-
tion may include some of these courses in their cur-
riculum to earn a specialized minor in one of the fol-
lowing areas: biotechnology, computers in chemical
engineering, energy technology, food technology, and
polymer engineering.
Since engineering is largely a team effort, we be-
lieve that development of the individual's ability to
work effectively as part of a team is important. To
accomplish this, the laboratory courses and the design
courses involve teams of students. Since individual in-
struction is so important to students' growth, labora-
tory sections are small and a high level of personal
contact between student and instructor is maintained.
Some of our best students also choose to work on inde-
pendent research projects during their junior and/or
senior years, which prepare them for graduate re-
search.

GRADUATE PROGRAM
The coursework for MS and MChE degrees must
include at least four of the following six core courses:
chemical reaction engineering, fluid mechanics, heat
transfer, mass transfer, process control, and ther-
modynamics. For a PhD degree, coursework in all six
areas is required so that the students will be equipped


At IIT we are developing courses to integrate
these emerging areas into our curricula while, at
the same time, leaving the emphasis on science
and engineering fundamentals unchanged.

to apply advanced principles from the entire spectrum
of chemical engineering irrespective of their research
specialization. A student pursuing an MS must com-
plete eight credit hours in research and thesis work.
The MChE is a professionally-oriented degree pro-
gram which permits a concentration in engineering
practice. The requirements are the same as those for
the MS degree, except that additional courses and/or
a project replace the eight credit hours of thesis.
Every prospective PhD candidate must take a qualify-
ing examination to determine fitness and aptitude for
further graduate study. Research qualities are judged
during the oral comprehensive examination over the
student's dissertation proposal, taken some time after
admission to candidacy. After successfully clearing
this examination, the candidate pursues the selected
research program in consultation with the research
adviser and advisory committee.
In addition to the core courses, IIT offers a wide
variety of elective graduate courses which in the past
three years have included biochemical engineering,
catalysis, computational techniques, electrochemical
engineering, polymer processing, reservoir engineer-
ing, separation processes and transport phenomena in
living systems. Many of the graduate courses, particu-
larly the core courses, are televised through the IIT/V
network. Remote centers for reception of these tele-
casts are located near several of the industrial centers
within a fifty-mile radius from the IIT campus for the
benefit of part-time students.

THE FUTURE
As the chemical engineering profession changes
and adapts to the technology needs of the next decade
and the next century, so must chemical engineering
education. In the coming years, the curriculum con-
tent will be modified to accommodate new technology
areas in chemical engineering. At IIT, we are develop-
ing courses to integrate these emerging areas into our
curricula while, at the same time, leaving the em-
phasis on science and engineering fundamentals un-
changed. As in the past, thorough training of the un-
dergraduate as well as graduate students will continue
to receive the highest priority.

REFERENCES
1. Kintner, R. C., D. T. Wasan, Chem Eng. Ed., 5, 108 (1971). [


CHEMICAL ENGINEERING EDUCATION








AIChE JOURNAL




Morton M. Denn, Editor Published Monthly


The AIChE JOURNAL, a monthly publication of the American Institute of Chemical Engineers,
is devoted to fundamental research and developments having immediate or potential value
in chemical engineering.


Increased to. 12 issues in 1985, this journal now provides more comprehensive coverage of
ongoing research and developing technologies in chemical engineering. As a permanent
repository of innovative processes, efficient analyses of data and current theoretical ideas,
it serves as an excellent reference source for researchers. Original papers are reviewed by
a board of peer scientists and engineers. Contributors come from industry, government
and university research groups in the U.S. and throughout the world.


Range is broad and includes such varied topics as: heat transfer, mass transfer, fluidization,
kinetics, adsorption, process dynamics, membrane technology, separation processes,
solid-liquid systems, reactor technology, electrofaction, polymer processing, filtration,
crystallization. Coverage impacts upon chemical, environmental and biotechnological
engineering.


1988 Annual Subscription Prices: AIChE Members $40 Others $275 Foreign Extra: $13
Subscriptions are on a calendar-year basis





Send Orders to: AIChE Subscription Dept. A, 345 East 47 Street, New York NY 10017. Prepayment in
U.S. currency is required (check, VISA, MasterCard, international money order, or bank draft drawn
on a New York bank). Members may order only one subscription at member price and must indicate
membership number when ordering. Credit card customers: Please indicate "VISA" or "MasterCard,"
include card number, expiration date, printed name of cardholder and signature.


AMERICAN INSTITUTE OF CHEMICAL ENGINEERS










i laboratory


A PILOT-SCALE HEAT RECOVERY SYSTEM


FOR COMPUTER PROCESS CONTROL


TEACHING AND RESEARCH


P. J. CALLAGHAN, P. L. LEE
and R. B. NEWELL
University of Queensland
St. Lucia, Qld, Australia 4067

N THE LAST two decades efficient energy utilization
and process integration have become increasingly
important to chemical plant economics. Plant design
for these factors has meant higher levels of process
interaction, often with positive feedback, which re-
quire improved control. A typical example is the heat
recovery circuit in reactor or distillation systems.
Process equipment used in the laboratory to inves-
tigate the control problems created by tighter process
integration is more convenient. This can be sum-
marized as

Lower cost of pilot-scale equipment, and
Loss of production and or product quality are not issues
during process experimentation.

However, instrumentation and control equipment
is "off the shelf" as is found in industry and hence is
a significant cost factor. The aim in building process
control rigs for undergraduate teaching and graduate


The pilot-scale heat recovery system is a valuable
teaching and research tool. It is sufficiently flexible
to demonstrate basic principles and yet sufficiently
complex to demonstrate common process control
problems such as nonlinearities and interactions
between variables.

research is to create rigs that are

Flexible enough to demonstrate a number of basic princi-
ples
Sufficiently complex to demonstrate common process con-
trol problems such as nonlinearities and interactions be-
tween variables
Economical

Due to the important role of real time computing
in process control, it is essential to provide teaching
and demonstration facilities of such techniques. In ad-
dition, most advanced control strategies require the
use of process control computers for successful im-
plementation.


TABLE 1
Equipment Details


FIGURE 1. Schematic diagram of pilot-scale heat recov-
ery circuit.
C Copyright ChE Division ASEE 1988


Valves V1 and V2

Valve V3
Double pipe exchanger


Plate heat exchanger
Rotameter
dp Cell
Pressure transmitter PIT1
Steam supply pressure to
experimental apparatus
Pressure transmitter PIT2


Fisher Control Valves, size 34
and 30 respectively
Badger Meter Inc, size B
2.43m long, 13mm diameter
copper pipe, 38mm diameter
copper jacket
Alpha Laval P20-HB
Metric Series 18 with S.S. plug
Taylor Instuments, 0-5"H20
Beckman, 0-500 kPa
0.2MPa

Honeywell, 0-500 kPa


CHEMICAL ENGINEERING EDUCATION


i









PILOT-SCALE HEAT RECOVERY CIRCUIT

A schematic diagram of the laboratory rig is shown
in Figure 1. Mains water is preheated in a pre-heat
plate heat exchanger, HE1. After preheating it flows
through a steam heated double pipe heat exchanger,
HE2. After the double pipe exchanger the water splits
either to be used in the preheat plate heat exchanger
or to bypass the heat exchanger. When the streams
rejoin they are fed to a constant head drain. A control
valve, V3, is installed in the bypass line and may be
used to regulate the bypass flowrate. Equipment de-
tails are given in Table 1.
There are two manipulated variables in the sys-
tem: a control valve in the steam line, V1, and a con-
trol valve, V2, in the hot line after the double pipe
exchanger leading to the preheat plate exchanger.
They can be used to control any of the dependent vari-
ables, but typically the final mixed temperature, TT3,




















P. J. Callaghan received his BEChem and his MEngSci degrees
from the University of Queensland in 1984 and 1986. His Master's
programme centered on the development and use of a predictive con-
trol algorithm for the heat exchange system described in this paper.
He is currently employed by Akoa Ltd. as a control engineer. (Not
pictured)
P. L. Lee received his BEChem from RMIT in Melbourne and his PhD
from Monash University in 1980. He worked in the design and commis-
sioning of computer control systems for both continuous and batch
plants for three years before coming to Queensland. His early research
was on the control of the unstable steady state in an exothermic CSTR.
His interests include multivariable self-tuning and adaptive control of
fermentation and heat recovery systems. (L)
R. B. Newell received his BScApp and BEChem from Queensland
and his PhD from the University of Alberta. He also has a DipEd in
Tertiary Education from Monash. He joined the staff at Monash Univer-
sity in 1974 and moved to Queensland in 1980. His early research
was in the multivariable control of a pilot plant evaporator, unstable
steady state control in a CSTR, and multilevel hierarchical optimization.
Current interests include optimization of the Australian oil refinery and
transportation system, combined fuzzy and deterministic control, and
selftuning and adaptive control of heat recovery systems. (R)


and the temperature out of the double pipe exchanger,
TT2, are used. A standard differential-pressure
bypass control loop is usually implemented to maintain
the pressure drop across the plate heat exchanger con-
stant.

PROCESS-CONTROL SYSTEM INTERFACE

An interface box between the process instrumen-
tation and the control system is located on the rig.
This box displays all rig variables, both manipulated
and controlled, i.e., the steam pressure, the hot side
water flowrate to the plate heat exchanger, and the
three temperature measurements and the valve posi-
tions. The interface box is supplied with 240V AC
power and supplies 24V DC for instruments and trans-
ducers. Standard signals are 0-5V for control signals
and 1-5V for measurements. The interface box also
has the facility to switch the manipulated variables
from remote to local, by interrupting the signal from
the control system and supplying a manual signal.
Figure 2 shows the responses in the exit tempera-


20 40 60 80 10 129 140 16t 188 288
Sample
(a)


28 40 69 869 10 120 148 10B IS0 260
Sample
(b)


FIGURE 2. Step test responses for TT2 due to a 10% step
(a) up and (b) down in VI after sample 20.


SPRING 1988









ture to 10% step changes up and down in the steam.
The sample time during the collection of this data was
five seconds. Note that both magnitude and shape of
responses differ, illustrating non-linear behaviour.
Also the need for filtering is apparent from this raw
data.

CONTROL SYSTEM EQUIPMENT
Basic Data Collection and Regulatory Control
The control system used to monitor and control
the pilot-scale apparatus is arranged as shown in Fig-
ure 3. This equipment is arranged in a hierarchical
manner, consisting of two layers: a primary data col-
lection and regulatory control level, and an advanced
control level.



IBM Supervisory







Regulatory
BAILEY NETWORK 90
Level







PROCESS



FIGURE 3. Control system equipment.


The primary data collection and regulatory control
functions are performed using a Bailey Network 90
distributed control system. The heart of the system is
a "multi-function controller," which is a micro-proces-
sor based controller. It performs all of the regulatory
control functions and coordinates the operation of the
analog-to-digital input and digital-to-analog output
cards. Operator interface is provided by means of a
color CRT display. This display is capable of showing
schematic representations of the pilot-scale equipment
with measurement data displayed, controller faceplate
displays, trend plots of measurement inputs, and com-
binations of the above types of displays.


Advanced Control Function
The advanced control functions, such as imple-
menting multivariable predictive controllers are per-
formed using the IBM Advanced Control System
(ACS) previously described in this journal by Koppel
and Sullivan [1]. However, this application uses ACS
in a real-time environment.
The University of Queensland installation consists
of an IBM 4341-2 mainframe with 8 Mbytes of mem-
ory, 12 terminals and 2 printers in two clusters, line-
printer, 1.8 Gbytes of disc storage, a tape drive and
a Series/1 minicomputer with direct process I/O and
connections to the Bailey Network 90 system.

UNDERGRADUATE COMPUTER-BASED PROCESS
CONTROL INSTRUCTION
The experimental rig has a number of interesting
characteristics that make it particularly suitable for
teaching real-time process control:

Non-linear behaviour. This is shown in Figure 2, which
illustrates the different time constants for a step increase
and a step decrease in the steam supply valve Vl. The time
constants are approximately 96 and 45 seconds respec-
tively.
Interacting behaviour. This is illustrated by examination
of the relative gain array for this process shown in Table 2.
The inherent noise in the measurements, again illustrated
in Figure 2.

The following teaching experiments utilize this rig:

1. Examination of digital filtering. Students are
asked to examine the effects of different filter
bandwidths and different filter types (exponential,
Butterworth, union). This short experiment is used
in teaching real-time computing and would be ex-
pected to take one hour of laboratory time.
2. A cascade loop controls the double pipe exit
temperature with a slave loop controlling the pres-
sure in the steam jacket via the steam valve. This ex-
periment only utilizes the Bailey Network 90 system
and students are expected to complete this experi-
ment in one and a half hours.
3. Feedforward control. A feedforward controller
compensates for disturbances in the feed flowrate by
adjusting both the steam and water valves. This ex-
periment, combined with a model identification exer-
cise usually takes about three hours of laboratory
time.
4. Decoupling control. Two feedback loops or two
interacting feedback loops are used as a basis for in-


CHEMICAL ENGINEERING EDUCATION









producing the concept of process interactions. The
students are then expected to design and implement
a decoupled controller using lead-lag compensators.
This experiment normally takes six to nine hours to
complete.
5. Adaptive control. An adaptive single loop con-
troller adjusts the steam valve to control the exit
temperature. This experiment, used in a fourth-year
elective control course, makes use of the ACS system
and would normally involve fifteen hours of labora-
tory time.

POSTGRADUATE PROCESS CONTROL RESEARCH
Current research utilizing the rig is on a predictive
control design technique described by Maurath, et al.
[2]. The basic predictive control algorithm uses a
model to predict the output for a number of future
moves due to the previous inputs. An error is formed
by subtracting the prediction of the output from the
setpoint. This error vector is then used to calculate a
change in the manipulated variables. The design tech-
nique being tested is to use a singular value decompo-
sition to condition the Dynamic Matrix [2] of step re-
sponses. Results from this study are reported else-
where [3].
Future work on the rig will include fuzzy identifi-
cation and control. A model-based controller designed
around a fuzzy model of the process will be used to
control the rig. Fuzzy identification techniques will be
used off-line to derive an initial model, and on-line to
provide continuous updating of the model. The on-line
use of fuzzy identification turns the model-based con-
troller into a fuzzy adaptive controller [4, 5, 6].
A reactor heat recovery system is a possible exten-
sion to the current rig. This is achievable by using an
exponential function on the exit temperature of the
double pipe exchanger to move the steam valve. This
allows for simulation of variable heats of reaction
without the danger and cost of reactants. In this mode
only one manipulated variable is possible, i.e., the
water valve (V2).

CONCLUSIONS AND SIGNIFICANCE
The pilot-scale heat recovery system is a valuable
teaching and research tool. It is sufficiently flexible to

TABLE 2
Relative Gain Array
TT2 TT3
VI 0.22 0.78
V2 0.78 0.22


demonstrate basic principles and yet sufficiently com-
plex to demonstrate common process control problems
such as nonlinearities and interactions between vari-
ables.
The process control system equipment allows stu-
dents to obtain experience in real time computing, at
several process control functional levels.

REFERENCES
1. Koppel, L. B. and G. R. Sullivan "Use of IBM's Advanced
Control System in Undergraduate Process Control Education,"
Chem. Eng. Educ. XX, No. 2, pp. 70-73, 106-07, 1986.
2. Maurath, P. R., D. E. Seborg, and D. A. Mellichamp, "Predic-
tive Controller Design by Principal Components Analysis,"
Proc. Acc., 1985, pp. 1059-1065.
3. Callaghan, P. J., M.Eng.Sc. Dissertation, University of
Queensland, 1986.
4. Graham, B. P. and R. B. Newell, "Computer Control of Dif-
ficult Processes," Chemeca 84, Vol 2, pp. 743-749 (1984).
5. Graham, B. P., P. L. Lee, and R. B. Newell, "Simulation of
Muitivariable Control of a Cement Kiln," Chem. Eng. Res.
Des., 63, Nov. 1985.
6. Graham, B. P., R. B. Newell, G. P. Le Page, and L. C.
Stonehouse, "Industrial Applications of Fuzzy Identification
and Control," Chemeca 86, pp. 221-226 (1986). D


letters

UPDATED REFERENCES

Dear Editor:

In a recent paper, "A Course in Mass
Transfer with Chemical Reaction," Chem. Eng.
Ed., 21, No. 4, 164 (Fall 1987), W. J. Decoursey
refers to Gas Liquid Reactions by P. V. Danck-
werts as a source in English for the seminal pa-
pers by S. Hatta which were published in
Japanese [Technol. Repts. Tokoku University, 8, 1
(1929); 10, 613,630 (1931); and 11, 365, (1932)].
Complete translations of the latter articles have
been published in International Chemical Engi-
neering, 18, 443-475 (1978). The readers of this
article may also be interested to know that a
translation of the closely related pioneering paper
by G. Damkohler, "The Influence of Diffusion,
Fluid Flow and Heat Transport on the Yield in
Chemical Reactors," Der Chemie-Ingenieur, 3,
Part I (1937), has recently been published in
International Chemical Engineering, 28, 132-198
(1988).

Stuart W. Churchill
University of Pennsylvania


SPRING 1988










In Memoriam

James J. Christensen
1931-1987


Dr. James J. Christensen's life was devoted to ser-
vice: service to wife and family; service to his profes-
sion in chemical engineering and to the field of ther-
modynamics; service to his university, the Brigham
Young University and to his students, both graduate
and undergraduate; and service to his church and com-
munity. This life of service was brought to a close by
a sudden heart attack on September 5, 1987.
Jim's worldwide recognition as an outstanding
thermodynamicist and a developer and builder of
state-of-the-art calorimeters gave him and his wife a
great opportunity to travel around the world. An out-
standing feature of the family room of their home is a
large map with the continents shown in bas-relief cut
from plywood. The map is studded with pins, indicat-
ing the many places they have gone where they had
the privilege of meeting new people and of teaching
his methods to many. Jim and Virginia shared each
other's life in this way as well as in taking dance clas-
ses together and in participating in the great books
program over the last 28 years. He was dedicated to
his four sons and a daughter and spent many hours
with them in such activities as hiking. One of his sons
followed him into the chemical engineering profession.
Although Jim loved his professional career, he loved
service to his family equally well.
Jim's service to the profession in the field of
calorimetry and thermodynamics is unparalleled. He
received a BS degree in 1953 and an MS degree in
1955 in chemical engineering from the University of
Utah. He then went on to complete his formal educa-
tion with a doctor's degree from Carnegie-Mellon in
1958. Part of his research work involved heat transfer
to coils which later showed up in the design of the
first-class calorimeters which he built. These


calorimeters have been developed commercially, are
sold around the world, and are a standard for making
precise calorimetry measurements. One of Jim's early
ambitions was to build a calorimeter precise enough
to measure the germination of a single seed but his
designs greatly exceeded this. The early designs were
crude compared with the later designs, but still re-
quired corrections for the energy input from a stirrer
motor. His latest design was a calorimeter which
would measure the heat of mixing at high pressure
and high temperature.
Professor Christensen was also a prolific author.
He has written or edited twelve books and co-au-
thored fifteen major review articles. He has also pub-
lished over 250 research articles in a wide variety of
journals. He was instrumental in the founding of inter-
national conferences dealing with the chemistry of
macro-cyclic molecules. He has been recognized by his
peers, who selected him for the Sigma Xi annual lec-
ture, a Blue Key lecture, and the Distinguished Fac-
ulty Annual Lecture at Brigham Young University.
He also received the Maiser Outstanding Research
Award and the Huffman Memorial Award as an out-
standing thermochemist. Last August he was asked
by the Chemical Engineering Division of the Amer-
ican Society of Engineering Education to give the 3M
lecture which is based on outstanding research and
contribution to the profession. (This lecture is sched-
uled for publication in a future issue of CEE).
Jim had a genuine love for his students, both
graduate and undergraduate, and they returned that
love through extra effort. He was recognized three
times by the students as being the outstanding lec-
turer in the department, and he was recognized by
the college as the outstanding teacher with the univer-
sity's Maesar Outstanding Teaching Award (which is
the counterpart of the outstanding research award.)
Jim took great delight in collecting "toys" which
demonstrated the principles of thermodynamics and
used them continually to interest the students. They
included the familiar bird dunking its beak into water,
rotating pinwheels, and his latest acquisition, a motor
which worked by dipping one end in hot water. Jim's
favorite courses were thermodynamics and creativity,
and he exhibited creativity in teaching. He taught the
students new ways of approaching old problems. He
was dynamic in his presentations and gave an ani-
mated demonstration of the energy content of gas
molecules by running back and forth, twirling and
jumping up and down, colliding with students and
walls. Almost every year during the student banquet,
this little demonstration was repeated by the students
as a sign of their regard. His annual backpacking trip


CHEMICAL ENGINEERING EDUCATION









to the Uintah mountains with the graduate students
was a highlight of each year.
He also served his community and church, the
Church of Jesus Christ of Latter-Day Saints, in many
ways. He was active in the education of his children
and in the processes by which parents can contribute
(P.T.A., etc.). He participated in great book discus-
sion groups, he was a cub master for eight years, and
demonstrated his love of teaching by teaching adult
priesthood groups.
Jim's life of personal service has concluded. But
his influence will continue to be felt in the lives of his
family, his students, and his friends for many years to
come.
Dee H. Barker


Roland Andrew Ragatz
1898-1987


Roland Andrew Ragatz, Professor Emeritus of
Chemical Engineering at the University of Wisconsin,
was born in Prairie du Sac, Wisconsin, in 1898, and
died in Madison, Wisconsin, on May 30, 1987, at the
age of 88. He is survived by his wife Nancy, who he
married in 1930, a son Andrew (Ellen), a daughter
Karen Roberts (Burnell), and seven grandchildren.
Professor Ragatz graduated from Prairie du Sac
High School in 1915. He completed his studies for a
BS degree (in the then new discipline of chemical en-
gineering) in 1920 at the University of Wisconsin,
Madison, and was immediately appointed instructor
in chemical engineering. He remained on the teaching
staff at Wisconsin until his retirement in 1969, except
for a one-year leave of absence in 1929-30. While serv-
ing as a full time instructor, Professor Ragatz earned
his MS degree in 1923 and his PhD degree in 1931,
also in chemical engineering. He served as depart-
ment chairman during three periods, totalling sixteen


years, and as associate chairman for five years there-
after.
Professor Ragatz' 49-year period of service to the
University is one of the longest on record. He contri-
buted significantly to his college and department and
to his students during the important early period
when the developing field of chemical engineering was
emerging from an empirical discipline and evolving
into a more scientifically based study. Basically de-
voted to teaching, Roland's areas of interest followed
the needs of the department and its students in pre-
paring them for industrial careers in a variety of often
traditional manufacturing enterprises. In 1920 he
began the development of introductory and advanced
metallography courses, including laboratory studies
related to the microscopic structures of metals, with
emphasis on materials of construction. When instruc-
tion in this area was transferred to the Department
of Mining and Metallurgy in 1948, Professor Ragatz
turned his interest in materials to developing courses
in plastics, again including a laboratory program. This
early attention to materials science established the
foundation on which the department's current strong
program is based.
Starting in 1935, Professor Ragatz joined the late
Olaf A. Hougen (and, during the 1940s, Kenneth M.
Watson) in giving courses in material and energy bal-
ances and in thermodynamics for chemical engineers
and participated in rewriting the corresponding parts
(I and II) of the text Chemical Process Principles,
which long has been a standard work in chemical en-
gineering.
During his extended service as department chair-
man, Roland alternated with Olaf Hougen, and, in the
words of Professor Hougen, "he assembled a staff well
balanced in special talents for teaching and research,
for undergraduate and graduate instruction, for vari-
ety in engineering and scientific interests, and with
balance in laboratory and classroom instruction."
Roland also served long and efficiently as chairman
of the College of Engineering committee on fellow-
ships and scholarships, as well as on numerous other
committees. He was a long-time member of the
ASEE, serving on its Council, as chairman of its
Chemical Engineering Division, and as organizer of
the important Summer School for Teachers of Chemi-
cal Engineering (at Madison in 1948).
Professor Ragatz exemplified the ideals of the fac-
ulty of the University of Wisconsin: scholarship, re-
search, public service, and above all, teaching.
Roger J. Altpeter
Wayne K. Neill
Charles C. Watson


SPRING 1988










curriculum


SAFETY AND LOSS PREVENTION

IN THE UNDERGRADUATE CURRICULUM

A Dual Perspective


DANIEL A. CROWL
Wayne State University
Detroit, MI 48202
and
JOSEPH F. LOUVAR
BASF Corporation
Wyandotte, MI 48192

R RECENT CHEMICAL PLANT accidents in India and
in Switzerland have increased public and indus-
try-wide concern and awareness of safety and loss pre-
vention. This is in spite of the fact that the chemical
process industries are safer than most other industries
[1]. However, due to the potential for a serious plant
accident, both the public and industry recognize that
safety must be increased.
The AICHE is taking considerable steps to im-
prove safety. First, it has formed a Center for Chem-
ical Process Safety with the charge to "address the
concerns about the handling of toxic or reactive mate-
rials and the safety of plant operating procedures" [2].
Second, it has formed the Design Institute for
Emergency Relief Systems (DIERS) User Group to
continue the cooperative industrial activities to extend
the DIERS technology [3]. And third, a Task Force
on Safety and Health in the Undergraduate Cur-
riculum has been formed under the Safety and Health
Division of the AICHE with the major objective to
"identify the key concepts of loss prevention, safety,
and health which should be considered essential for
accreditation of the curriculum by 1990" [4]. As a re-
sult of this new awareness and concern for safety, it
is apparent that safety and loss prevention will be-
come a part of the future undergraduate chemical en-
gineering curriculum.
Most undergraduate chemical engineering cur-
ricula in the United States contain little in the way of
safety and loss prevention. In fact, university
laboratories are, typically, serious safety offenders.
Copyright ChE Division ASEE 1988


Great Britain has had an ambitious safety and loss
prevention program in their undergraduate cur-
riculum for some years now. This is a result of a sub-
stantial chemical process accident that occurred at
Flixborough, England, in 1974 [1]. All British chemi-
cal, engineering curricula are presently required to
contain a significant amount of safety related content
for accreditation. This is achieved through a combina-
tion of dedicated courses or by demonstrating a cer-
tain fraction of safety related content in the existing
courses.
This article will provide a dual perspective on the
need for more emphasis on safety in our chemical en-
gineering undergraduate curricula. Both the academic














Daniel A. Crowl is an associate professor of chemical engineering
at Wayne State University. He received his BS degree in fuel science
from the Pennsylvania State University and his MS and PhD in chem-
ical engineering from the University of Illinois. After graduation he
spent two years as a senior process control engineer for St. Regis Paper
Company before returning to academic work. His present research in-
terests are in chemical process safety, process simulation, and the ap-
plication of CAD techniques to process design. (L)
Joseph F. Louvar is director of the Chemical Engineering Depart-
ment (Research and Development) of BASF Corporation, Wyandotte,
MI. A member of AICHE's Loss Prevention Committee, he also teaches
experimental statistics at Wayne State University. He received a PhD
in chemical engineering from Wayne State University, an MSChE from
Carnegie-Mellon University, and a BSChE from the University of Mis-
souri (Rolla). (R)


CHEMICAL ENGINEERING EDUCATION










This article provides[s] a dual perspective on the need for more emphasis on safety in our undergraduate
curricula. Both the academic and industrial viewpoints [are] presented. We .also discuss a unique safety
related course development effort being undertaken at Wayne State and Michigan Tech Universities ...


and industrial viewpoints will be presented. We will
also discuss a unique safety related course develop-
ment effort being undertaken at Wayne State Univer-
sity and Michigan Technological University, with sub-
stantial technical assistance from BASF Corporation.
This program is being supported by the National Sci-
ence Foundation under its University/Industry/Gov-
ernment (UIG) program.

THE UNIVERSITY PERSPECTIVE
Dan Crowl
In the summer of 1986 I had an opportunity to
spend a few months at BASF Corporation working on
a computer simulation project. One day I was visited
by Joe Louvar, Director of Chemical Engineering,
who asked me if I was interested in working on a few
safety related projects. In total ignorance I replied,
"You mean hard hats and safety shoes?" Joe went on
to explain some of the more fundamental aspects of
safety, including reactor dynamics, two-phase flow
during reactor venting, gas dispersion models, and so
forth. It had never occurred to me that safety had a
fundamental aspect! I had always assumed safety was
simply a set of rules developed as a result of practical
experience.
During the remainder of my stay at BASF, I ob-
served how safety was practiced in an industrial envi-
ronment. It became apparent to me that practicing
chemical engineers spend a considerable amount of
time on safety related activities. In spite of all my
years of academic training, I felt woefully inadequate
with respect to safety. I could indeed understand
pieces of the fundamental components, but using those
pieces in the practical application and management of
safety seemed a mystery to me.
Despite a continuing feeling of inadequacy, I was
now aware of safety. I began to look at my past
academic experiences and at the experiences we are
providing for our present students. I found little in
the way of safety instruction. Many of you will argue
that the academic experience is designed to provide
only a fundamental knowledge, with practical applica-
tion being the responsibility of industry. But safety is
a systems science, involving the application of a broad
range of fundamental skills strongly coupled with
practical application. Why not adjust the fundamental
skills taught to our undergraduates to strengthen the


safety aspect? Can the academic community continue
to neglect an area that is already very important to
the industrial community?
As an undergraduate I learned that process design
is motivated by savings in material usage and capital
investment. In the 1970's energy also became an im-
portant driving force in design. Now, safety consider-
ations are becoming just as important.
The academic community is in a difficult position
with respect to safety. First, most faculty have little
safety experience. This is either a result of little indus-
trial experience or of participation in industry during
a time when safety was a lower priority and was prac-
ticed in a different fashion than it is today. Second,
we have inadequate equipment and facilities for dem-
onstrating safety. Finally, our curricula are already
at their practical maximums. Where can we find room
to add a safety component?
One possibility is to add safety instruction
throughout all of the chemical engineering courses.
Safety is practiced in every aspect of the industrial
experience, so why not practice it throughout the en-
tire undergraduate course sequence? I agree with this
approach. However, a final capstone course is neces-
sary to culminate the course sequence. One can teach
reactor safety (runaway reactors) in the kinetics and
reactor design course, but a capstone course is essen-
tial for the teaching of safety reviews, hazard identifi-
cation, risk assessment, and other topics that are
special to safety.
I recently attended the International Symposium
on Preventing Major Chemical Accidents, organized
by the Center for Chemical Process Safety of the
AICHE. This symposium was held in Washington,
DC, during the week of February 3, 1987. Of the over
four hundred participants, less than five were from
academia. This was surprising since the papers pre-
sented were as fundamental and research-oriented as
those presented at academic conferences. I believe
that safety can become an important academic area if
faculty become aware of 1) the potentially fundamen-
tal nature of safety, and 2) the opportunities for re-
search and funding. Safety is an excellent area for
industry/academic research collaboration since most
of the industrial proprietary barriers dissolve.
Safety is an essential part of the industrial experi-
ence, more now than some of the traditional funda-
mental chemical engineering areas. I believe it is time


SPRING 1988









to include safety as an important part of the under-
graduate chemical engineering experience as well.


THE INDUSTRIAL PERSPECTIVE
Joe Louvar

All of my safety knowledge has been acquired in
an industrial setting. It came in bits and pieces, and
it took several years to really appreciate the signifi-
cance, the subtlety, and the technical complexity of
chemical process safety. In hindsight, I recognize that
during those formative years I made many serious
safety errors. It was only the result of pure chance


that I escaped having serious accidents and/or in-
juries.
Using today's standards, most of my past errors
would have been identified and corrected by supervi-
sion or by the safety review process which is used
very effectively within industry. Unfortunately, I am
convinced that the self-motivation necessary to take
on the responsibility for safety (responsibility of every
engineer), and to learn the technology of safety, still
requires years within an industrial setting.
From a management standpoint I have also recog-
nized that supervision has a serious handicap when
working with fresh graduates. Although they have an


TABLE 1
Course Outline


1. Introduction
A. Course syllabus
B. Introduction
C. Accident and loss statistics
D. The accident process
E. Three significant disasters
F. Personal and management responsibilities
G. Legislative responsibilities
H. Employee and community "right to know"
2. Fundamentals of Safety
A. Toxicology, industrial hygiene and exposure control
1. Toxicology
History
Dose vs response
How the body responds to exposure
Determining safe working levels
2. Industrial hygiene and exposure control
Types of exposure
Methods for exposure control
Administrative and engineering methods
Personal protection
B. Fires
1. Flammability of vapors and liquids
2. Minimum oxygen concentration
3. Inerting
4. Autoignition temperature
5. Auto-oxidation
6. Adiabatic compression
7. Effects of sprays and mists
C. Explosions
1. Definitions: Explosion, detonation, deflagration, con-
fined and unconfined vapor explosion, BLEVE
2. The nature of the explosion process
3. Effects of explosions
4. Calculations relating to explosions
D. Liquid and Vapor releases
1. Gaussian distribution
2. Gaussian plume model
3. Gaussian puff model


4. Spill models
5. Inhalation exposure: spills or from vessels
6. Inhalation exposure: filling containers
3. Applications for safety
A. Introduction
B. Designing for safety
1. Intrinsic design
2. Extrinsic design
3. Maintenance
C. Engineering to Prevent Fires and Explosions
1. The Fire Triangle
2. Passive Protection Methods
Types of environments
Eliminating sources of ignition
Atmospheric control, including inerting and purging
Plant siting
3. Active Protection Methods
Shutoff and check valves
Combustible gas monitors
Emergency material transfer
Sprinkler systems
Water vs. foams for fire control
D. Engineering to Prevent Toxic Release
1. Relief systems
2. Design and selection of relief valves
3. Flares, vents and scrubbers
E. Process hazards identification and risk assessment
1. Hazards identification
Hazards surveys, including the Dow Index
HAZOP
SFMECA
2. Risk Assessment
Event trees
Fault trees
Consequences analysis
Limitations to risk assessment
3. Safety reviews
4. Accident Investigations
5. Case histories


CHEMICAL ENGINEERING EDUCATION









excellent foundation in the traditional chemical en-
gineering fundamentals, the concepts of safety are
foreign to them. In fact, they perceive safety to be
unimportant because the subject is hardly mentioned
during their schooling. In some universities, safety
practices are totally neglected. Consequently, indus-
try starts with graduates who need a basic improve-


TABLE 2
Video Session 1

INTRODUCTION
1. Introduction
2. Brief description of five video sessions
3. BASF Corporation's safety program
a. Policy
b. Committments
c. Training
d. Activities
e. Audits
4. Introduction to safety terminology and principles
a. XP vs. non-XP
b. Relief vs. rupture disc
c. Runaway reaction
d. Two phase flow
e. Design Institute for Emergency Relief Systems
(DIERS)




TABLE 3
Video Session 2

BASIC SAFETY EQUIPMENT AND PROCEDURES
1. Introduction
2. Personal protection equipment
a. Motivation
b. Face and eye protection: glasses, goggles, face shield
c. Clothing: aprons, gloves, suits
d. Respirators: dust, cartridge, canister, air-line
e. Miscellaneous: hats and shoes
3. Sources of toxicity and safety information
a. Material Safety Data Sheets (MSDS)
b. Vendor information
4. Safety procedures
a. Hot work permits
b. Lock Tag Try
c. Vessel entry
d. Grounding and bonding
e. Fail safe
f. Safety reviews
5. Safety features
a. Sprinklers
b. Alarms
c. Showers
d. Color indicator tubes
e. Extinguishers


ment in attitude prior to addressing the more complex
features of safety.
The chemical industry is entering an era where
processes will be even more complex. To prepare our
future engineers for this era, I believe we must begin
to educate them on the principles of safety. This edu-
cation must begin at the university, with an emphasis
equal to heat transfer, mass transfer, ther-
modynamics, etc., and should be given during the
same period as the core courses. This would be an
effective way of emphasizing the importance of safety.
I believe that there should be a three-hour course
during the third year which is dedicated to safety.
Some educators believe that safety should be a part
of every course, and I agree. But the more technically
complex areas of safety need more time and attention
than could possibly be allotted in core courses. The
following subjects could only be adequately covered in
a separate safety course:

Dispersion studies
Relief valve sizing (including 2-phase flow)
Safety reviews and hazard identification
Flammability of chemicals
Toxicity of chemicals


The benefits of teaching safety in the university
exceed those mentioned above. An important spin-off
of safety courses could be the development of more
interest in initiating safety-oriented research. In a re-
view of university research (PhD dissertations), it is
apparent that there is very little research devoted to
safety. Topics which could be fruitfully addressed by
universities include:

Advanced adaptive reactor control methods for hazardous
reactions
Expert systems for monitoring reactor and/or plant safety
Expert systems to improve the reliability of fault analysis
Advanced relief methods for runaway two-phase flow sys-
tems

From my perspective, the chemical industry will
continue to stress faster reactions, more complex
reaction systems, and a more complex utilization of
investments via multiple product reactor systems.
The success of these complex processes will depend
upon our ability to design modern systems with the
safety features demanded by both industry and the
public. To meet these challenges of the future, we
must give more attention, concern, and respect to
safety.


SPRING 1988


77


77









NEW COURSE DEVELOPMENT

We have developed a senior level course entitled
"Safety in the Chemical Process Industries." This
development effort was supported by the National
Science Foundation under their UIG program. The
unique feature of the course was that it included five
two-hour lectures, broadcast live from the chemical
pilot plant facilities at BASF Corporation in Wyan-
dotte, MI. The format supported interactive question-
ing between the students in the classroom and the
practicing chemical engineering professionals in the
pilot plant. The broadcasts were uplinked to a satellite
for downlinking at both Wayne State University in
Detroit and Michigan Technological University in the
upper peninsula. The course was taught simultane-
ously at both locations during the fall of 1987.
The course provided a rare opportunity for stu-
dents to see safety being practiced by professional
chemical engineers in an actual chemical plant envi-
ronment. Demonstrations were provided using actual
equipment and safety situations that could never be
shown in the classroom or laboratory. The video lec-
tures were in addition to a series of 25 one and one-half
hour classroom lectures. These lectures presented the
fundamental and theoretical features of safety, sub-
jects requiring problem solving and discussion within
the classroom environment.


TABLE 4
Video Session 3
INSPECTION OF LABORATORY AND PROCESS AREA
1. Introduction
2. Inspection concepts
3. Review film on Stop Observe Act Report (SOAR)
4. Inspection of laboratory (several examples of poor safety
practices will be staged)
a. Storage of solvents
b. Storage of glass equipment
c. Safety equipment and operation
d. Principles of using hoods
5. Inspection of Process Development (PD) area (several ex-
amples of poor safety practices will be staged)
a. Reliefs and rupture discs
b. Nitrogen vented in room
c. Non-XP in XP room
d. Poor grounds
e. Belts not guarded
f. Bad drum vent
g. Incorrect tools
h. Poor ventilation
i. No double block and bleed (show correct configuration
via glass system)
j. No hot-work permit


A total of four industrial and seven academic con-
tributors were involve with the effort. The BASF
Corporation industrial participants were: G. W.
Boicourt, M. Capraro, J. F. Louvar, and J. Strick-
land. Their effort was directed mostly towards the
video lecture material and scripts. J. F. Louvar also
contributed towards the development of lecture mate-
rial. The academic participants were D. A. Crowl and
R. W. Powitz from Wayne State University and M.
Banks, B. A. Barna, E. R. Fisher, N. K. Kim, and D.
G. Leddy from Michigan Technological University.
The major focus of the academic group was toward



TABLE 5
Video Session 4

EXPERIMENTS FOR SAFETY
1. Introduction
2. Vent Sizing Package (VSP) for sizing reliefs
a. Show features of equipment
b. Show types of data collected
c. Illustrate specific tests for various objectives
d. Introduce relief valve sizing concepts
3. Characterizing dusts
a. Deflagration index
b. Illustrate features of equipment
c. Describe test procedures
d. Describe data from tests
e. Introduce relief sizing concepts
4. Flammability limits
a. Illustrate features of equipment
b. Describe test procedures
c. Describe data from tests
d. Introduce relief sizing concepts




TABLE 6
Video Session 5

SAFETY REVIEWS
1. Introduction
2. Informal safety review in laboratory
a. Describe phosgenation system
b. Analyze procedures relevant to safety
c. Progressively make improvements via team dialogue
concept
3. Formal safety review
a. Introduction
b. Safety Review Meeting
c. Equipment inspection
d. Wrap-up meeting and development of Action Plan
4. Wrap-up of video sessions
a. Summary remarks
b. Open questions and answers


CHEMICAL ENGINEERING EDUCATION









development of the lecture materials. The overall pro-
ject was managed by D. A. Crowl.
The course outline is shown in Table 1. It is divided
into two major parts. The first part presents the fun-
damentals of safety and includes discussions of tox-
icology, fire, explosion, and toxic release. The second
part deals with using those fundamentals in practice
and includes a discussion of "designing for safety" and
using various safety review procedures (such as
hazards and operability studies). The course also in-
cludes a discussion on case histories and accident in-
vestigations.
The outlines for the five video lectures are shown
in Tables 2 through 6. Except for video session 4, the
videos are not dependent on the lecture material. The
emphasis of the videos was to show the students how
safety is practiced on real process equipment. The
fourth video lecture on "Experiments for Safety" re-
quired some fundamental lecture material prior to
broadcast.


SUMMARY
This paper has presented both the industrial and
university perspectives regarding the need for teach-
ing safety in the chemical engineering undergraduate
curriculum. We have also presented one approach to
teaching safety and loss prevention. As a result of
NSF support we had a unique opportunity to bring
the students into an operating chemical pilot plant,
through the use of live TV.
We hope that this approach, and others, will im-
prove the engineers of the future and result in safer
chemical process plants.

REFERENCES
1. Lees, F. P., Loss Prevention in the Process Industries, Butter-
worths, Boston (1986).
2. Anonymous, AICHE Environmental Division Newsletter,
February, 1986.
3. Fisher, H. G., "DIERS Research Program on Emergency Re-
lief Systems," Chemical Engineering Progress, August (1985).
4. Glass, Arnold J., Letter of February 2, 1986. O


IM book reviews

INTRODUCTION TO POLYMER
VISCOELASTICITY, Second Edition
by John J. Aklonis, William J. MacKnight
John Wiley & Sons, Somerset, NJ 08873; $39.95 (1983)
Reviewed by Albert Co
University of Maine
This book introduces various fundamental concepts
in studying the viscoelastic behavior of polymers, with
an emphasis on the molecular approach. The book con-
sists of nine chapters.
Chapter 1 introduces the reader to several exper-
iments that display the viscoelastic nature of poly-
mers. In Chapter 2, viscoelastic material properties
in transient and oscillatory experiments are defined
and are illustrated clearly with simple experiments.
The Boltzmann superposition principle is stated; its
applications in relating the creep compliance and the
stress relaxation modulus and in relating transient
and oscillatory properties are demonstrated.
In Chapter 3 the regions of viscoelastic behavior
are described and the effects of molecular weight,
crystallinity, and plasticizing agents are explained.
The concept of time-temperature superposition, the
master curves, and the WLF equation are then pre-
sented. In Chapter 4, the phenomenon of glass trans-
ition is examined, and explanations based on free vol-
ume, thermodynamics, and kinetic theories are pre-
sented. The effects of structural parameters on glass


transition temperature and the relaxation occurring
in the glassy state are rationalized in terms of molecu-
lar motion and chain mobility. In preparing the reader
for subsequent chapters, the statistics of a polymer
chain are reviewed in Chapter 5.
In Chapter 6, various treatments of rubber elastic-
ity and the structural factors that affect rubber elas-
ticity are discussed. In Chapter 7, the behavior of typ-
ical mechanical models is analyzed and the Rouse-
Zimm molecular theories for polymer solutions are
discussed. Extensions of these molecular theories to
bulk polymers are then considered and the reputation
theories are briefly described. In Chapters 8 and 9,
the phenomena of dielectric relaxation and chemical
stress relaxation are examined, respectively.
Throughout the book, the mathematical treat-
ments are maintained at a level comfortable for under-
graduates. Advanced mathematics required for the
discussion of a subject matter are elaborated in the
corresponding appendices. The problems at the end of
each chapter range from simple calculations to ad-
vanced problems requiring a certain degree of
mathematical sophistication. Readers will find the so-
lutions located at the end of the book to be helpful.
Overall, this book is an excellent introduction to
polymer viscoelasticity. However, the treatise is re-
stricted to amorphous polymers. The treatment on
crystalline polymers is very limited, and topics such
as solution behavior, melt rheology, and birefringence
are not covered. Nevertheless, it is a good choice as
a textbook for one of a series of courses on polymer
viscoelasticity. [


SPRING 1988










M N curriculum


ENGINEERING MANAGEMENT

A Course to Minimize Functional Handicaps of Graduates


REX T. ELLINGTON
University of Oklahoma
Norman, OK 73019

THE INDUSTRIAL ECONOMY in the U. S. has un-
dergone significant changes since 1982. Some of
the changes have been initiated by recent intensive
overseas competition, some by the drop in oil prices,
and some by mergers. In the 'right sizing,' or downsiz-
ing, that has taken place in most companies in this
decade in order to maintain competitiveness, whole
layers of management have been expunged. Other
changes resulted from technological developments in
the late 1970s and in the first half of this decade. De-
velopments in computer workstations and software
have added such power to the individual engineer that
companies in chemical processing and the engineering
companies serving them will never again need as
many engineers of any discipline for the same volume
of business as they needed in the 1975 era. As a conse-
quence, there has been a reduction in hiring by most


Rex T. Ellington has been professor of chemical engineering at the
University of Oklahoma for five years, after nine years at IGT (Illinois
Tech) and nearly thirty years in industry. He received his BS (ChE) from
the University Colorado and the MGT, MS (ChE), and PhD from IIT. He
held research and general management positions in Sinclair Oil, At-
lantic Richfield, and Transco Energy and had project management po-
sitions in Sinclair, Transco, Fluor Engineers and Constructors and Occi-
dental. Along the way, he had a private consulting practice for a year.
His teaching has concentrated on senior design, process control, and
engineering management.


companies, with the result that many students have
not been able to find jobs even months after gradua-
tion. Competition for the available jobs has become
intense, and many young engineers have had to go
into fields where they may never be able to use their
hard-won training. The purpose of this paper is to
review a course, "Engineering Management," that has
been developed at the University of Oklahoma to im-
prove the students' ability to get jobs and to perform
better in any company.
Important technical consequences of the industrial
changes which were described in a recent article in
this journal [1] agree with my own observations as a
senior manager that

A smaller percentage of all engineers, especially in larger
companies, will be able to move into management.
All engineers will have to be more competent technically
and to remain competent longer before they can become
managers.
While they practice their specialty, engineers will have to
function more effectively in a company environment to
have job security.
Engineers will need far greater organization- and people-
skills than past graduates possessed.

The needs represented by the last item have been
voiced by every industrial advisory group that I have
known or been a part of in the last twenty years.
As soon as recruiters identify a lack of understand-
ing of the industrial environment and an obvious lack
of people-skills in interviewees, they lose interest.
These graduates may never get an engineering job,
or they may be shuffled aside at an early date. As
recently as April 1987, senior managers advising our
department told about students who were essentially
unemployable, no matter what their GPA, because of
their obvious lack of people-skills. The student needs
to understand how the industrial world operates and
should develop people-skills before leaving college so
he/she can function effectively immediately after
being hired. Costs are simply too high for businesses
Copyright ChE Division ASEE 1988


CHEMICAL ENGINEERING EDUCATION









to tolerate a one-year shakedown period for new en-
gineers, as they did in the past.
Another educational need is based on the convic-
tion of most senior managers that learning to manage
is a lifetime job. This learning process must be started
very early by engineers in order to meet the greater
competition for managerial positions or they will be
left behind, and business majors will skim off an even
larger share of the top jobs in the future.

CHANGING THE CURRICULUM
C. T. Sciance [1] cites a significant need to aug-
ment the technical content of the chemical engineering
curriculum, but it could be even more important to
improve people-skills and knowledge of how industry
functions. Many students gravitate toward engineer-
ing because their natural aptitudes lie more in math
and science and less in communication and interper-
sonal relations. Throughout their engineering study
this uneveness in skills probably increases. Students
are shocked to find, while being interviewed or upon
being hired, that the practice of engineering is a very
people-oriented profession. They soon realize they are
woefully unprepared in many areas that could affect
career achievement even more than their calculation
ability. A serious gap in their education soon becomes
apparent.
The effects of the new employment environment
have only lately reached the universities in the form
of reduced job opportunities, but few changes have
been made in the educational program in response to
the needs of those now graduating and to the next
generation of students. Why has the gap in organiza-
tional know-how not been filled? Many professors
have no industrial experience, or it was of such short
duration, or at such a junior level, that they do not
know what is really involved and are uncomfortable
trying to teach such material. They may agree with
the need, but they lack the tools to deal with it.
An effort has been underway at the University of
Oklahoma to develop a local solution to the problem
for students willing to devote three semester hours to
the effort, and to draft a text which could be used by
other faculties. Further, we hope that the text might
ultimately become a desktop reference which will be
consulted more frequently than many technical refer-
ences.

COURSE SPECIFICS
Ideas for this course have gestated for more than
a decade, through thousands of goal-settings, perfor-
mance reviews, hirings, firings, and many mistakes


The course then becomes one of high
student-participation. Basic value systems are
discussed to help the student understand why
others react to life's inputs differently
than the student himself does.


by the author. They came from fellow executives of
all ages who urged that students be made aware of
the things that cause so much difficulty in industry
and handicap so many careers. Time and time again,
the root causes of those difficulties are poor communi-
cation and lack of interpersonal relation ability, in-
adequate knowledge of company workings, or dis-
torted concepts of management. Lately, new ideas
have come as feedback from recent graduates.
This is a course in what life is really like "out
there." Many of the rough edges have been ground
off. Attendance has been interdisciplinary, and the
first results have been good. Feedback and requests
from earlier graduates have resulted in the develop-
ment of a text. Graduates have already reported on
more comfort in working in team situations, a greater
ability to determine supervisors' needs by quickly de-
termining how an organization works, and less fear of
moving into project work immediately. These results
suggest that the effort is moving in the right direction.
The first draft of the text was developed to meet
classroom needs and it is outlined in Table 1. The first
part of the course is very important for the student
and new graduate. It starts with a brief perspective
of the entire course and then presents a view of the
types of behavior and functioning expected of all em-
ployees. The characteristics of effective employees
and good supervisors are discussed. There are numer-
ous "horrible examples" to awaken the student to
things that can happen which develop a bad reputa-
tion. The intent is to open the student's mind to the
concepts that lead to being an effective person.
The course then becomes one of high student-par-
ticipation. Basic value systems are discussed to help
the student understand why others react to life's in-
puts differently than the student himself does. In the
main, students seek the company of peers and cliques
with whom they are comfortable. It is often a shock
for them to realize that they may have to get along
and work with people who see life differently. The
value system is a convenient mechanism for identify-
ing differences in reaction to one's environment.
Next, attention turns to getting a job-the right
job for the individual. The student is taught how to
take full, personal charge of job-finding. Careful as-
sessment of the student's attributes leads to the de-


SPRING 1988









velopment of lists of job opportunities, location of
target companies, investment of significant time on
company research, and improved communication in
resumes and interviews. Effort is expended to
heighten the student's awareness of his/her impact on
recruiters, interviewers, and other people during
plant trips.
Then, important things that the newly employed
engineer should consider in order to get the best pos-
sible start on the job are reviewed. These are obvious
truisms, but they are often overlooked or ignored.

It is necessary to develop a different
frame of mind in the students in order to attack
the deficiency in communication ability which
is mentioned by nearly all supervisors.

Some of the most important areas concern the study
of personal ethics, business ethics, technical and intel-
lectual property of a company, liability problems (to
give the student perspective regarding what must be
protected), and obligations during and after employ-
ment. Then they examine one problem all of them will
encounter some day in some way: whistle blowing.
Moral obligations are analyzed in detail.
It is necessary to develop a different frame of mind
in the students in order to attack the deficiency in
communication ability which is mentioned by nearly
all supervisors. Communication theory is examined
briefly to learn why some of the best intentions go
awry in speaking and writing. Throughout the course
there are opportunities to make brief formal presenta-
tions, and each is critiqued by the class for its organi-
zation and delivery. At least one presentation is re-
peated several times in order to force the individual
to revise and improve it. Written assignments are re-
quired frequently. The student is taught to observe
audience awareness, whatever the occasion.
The true function of management is examined next
and it proves to be a revelation for many students. As
a result, some students are less anxious to become
managers, but at the same time they develop a better
appreciation of the needs of supervisors and mana-
gers. To help the student understand his/her new en-
vironment, subjects such as types of company organi-
zation, job descriptions, job evaluations, goal setting
for individuals, performance reviews, and merit
budgeting are taken up. Leadership types are
examined in some detail so that young employees can
better characterize the behavior of supervisors and
work with individuals more effectively. Equal employ-
ment opportunity requirements are carefully dis-


cussed to help people avoid problems and/or to insure
that their rights are observed.
The added awareness provided by these subjects
makes the student a better employee. People who
have had the course seem to get a faster start on the
job and to have an edge on the competition-befitting
the adage, "If at first you don't succeed, you may
never." They are more aware of the importance of
fitting into group activities, have greater empathy for
a supervisor's needs, and are able to develop a per-
sonal program for advancement at an early date. The
rate of return for the time and money invested in this
part of the course appears to be very high.

MODERN TECHNOLOGY EMPHASIZED
The middle section of the course deals with modern
methods of modeling, planning, and analysis. Tech-
niques and sophistication vary widely from company
to company, so the treatment provides a look at the
utility of these tools without taking a course in opera-
tions research or a business course in modeling. Each
method is examined from a management rather than
a program writer's viewpoint, to show how the analyst
and manager must work together and how model com-
puter outputs or "decisions" are used. General
economics and spreadsheets are reviewed quickly to
provide the basis for economic decisions. Reading and
analysis of company financial reports is stressed as an
ability that every young engineer should have. The
linear programming, which is one of the most widely-
used optimization aids-from refinery operation to
vegetable planting-is taken through graphical inter-
pretation, matrix methods, analysis of computer out-
puts, and sensitivity studies. Study of integer pro-
gramming provides the basis for analyzing supply,
transportation, and other networks. Among these are
the critical path and PERT techniques. Next, the stu-
dent is presented with the structural basis for making
decisions on a more rational basis. Decision theory
considers the different paths electable in solving a
problem and deals with the probable cost or conse-
quence of picking each. Several types of business
examples are used.
Reliability analysis is a field which the Japanese
have adopted more widely than we have in the U.S.
for a number of years. Failure, repairability, main-
tainability, and availability are taken up in terms of
process plant units and design for reliability. This
leads to consideration of risk analysis, fault trees, and
failure mode and effect analysis. The fact that the
rigor of the methods clarifies thinking is immediately
clear. Results of these analyses do no good in the files,


CHEMICAL ENGINEERING EDUCATION










however, and emphasis is placed on continued think-
ing by those involved in operations in order to prevent
future catastrophes such as those which have occurred
in the last decade.
By this time, students realize the limitations of
deterministic inputs to models. Distributions are con-
sidered for input values, and Monte Carlo methods
are applied to such things as cost estimates and se-
quential operation simulations. This section of the
course provides a new awareness and perspective of
tools most engineering students have not seen before.
With this awareness comes the ability to help super-
visors through the use of tools lacking in other en-
gineers and the ability to sharpshoot with further
study when the time comes.

PROJECT MANAGEMENT TOOLS TAUGHT
Since so many tasks in business and engineering
are of a project nature, the third part of the course
treats management of projects from beginning to end.
Few young engineers know how projects get started;
therefore, design bases, requests for proposals, pro-
posals, contracts, and agreements are carefully re-

TABLE 1
Table of Contents

PART I. Functioning with People in Organizations
1. Introduction
2. Value Systems
3. Getting the Correct Job
4. Ethics, Technical Property and Legal Issues
5. Getting Along on the Job
6. Communication
7. Functions of Management
8. Company Organization
9. Managing People
PART II. Planning and Analysis Methods
10. General Economics
11. Linear Programming
12. Integer Programming and Networks
13. Decision Theory
14. Reliability, Fault Trees, FMEA
15. Probabalistic and Simulation Models
PART III. Work Management (Project Basis)
Glossary
16. Project Initiation
17. Project Organization/Kickoff
18. Engineering and Detailed Design
19. Equipment and Material Procurement
20. Time and Money control
21. Quality Assurance
22. Construction and Manufacture
23. Project Completion


viewed and discussed to show that projects don't "just
happen." A typical project is organized and engineer-
ing and detailed design activities are defined. Project
activities and individual engineering discipline ac-
tivities are related to total project progress curves,
with emphasis on the interrelationship of activities.
Probability of success on any project is related to the
amount of planning, to the handling of details, and to
the accuracy of measurements that go into it. All pro-
curement activities (spec sheets, inquiries, bids, bid
tabs, vendor selection, expediting, inspection, vendor
data acquisition, and transport) are put into time per-
spective. Time and money control discussions involve
activity definition, loading of effort on activities, crit-
ical path analysis, time and cost estimates, and cost
control. Emphasis is placed on the need to measure
progress in the same terms as the original estimate
for accurate progress reporting. Product and service
quality assurance needed for better reliability,
maximum productivity, and lower costs in the face of
overseas competition, provide background for treat-
ment of construction or manufacturing and project
completion.
At the end of this section, many students under-
stand the unusual characteristics of good project en-
gineers. Although they may not want to enter this
specialty, they can function much more effectively on
any project and realize the benefit of a project-man-
age-approach to their own lives.

CONCLUSIONS
Obviously, many topics are treated quickly, and
not all are treated every semester. But the objective
is perspective and development of the "big picture"
viewpoint. With this, the young engineer knows
where to dig when the need for details of a given
method or topic arises. Over the years, the author has
found that the engineer who has this perspective is
most often the one who defines the problem correctly
and comes up with the needed solution at the right
time, i.e., is the most promotable individual.
The chapter listing for the text is given in Table
1. Comments and ideas from educators or indus-
trialists are always welcome. As a veteran project
manager, the author's primary interest is improving
the product and serving the customer (the student
and the employer).

REFERENCES
1. Sciance, C. T., "Chemical Engineering in the Future," Chem.
Eng. Ed., XXI, No. 1, Winter 1987. [


SPRING 1988


---~-~--










I curriculum


A COURSE ON PRESENTING TECHNICAL TALKS



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


N THE ONGOING debate about what should be put
in and taken out of the chemical engineering cur-
riculum, one of the few points of universal agreement
is that oral communication skills are essential for all
practicing engineers and not enough is done to develop
them in most engineering curricula.
Like many other departments, we used to have a
senior seminar course in which each student presented
a single 30-45 minute talk sometime during the semes-
ter. The results were not particularly impressive: the
students whom you would have expected to give good
talks gave them; the other talks ranged from poor to
average; evaluations were superficial; and the one-
shot nature of the talks provided little opportunity for
individual improvement. Also, the worst of the talks,
being as long as 45 minutes, were excruciating experi-
ences for everyone involved.
When Harold Hopfenberg became Department
Head eight years ago, one of the first changes he in-
troduced was a complete reorganization of the seminar
course. Under the new system the senior class is di-
vided into groups of six to eight students and a faculty
member is assigned to each group. The groups meet
once a week for most of the semester. Each faculty
member runs the seminar in any way he or she
chooses; the next section describes a structure that
has worked particularly well for the author.

COURSE STRUCTURE
Each student prepares and delivers two talks dur-
ing the course. Talks in the first round are each fifteen


. one of the first changes [Harold Hopfenberg]
introduced was a complete reorganization of the
seminar course. Under the new system the senior
class is divided into groups of six to eight students
and a faculty member is assigned to each group.


0 Copyright ChE Division ASEE 1988


Richard M. Felder is a professor of ChE at N.C. State, where he has
been since 1969. He received his BChE at City College of C.U.N.Y. and
his PhD from Princeton. He has worked at the A.E.R.E., Harwell, and
Brookhaven National Laboratory, and has presented courses on chem-
ical engineering principles, reactor design, process optimization, and
radioisotope applications to various American and foreign industries
and institutions. He is coauthor of the text Elementary Principles of
Chemical Processes (Wiley, 1986).


minutes long for a 50-minute class period and twenty
minutes long for a 75-minute period. Talks in the sec-
ond round are ten minutes long.
The first meeting is organizational. The instructor
explains the course structure and hands out and re-
views a list of suggestions for good presentations (to
be given subsequently in this paper). He then lists all
of the dates on which the course will meet and re-
quests volunteers for the first two presentations, to
be given two weeks from the current date. If there
are volunteers the rest of the calendar can usually be
filled in on a voluntary basis; if not, a lottery is used
to assign presentation dates.
In the second period the instructor delivers an il-
lustrative seminar. He begins by explaining that he
has been giving technical talks for a long time and the
students should not expect to be able to duplicate his
skill at this stage of the game. He adds that even
experienced speakers can find room for improvement,
and he requests that the students take notes on things
that might be improved in the presentation. He then


CHEMICAL ENGINEERING EDUCATION











He provides no real introduction but launches directly into a long monologue read verbatim from notes. He
shows a crudely drawn flow chart with no units or streams labeled; several data plots with no apparent
relationship to anything; and one transparency that looks like a facsimile of the Dead Sea Scrolls ...
no eyecontact is made; and gum is chewed continuously and ostentatiously.


proceeds to give the worst talk he can possibly give.
He provides no real introduction but launches directly
into a long monologue read verbatim from notes. He
shows a crudely drawn flow chart with no units or
streams labeled; several data plots with no apparent
relationship to anything; and one transparency that
looks like a facsimile of the Dead Sea Scrolls, filled
from top to bottom with handwritten equations in tiny
print. The talk is crammed with undefined technical
jargon; no eye contact is ever made; and gum is
chewed continuously and ostentatiously. The speaker
concludes by fumbling around with his transparencies
and then muttering, "Well, I guess that's about it."
The talk lasts for approximately eight minutes and
by the end of it the students have all caught on to
what is happening. The class then brainstorms the
things that were wrong about the talk and what
should have been done differently. The instructor sub-
sequently gives a coherent version of the same talk,
complete with introduction, body, summary, and in-
telligible transparencies, and the class briefly discus-
ses the things that made a difference relative to the
first presentation.
At least one week before students are scheduled
to present their talk they are required to submit a
topic. They may talk about published papers or about
work they did on projects or summer jobs; the only
ground rule is that the talks should be reasonably
heavy in technical content at a level appropriate for
chemical engineering seniors. The instructor reviews
the chosen topics and tells the students either to go
ahead with them or to find alternatives with more
technical content. During the week before the presen-
tation the students write, duplicate, and hand out
seminar announcements and abstracts to the instruc-
tor and to all group members.
A typical class session begins with the presenta-
tion of the first talk. During the talk the class mem-
bers jot down questions about the content and com-
ments on the presentation. A five-minute questioning
period follows the talk. The students and instructor
then fill out a checklist rating various aspects of the
talk (introduction, body, and summary; level of the
material presented; use of time; quality of transparen-
cies; clarity of presentation; speaking style; strong and
weak points of the talk). Finally, the students and the
instructor each present brief oral critiques. The


checklists are given to the speaker to review and are
handed in to the instructor at the following class ses-
sion. The procedure is then repeated for the second
speaker.
Since the course was last offered the department
has acquired a videotaping facility. In the future, each
talk will be taped and the students will be requested
to view their presentations and critique themselves.

SUGGESTIONS TO SPEAKERS
Preparation and Organization
Know the technical background of your audience and gear
your talk to that level. Do not use a lot of unexplained
technical jargon unless you are sure the audience already
knows what it means, and don't explain what a material
balance is to a group of chemical engineering seniors.
Make sure your talk has a distinct introduction (outline
what you are going to say and why it might be important
or interesting to your audience), a body, and a summary
(repeat what you particularly want your audience to retain
from the talk).
Use overhead transparencies or slides to present main
points and provide explanatory details in the talk. Trans-
parencies work well for informal seminars, and you can
easily make them yourself. Slides are more difficult to pro-
duce, but they are often more convenient for short tightly-
timed presentations and they are required at some national
and regional conferences.
Never present a large body of information orally without
summarizing its main points on a transparency.
Do not present more than about eight lines on a single
transparency. Transparencies crowded with information
are useless. It should take about two minutes to go through
a single transparency in the talk.
Use large type on transparencies-a label maker or the
Orator ball on a Selectric typewriter or a word processing
program with variable type size. Ordinary size type doesn't
look good. If you handwrite the transparency, use large
block lettering with horizontal guidelines to keep your
lines straight. Never use script unless you're Octave
Levenspiel.
If you show a process flow chart, make sure the units and
streams are labeled. A bunch of unlabeled boxes and lines
with arrows is worthless to the audience.
Try to avoid complex equations, which can rarely be
explained intelligibly in the amount of time available. If
you are talking about a mathematical model, focus on
what it does (input and output variables, assumptions) and
provide, at most, qualitative summaries of the mathemati-
cal and computational details. (If listeners want more de-


SPRING 1988









tails they can ask you for them in the question period.)
If you show data plots be sure the axes are clearly labeled.
Rehearse your talk several times with a friend or in front
of your mirror, and make sure the time it takes is within
one minute of the time allotted for the talk. Running long
can be a disaster in a formal presentation and running
short may not win you any friends if you're at a meeting
where consecutive talks are scheduled at set times.

Presentation
Never read directly from prepared text-there is nothing
more deadly to an audience.
Make frequent eye contact with your audience throughout
the talk. Do not stare at your notes or at the screen.
Sound enthusiastic about your subject, or at least in-
terested in it. Do not speak in a monotone. Gesture occa-
sionally. If you seem bored by your material you can be
guaranteed your audience will follow your lead.
Make sure your watch is visible and check it occasionally
to see how the time is running. If you see you are running
short or long, try to adjust the speed of your presentation
to compensate.

DISCUSSION
The improvements in the student presentations as
the semester progresses are clear and frequently
dramatic. Almost invariably poor speakers become
adequate, adequate speakers become good, and good
speakers become better. During the past six years a
student from our department has won the regional
AIChE student chapter paper award competition
three times. We can't prove it, but we are convinced
that the seminar course has a lot to do with this re-
cord.
The oral critiques are a valuable and interesting
part of the course. The natural student tendency is
to be excessively polite, to avoid criticizing harshly
lest they themselves come in for the same treatment
when it's their turn to speak. As a result, in the first
few sessions the principal burden of criticism falls on
the instructor. However, as the semester pro-
gresses the student criticisms become more and more
germane and incisive, although courtesy is always ap-
propriately retained. (We are Southern here, after
all.) By the end of the semester the instructor is al-
most redundant: the points he is prepared to make in
his critique are usually made first by the students.
Requiring each student to give a fifteen- or
twenty-minute talk and subsequently a ten-minute
talk seems to work very well. It is usually difficult
(even for seasoned professionals) to present a signifi-
cant body of technical material in twenty minutes;


having to do so provides the students with excellent
practice in preparing technical seminars such as those
at national AIChE meetings. Cutting the material
down to ten minutes presents a whole different set of
problems, as the students quickly discover. The latter
exercise is good preparation for, say, company staff
meetings at which many people must summarize their
work in a relatively short time.
Finally, it is critically important for the course in-
structor to remember that the students taking the
course are particularly vulnerable: they are nervous
about public speaking in general and they are espe-
cially not used to being publicly critiqued. If the criti-
cism is destructive or unduly harsh, or seemingly arbi-
trary and unfair, the course has the potential of doing
much more harm than good. However, as long as the
instructor establishes firm ground rules about criti-
cism and takes the lead himself in creating a suppor-
tive environment, the course can be among the most
positive and rewarding educational experiences the
students experience in their academic careers. D



book reviews

CATALYST DESIGN:
PROGRESS AND PERSPECTIVES
by L.L. Hegedus, A.T. Bell, N.Y. Chen, W.O. Haag,
J. Wei, R. Aris, M. Boudart, B.C. Gates, and G.B.
Somorjai
John Wiley & Sons, Somerset, NJ 08873; 288 pages,
$47.50 (1987)
Reviewed by
R. J. Gorte
University of Pennsylvania
While there are a number of books on catalysis, it
is very difficult to find a book which gives a balanced
presentation of the many topics in this field. The prob-
lem is that everyone working in catalysis has a differ-
ent view of what the subject is and what is important.
People working in surface physics view catalysts as
adsorption on single crystals in ultra-high vacuum,
mathematical modellers view it as concentration and
temperature gradients across a catalyst pellet, and
traditional workers in catalysis view it as the turnover
number or selectivity for a reaction carried out over
a fixed bed. While not written specifically as a
textbook, Catalyst Design: Progress and Perspectives
has tried to give an overview of work carried out by


CHEMICAL ENGINEERING EDUCATION









all types of catalyst researchers by bringing together
leaders from several of the important areas in the field
and having each write a brief review of the important
aspects of their particular area.
The book itself is a series of short review articles,
each written by a different author. The first chapter,
written by L.L. Hegedus, very briefly discusses the
continued importance of heterogeneous catalysis to in-
dustrial practice and lists the applications which uti-
lize the largest quantities of catalysts.
A microscopic viewpoint of catalysis on single-
crystal, metal surfaces is presented by G.A. Somorjai
in Chapter 2. The work cited is mainly from Professor
Somorjai's own research and discusses the results of
reaction and adsorption studies on single crystals, in-
cluding topics such as the importance of crystallo-
graphic structure for reactions on metals and the influ-
ence of surface modifiers on several example reac-
tions. It should be noted, however, that some of the
conclusions reached in this chapter are still controver-
sial within the surface science community.
The third chapter provides a discussion of sup-
ported, organometallic clusters by B.C. Gates. The
chapter begins with a review of catalysis by transition
metal clusters and continues with a discussion of work
carried out to anchor these compounds to a support.
This second part reviews the synthesis of supported
complexes and concentrates on the spectroscopic tech-
niques which have been utilized in characterizing
these catalysts. Following the section on synthesis
and characterization is a discussion of the catalytic
properties for several example catalytic systems.
Chapter 4, by A.T. Bell, is a review of supported
metal catalysis, with an emphasis on the effect that
the support can have on the metal. The chapter re-
views a wide range of topics, including support acid-
ity, preparation procedures for introducing metals
onto a support, and the influence that a support can
have on a metal's adsorption and reaction properties.
Most of this last section involves a discussion of the
unusual properties which can be observed with titania
supported metals. It should be noted that Professor
Bell presents certain conclusions concerning the role
of titania which are still being debated in the litera-
ture.
A discussion of reaction kinetics and the design of
catalytic cycles is given in Chapter 5 by M. Boudart.
Since most reactions involve several elementary
steps, Professor Boudart suggests ways for logically
designing catalysts assuming that the intermediate


steps can be selectively altered by judicious choice of
catalyst or operating conditions.
W.O. Haag and N.Y. Chen have written a review
of acid catalysis by zeolites in Chapter 6. Their chap-
ter starts by describing what zeolites are, followed by
a discussion of zeolite properties including sorption
behavior, diffusional phenomena, and catalytic activ-
ity. The chapter includes a concise introduction to
preparation methods for zeolites, to techniques for
characterization of the acid sites, and to methods for
changing zeolite acidity. Following this introduction,
the role of zeolites in several commercial processes is
described, with a particular emphasis on the impor-
tance of molecular shape selectivity in those proces-
ses. The chapter ends with an overview of the design
principles which were incorporated into the develop-
ment of the first zeolite hydrocarbon cracking
catalysts.
The section on mathematical modelling of trans-
port properties in catalysis, Chapter 7, was written
by R. Aris. The chapter begins with the history be-
hind calculations of catalyst effectiveness and follows
with a tutorial on how to determine the influence of
catalyst geometry, reaction kinetics, and other factors
on the observed reaction rates. The chapter includes
a short section on methods for controlling the distribu-
tion of catalytic activity within a catalyst pellet and
concludes with a discussion of rate multiplicities and
stabilities.
The final chapter, written by J. Wei, presents the
design considerations used for hydrodemetallation
catalysts. The chapter begins by introducing the
reader to the complex structure of metal-containing
molecules which are present in petroleum. The rest of
the chapter reviews the problems associated with de-
hydrodemetallation in the presence of hydrodesulfuri-
zation and discusses the principles used to design
catalysts which have a high activity for long periods
of time.
Over all, this book provides a good review of a
wide range of topics in heterogeneous catalysis. While
the book could be used as a text for a course in
catalysis, it would be necessary to provide supplemen-
tary materials to provide background on the different
techniques which are discussed. As with any book on
topics for which research is ongoing, one should not
consider any of the chapters as being the final word.
However, each section does provide a good beginning
for the interested reader. There is clearly a need for
a book of this type. E


SPRING 1988










1[6 Y classroom


FLOW SHEET IS PROCESS LANGUAGE

MANFRED FEHR
Federal University at Uberldndia
38400 Uberldndia MG Brazil
220* F t.-n,
TEXTBOOKS WRITTEN FOR chemical engineering
core courses do not generally provide links be-
tween the theory presented and industrial practice of
process engineering. Many surprises are waiting for
students in the plant environment. We have found
that this situation can be considerably improved by
incorporating flow sheet conception into the lecture
program, mainly in unit operations and design
courses. The flow sheet forces the student to consider ("1 .
a particular piece of equipment in its true industrial
context where it ties in with utilities and basic control
strategy. The method not only motivates students to
participate creatively, but it also generates questions
on process engineering logic that are a welcome en-
richment of class activity. This paper will illustrate
with examples how students can be induced to ask T -
these questions and to find answers to them. ope

FIRST HEAT EXCHANGER EXAMPLE


To illustrate the calculations for heat transfer to
jacketed tank reactors, we went through an example
solved in a well-known textbook [1]. Knowing the heat


Manfred Fehr received his BS (1967) from the Universite Laval, his
MS (1969) from the University of Alberta, and his PhD (1977) from
University Laval. He is a registered professional engineer in Alberta,
Quebec, Sao Paulo and Minas Gerais. After short industrial experiences
in Canada and Germany, he lectured at various South American uni-
versities and is now engineering consultant and professor at the Fed-
eral University, Uberl6ndia, Brazil. His research interests are in the
area of local energy sources.


FIGURE 1. Steam supply to reactor jacket. Above: book
approach. Below: flow sheet approach.

load and the available surface, the steam temperature
required in the jacket is found to be 220F. The sub-
ject matter of the book goes this far, but thinking that
this was not a real challenge to my students, I went
on to speak some process language: our plant has two
steam headers, 50 psig and 150 psig, and a condensate
return header at 20 psig. I asked the students to pre-
pare a flow sheet with enough details to show exactly
how steam flows from which of the supply headers
through the jacket and back into the return header.
The jacketed reactor then acquires its true identity as
part of the processing unit. This is process language.
Some quite common "nonsense situations" were sub-
mitted from which a lot could be learned about the
logic of process engineering. Can you imagine a pump
at the jacket exit that simply pulls the steam through
from the supply header? Figure 1 shows a workable
solution. It also depicts the difference between a book
approach and a flow sheet approach to the problem.
Copyright ChE Division ASEE 1988


CHEMICAL ENGINEERING EDUCATION










SECOND HEAT EXCHANGER EXAMPLE
When studying the thermal design of shell and
tube exchangers, we went through an example solved
in another well-known textbook [2] which states that
"atmospheric pressure steam condenses on the shell
side." Nobody objected to this as long as we were
busy figuring the number of tubes and passes. The
surprise came when I again asked for the flow sheet.
Only then did it become clear that book language is
not process language. Even the simplest steam trap
requires a pressure drop to function, which means
that in addition to the equipment shown in Figure 1,
it will be necessary to pull a vacuum on the condensate
collection tank in order to make the steam flow at all.
What is more, this vacuum has to be monitored if a
pressure of 1.00 atm. abs. is to prevail in the shell.
The flow sheet question helped integrate the ex-
changer into the processing environment, and it be-
came evident that atmospheric pressure was a poor
choice for the shell.

EVAPORATOR EXAMPLE
The typical textbook evaporator calculation [3] will
state the pressure in the last effect to have a certain
value, such as 100 mm Hg absolute or 26 in Hg vac-
uum. All calculations are based on this number, al-
though no effort is made to explain its source. In addi-
tion, our school is located 3500 feet above sea level
where 26 in Hg is considered perfect vacuum. This is
another excellent situation to be explored with pro-
cess language. I got surprised looks when I asked,
"Would you please supply a flow sheet?" and, "What
would you do if some day you were asked to operate
at 23 in Hg vacuum?" The problem statement in the
book did not require that the student think this far
ahead. Many misconceptions about vacuum equipment
can be dispelled by simple confrontation of sense and
nonsense on the flow sheet. Can you imagine a pump
that sucks out all the vapor produced in the last effect?
The surprised looks disappeared as the answers
slowly took shape. A workable solution is presented
in Figure 2. Again, the figure compares book approach
to flow sheet approach to show the enrichment ob-
tained through process language.

DISTILLATION PRESSURE EXAMPLE
An uncommon situation appeared in the study of
the conversion of fermented beer into 96 volume %
ethanol in two distillation towers. The equilibrium
conditions are such that no reflux is required in the
first column. The overhead vapor containing 24 mol %


The method not only motivates students
to participate creatively, but it also generates
questions on process engineering logic that are a
welcome enrichment of class activity. This paper
will illustrate ... how students can be induced to
ask these questions and to find answers ...


20 In Hg ou


FIGURE 2. Condensate collection from evaporator.
Above: book approach. Below: flow sheet approach.

ethanol is fed directly to the appropriate plate of the
second column. On paper, the two columns looked like
any column depicted in textbooks, with the number of
plates determined by the method of McCabe-Thiele
and the feed plates chosen accordingly. Mass and heat
balances closed with the correct amount of open steam
at the bottoms and the correct amount of cooling
water in the second column condenser. There was but
one little problem: At the start, on deciding the
operating pressure, there was no reason for it to be
much different from atmospheric. A distilling com-
pany was consulted and the information obtained that
the first column runs at 3 psig, which confirmed our
reasoning. Consequently, the design proceeded with
this pressure. When the time came to provide a flow
sheet, there was no variable available for manipula-
tion to control the first column pressure. Textbooks
were consulted but no answer was found. On combin-
ing the flow sheets of the two towers and after a brain-
storming session, the fact slowly emerged that the
first column pressure in this scheme is not an inde-


SPRING 1988









pendent variable as it is in standard reflux columns
discussed in the books. The second column pressure
has to be controlled at the condensate drum, and the
first column pressure would then find its own equilib-
rium value, much like it does in a double effect
evaporator. Had the problem been terminated at the
textbook stage, this "discovery" would have been lost.

CONCLUSIONS
Flow sheet is process language. Process language
is exciting. Process language transforms book exam-
ples into process plant examples. Process language
resorts to the use of utilities to vitalize unit operations
teaching. Process language is a stimulating teaching
tool.

LITERATURE CITED
1. Kern, D. Q., Process Heat Transfer, McGraw-Hill (1950).
Example 20.1
2. Foust et al., Principles of Unit Operations, 2nd Edition, John
Wiley & Sons (1980). Example 15.1
3. McCabe, W. L., J. C. Smith, P. Harriott, Unit Operations of
Chemical Engineering, 4th Edition, McGraw-Hill (1985).
Example 16.1


Son book reviews

DISTILLATION TRAY FUNDAMENTALS
by M. J. Lockett
Cambridge University Press, Cambridge and
New York, 1986. 224 + xxiii pp. $54.50
Reviewed by
James R. Fair
The University of Texas at Austin
The distillation column continues to be the princi-
pal separation device for the chemical and petroleum
processing industries. For many years it was charac-
terized as a vertical, cylindrical column containing
plates or trays upon which rising vapor and descend-
ing liquid were brought into intimate contact, for pur-
poses of effective mass transfer. In recent years de-
signers of distillation columns have shifted some at-
tention to the use of structured (as opposed to ran-
dom) packing as vehicles for effecting intimate phase
contacting. However, such packing are considerably
more expensive than trays, and their cost is often jus-
tified only when their lower pressure drop carries an
economic advantage, as in some vacuum distillations.
Thus, the tray column remains as the standard and
basic device for vapor-liquid contacting.
Despite the title, this book deals not only with


trays for distillation services but also covers applica-
tions in absorption and stripping. It covers all of the
important aspects of tray design, those of a more hy-
drodynamic nature as well as those relating directly
to the mass transfer propensity of the two-phase mix-
ture on the tray. Considerable space is devoted to the
characterization of this mixture: foam, emulsion,
froth, spray, and so on. The overall coverage is quite
complete, with no detail of design left unaddressed.
Such important topics as phase flow distribution, ca-
pacity limits, pressure drop and interphase mass
transport are dealt with on quantitative bases. A com-
mendable effort has been made by the author to con-
sider all historical approaches (mostly empirical) that
deal with the various design parameters. The litera-
ture coverage is near exhaustive, and the reader will
not find elsewhere as complete a bibliography on the
distillation tray as is provided here. For each design
consideration, a method with some fundamental and
mechanistic support is provided-and for practition-
ers of distillation system design this is a welcome ad-
vance from the art and empiricism that have often
prevailed.
There are some limitations to the treatment that
should be mentioned. First, the author has not always
found it possible to make a forthright recommendation
when several alternate models or procedures are
available for a particular design step. The reader
must then make his own choice. Also, despite the title,
all tray-type contacting devices are not considered.
There is very little on valve trays, essentially nothing
on bubble-cap trays, and complete silence on dualflow
trays (those without downcomers). Emphasis is
clearly on crossflow sieve trays, but this is not all bad.
While there are still many bubble-cap tray columns in
operation, very few new ones have been designed dur-
ing the last few decades. The valve-tray is really a
proprietary contactor, with design often left up to the
proprietor. The dualflow tray is a rather specialized
device (and tricky to design), used mostly for fouling
services. On the other hand, the sieve tray is an effi-
cient and relatively inexpensive non-proprietary de-
vice that has been the object of many basic studies,
and its simple geometry (in effect, one or more sheets
of perforated metal, joined to a downcomer for hand-
ling liquid passage) makes it reasonably amenable to
fundamental modeling. Still, the title might have read
"Distillation Sieve Tray Fundamentals."
The book might have been improved by the inclu-
sion of some worked-out design examples, some ad-
vice on laboratory or pilot plant scaleup procedures,
and an author index. Still, the development of ra-
tional, fundamental-based approaches to the handling


CHEMICAL ENGINEERING EDUCATION









of complex two-phase mixtures, as are found on trays,
is refreshing and encouraging. The author has an ex-
tensive background in the research, testing and mod-
eling of distillation devices, and his authoritative text
reads very well. There is no equal to the book pre-
sently on the market. Anyone concerned with the de-
sign or analysis of distillation, absorption or stripping
columns of the tray type will want to take advantage
of the modern approaches presented in this book. EO

GAS SEPARATION BY
ADSORPTION PROCESSES
by Ralph T. Yang
Butterworth Publishers,
80 Montvale Avenue, Stoneham, MA 02180;
352 pages, $52.95 (1986)
Reviewed by
D. M. Ruthven
University of New Brunswick
The importance of adsorption as a separation pro-
cess in the chemical and petroleum industries has in-
creased dramatically in recent years, but the subject
is still not covered in any significant way in most
chemical engineering curricula. There have been three
recent books on the subject: Principles of Adsorption
and Adsorption Processes, by this reviewer, pub-
lished by Wiley in 1984; Large Scale Adsorption and
Chromatography, by P. C. Wankat, published by
CRC Press in 1986; and the present volume by Ralph
Yang, published by Butterworth Publishers in 1987.
None of these is really a textbook in the formal sense,
but any of them could be used as the basis for a
graduate level (or possibly a final year elective) course
on the subject.
The coverage of the present volume is broadly
similar to that of Principles of Adsorption and Ad-
sorption Processes, and there is considerable overlap,
which is probably inevitable since many of the source
references are common. The emphasis is, however,
different-reflecting the different areas of interest
and expertise of the authors. The book provides a
coherent and comprehensive account of the subject,
including the basic physico-chemical principles as well
as process technology. Although the title is Gas Sep-
aration by Adsorption Processes (and this is indeed
the main focus), liquid phase separation processes
such as the "Sorbex Process" and parametric pumping
are also covered briefly. As with its predecessors, the
emphasis is on fundamentals rather than on technolog-
ical details, and the level of background knowledge
which is assumed is also similar.
The book is divided into eight chapters: 1. Intro-


duction; 2. Adsorbents and Adsorption Isotherms; 3.
Equilibrium Adsorption of Gas Mixtures; 4. Rate Pro-
cesses in Adsorbers; 5. Adsorber Dynamics-Bed
Profiles and Breakthrough Curves; 6. Cyclic Gas Sep-
aration Processes; 7. Pressure Swing Adsorption-
Principles and Processes; 8. Pressure Swing Adsorp-
tion-Models and Experiments.
It is not a book for the undergraduate, but it
should be easily understood by graduate students and
those with some experience in research and develop-
ment. Since most adsorption processes operate under
transient conditions, some familiarity with partial dif-
ferential equations is needed to follow the sections
dealing with column dynamics and process modeling
(Chapters 5 and 8). I found Chapter 3, which contains
a good review of the various approaches to the corre-
lation and prediction of multicomponent adsorption
equilibria, and Chapters 7 and 8, which provide an
authoritative summary of PSA technology and model-
ing, to be most useful.
There is no discussion of membrane separation
processes which compete directly with pressure swing
adsorption in a number of applications. While mem-
brane separations may not be included within the nar-
rower definition of adsorption processes, some such
discussion would have been useful to allow the reader
to assess the relative merits of either approach, par-
ticularly in view of the publisher's claim that this is a
"complete treatise covering all aspects of adsorption
processes ."
Inevitably in a book which covers such a wide
range of subjects, one can expect controversy over
the treatment of certain topics. For example, in the
discussion of surface diffusion and intracrystalline dif-
fusion in zeolites (pp. 113-121) it should probably have
been pointed out that the kinetic treatment (which is
emphasized) and the quasi-thermodynamic treatment
(which is criticized) are not necessarily in conflict, but
merely represent different ways of looking at the
same phenomenon. The advantage of the quasi-ther-
modynamic treatment is that it allows meaningful
transport co-efficients to be derived without know-
ledge of the detailed diffusion mechanism. This may
not be obvious to the casual reader. Such criticisms
are, however, minor, and any lack of balance is more
than offset by the advantages in the presentation of
coherent perspective.
Taken as a whole, the book presents a concise and
readable summary of the voluminous literature of the
subject. It will no doubt become required reading for
those working in this area, both in universities and in
industry. At US $52.95 it is (just about) within the
affordable price range for individuals. [


SPRING 1988










mn lecture


THE MYSTIQUE OF ENTROPY


B. G. KYLE
Kansas State University
Manhattan, KS 66506


STUDENTS OF thermodynamics soon learn to ap-
preciate the utility of entropy in making various
calculations involving process heat and work effects.
These are direct applications found in engineering and
physical science. However, in addition to these quan-
titative applications, one finds qualitative and
metaphoric uses of entropy in a wide diversity of
fields. The extent to which the concept of entropy has
suffused contemporary thought is well illustrated by
Lord C. P. Snow's assertion that any definition of cul-
ture should include a technical component and that an
understanding of the second law of thermodynamics
is the cultural equivalent of a familiarity with the
works of Shakespeare [1].
Igoring incidental uses, a few examples will be pre-
sented here in which the concept of entropy is central
to the development of a theme or is thought to provide
insight. The intent is to illustrate the pervasiveness
of the fascination evoked by the entropy concept-the
entropy mystique.

COSMOLOGY
Of the many formulations of the laws of ther-
modynamics, the boldest and most provocative was
advanced in 1865 by Rudolf Clausius:

The energy of the universe remains constant.
The entropy of the universe tends to a maximum.

Here Clausius has taken concepts arising from
limited, earth-bound experience and with a great leap
of imagination has vested them with cosmic signifi-
cance. While both statements are daring, the second-
law statement has sparked the most interest and con-

... a few examples will be presented in which
the concept of entropy is central to the development
of a theme or is thought to provide insight. The intent
is to illustrate the pervasiveness of the fascination
evoked by the entropy concept-
the entropy mystique.


Benjamin G. Kyle is professor of chemical engineering at Kansas
State University and is enjoying his thirtieth year of teaching. He holds
a BS from the Georgia Institute Technology and a PhD from the Univer-
sity of Florida. He has not outgrown an early fascination with ther-
modynamics and is interested in practically all aspects of the subject.
He is the author of a recently published (Prentice-Hall) thermodynamics
textbook.

troversy because it leads to speculations regarding the
birth and death of the universe.

Birth and Death of the Universe
By definition, the universe must be a closed sys-
tem, and the uncritical extrapolation of our terrestrial
experience would suggest that its entropy is increas-
ing toward a maximum. Because an increase in en-
tropy is associated with a decreased ability to perform
work, the second law implies that the universe will
ultimately reach a dead state referred to as thermal
death. This dead state is an equilibrium state in which
all thermodynamic potentials have been leveled and
processes yielding work are no longer possible.
If one accepts Clausius' statement of the second
law with its implication of thermal death, then by the
following simple argument one can show that the uni-
verse had a beginning. A maximum is characterized
by a first derivative of zero. However, the derivative,
or rate of change, is finite until the maximum is
reached. Thus, if the entropy of the universe is tend-
ing to a maximum, it is doing so at a finite rate and
will reach its goal in a finite time. A universe of infi-
C Copyright ChE Division ASEE 1988


CHEMICAL ENGINEERING EDUCATION









nite age would have already reached its equilibrium
state of maximum entropy, and since this is not the
case, the universe must have a finite age. Thus, the
universe had a beginning and we have a position that
is at least congenial to the inference of creation and a
creator. Undoubtedly, the inclination to draw this in-
ference is strengthened by a scientifically more ac-
ceptable theory originated by Lemaitre in 1927 and
more recently popularized [2] as the "Big Bang"
theory.
The widely published English prelate, William
Inge, eagerly accepted the proof of a creator as im-
plied by the "law of entropy" [3]. Inge used the pros-
pect of the thermal death of the universe to argue that
God the Creator could not be merely a pantheistic god
found only in nature because such a god would be
under an inexorable death sentence. Such would not
be the fate of the transcendent god of Christianity.

It is not Christianity but modern pantheism and the myth
of unending progress which are undermined by the de-
gradation of energy.

Most notable among those espousing the creation view
was Pope Pius XII, who stated that Clausius' law of
entropy provides "eloquent evidence of the existence
of a Necessary Being." [4]
An argument against the inference of creation
from the laws of thermodynamics was advanced by
the Russian physicist, I. P. Bazarov [5], based on the
dialectic materialism of Engels. Identifying flaws in
the thermal death argument for the existence of a
creator, Bazarov points to the unfounded assumption
that the laws of thermodynamics apply to the entire
universe and to Engels' argument that the creation
implied by the second law would be in violation of the
first law. In Engels' words, those advocating the pro-
creation view saw the universe as winding down and
thus assumed that an initial winding up had been pro-
vided by "a stimulus from without." However, the
process of winding up imparted energy to the universe
and hence the total energy of the universe has not
always been constant as required by Clausius' first-
law statement.
It now appears that Engels' argument of a first-
law violation may be vitiated by recent calculations
that indicate the total of all energy in the universe is
zero [6]. This is possible because energy of motion and
the energy equivalent of mass are positive while
energy of gravitational or electromagnetic attraction
is negative. A zero energy sum would imply that cre-
ation could have occurred without an energy input and
brings to mind the theological doctrine of creation ex
nihilo.


The theory of relativity shows that time and
space are part of the structure of the universe and
not an empty stage on which physical events are
enacted. However, within this purely physical
picture there remains the basic question
of the direction of time.

Bazarov's criticism regarding the unwarranted ex-
tension of the laws of thermodynamics to the entire
universe seems well taken. This is especially so con-
sidering that on a cosmic scale the predominant
energy effects are associated with gravitation and
radiation, types of energy which are usually neglected
when applying thermodynamics in its usual terrestrial
context. In spite of this obvious incongruity, the ques-
tion is still unresolved for today one finds astrophysi-
cists concerned with entropy changes of various cos-
mic processes in which quantum and relativistic ef-
fects predominate [6].

Time
Of all the basic quantities in the physical world,
time is the most elusive. This is because we not only
experience it as an abstraction useful in physical sci-
ence, but also in a psychological sense. Because the
latter lies beyond the realm of physics, any strictly
physical description of time can never be entirely
satisfactory.
The theory of relativity shows that time and space
are part of the structure of the universe and not, as
pictured by classical physics, an empty stage on which
physical events are enacted. However, within this
purely physical picture there remains the basic ques-
tion of the direction of time. The equations of classical
mechanics, quantum mechanics, and relativity theory
are symmetric in time and do not preclude the rever-
sibility of processes. On the other hand, we know that
all naturally occurring processes are irreversible and
hence travel only in what we have designated as the
forward direction of time. For this reason the second
law of thermodynamics is said to define the "arrow of
time." Time increases in the direction of increasing
entropy.
The problem of reconciling the reversibility of the
microscopic world of colliding molecules obeying the
laws of classical mechanics with the irreversible be-
havior of the macroscopic world was first undertaken
by Boltzmann. This search for the origins of irreversi-
bility provides an interesting chapter in the annals of
science [7] and has led to the development of statistical
mechanics, an area of science which has proved quite
useful in the calculation of thermodynamic and trans-
port properties. However, despite this practical suc-


SPRING 1988









cess, the origins of irreversibility have yet to be deter-
mined to the satisfaction of scientists of a philosophical
bent [8].

LIFE AND EVOLUTION
At present there is disagreement among scientists
as to whether living systems can be completely de-
scribed by the laws that apply to inanimate matter or
whether additional, but as yet undiscovered, laws ap-
plying only to living matter are also needed. However,
the various identifiable physical and chemical proces-
ses occurring within living systems require no addi-
tional laws for their explication and therefore conform
to the laws of thermodynamics. Additionally, it is
known that the laws of thermodynamics are obeyed
by each living system as it interacts with its environ-
ment. Our present societal obsession with diets and
calorie counting serves as a monotonous reminder of
our bondage to the first law, but the dictates of the
second law are not so obvious and, in fact, may seem
counter-intuitive. The conceptual difficulty appears
when we regard life as a striving to maintain order
and the second law as a principle of degradation. This
apparent conflict is due to our failure to recognize liv-
ing systems as open systems. In an open system the
combined entropy of the system and surroundings
must increase. Thus, a living system can maintain it-
self or grow, and thereby decrease its entropy, if the
surroundings undergo the appropriate increase in en-
tropy. This leads directly to the statement that life is
sustained by the conversion of low-entropy resources
into high-entropy wastes. The ultimate origin of low-
entropy resources is the sun which drives the photo-
synthetic mechanism in plants whereby solar energy
is converted into chemical energy and stored as carbo-
hydrates.
Despite reconciliation of the life force with the law
of entropy, there is still a persistent undercurrent of
nagging doubt when the origin and evolution of life
are considered. These activities imply purpose, a goal
that seems totally opposed by the entropy principle.
This contradiction was recogized by Teilhard de Char-
din in formulating his grand scheme of evolution to-
wards an Omega Point of pure consciousness and
ecstatic union with God [9]. Because of the law of
entropy with its opposing tendencies and prospect of
thermal death, Teilhard postulated two classes of
energy: tangential and radial. Tangential energy is
simply the energy identified by physical science and
subject to the laws of thermodynamics. On the other
hand, radial energy is of a psychic or spiritual nature
and is not subject to the laws of thermodynamics. The
evolution of life toward the Omega Point is said to be


driven by radial energy and thus is free of second-law
restrictions. The relationship of the two classes of
energy is tenuous and is best illustrated by Teilhard's
words:

To think, we must eat. But what a variety of thoughts we
get out of one slice of bread! Like the letters of the alphabet,
which can equally well be assembled into nonsense as into
the most beautiful poem, the same calories seem as indif-
ferent as they are necessary to the spiritual values they
nourish.

Teilhard was a Jesuit priest as well as a scientist,
but he wanted his work to be judged on the basis of
its scientific merit. His Phenomenon of Man is a
monumental attempt to meld science and metaphysics
to provide an answer to the question of meaning. De-
spite his heroic effort to fuse faith and reason, science,
because it recognizes only reason, could do nothing
but render a negative verdict. Nevertheless, his work
has wide appeal today, some thirty years after his
death. Perhaps this is because the soul of a poet and
mystic shows through. After all, who but a poet or
mystic would dare address such questions.

SOCIAL SCIENCE
Henry Adams, the noted American historian and
man of letters, was the first to view the course of
civilization (i.e., history) from a second-law perspec-
tive [10]. In the opening years of the twentieth cen-
tury he expounded a theory of history in which the
Social Energy of civilization is continually dissipated.
According to Adams, the second law of ther-
modynamics required that, ". the higher powers of
energy tended always to fall lower, and that this pro-
cess had no known limit." In the category of energy
Adams included the Vital Energy of an individual and
the Social Energy of society and stated that, "The law
of entropy imposes a servitude on all energies, includ-
ing the mental." All this naturally resulted in a pes-
simistic prognosis for civilization where ". .. the ash-
heap was constantly increasing in size."
According to one of its most famous practitioners,
Claude Levi-Strauss, the field of anthropology could
appropriately be called "entropology." Levi-Strauss
[11] sees man and his civilizations as "instruments in-
tended to create inertia, at a rate and in a proportion
infinitely higher than the amount of organization they
involve." He further states

Thus it is that civilization taken as a whole, can be de-
scribed as an extraordinary complex mechanism, which we
might be tempted to see as offering an opportunity of sur-
vival for the human world, ifits function were not to produce


CHEMICAL ENGINEERING EDUCATION










what physicists call entropy, that is inertia. Every verbal ex-
change, every line printed, establishes communication be-
tween people, thus creating an evenness of level, where be-
fore there was an information gap and consequently a
greater degree of organization.

We find familiar the concept that the net result of
human life is an increase in entropy, but the idea of a
leveling of social organization through communication
is a bit unexpected and would appear to be a
metaphoric extension of the thermal death concept.
The second law of thermodynamics has been shown
to provide a realistic perspective for economics [12].
According to Nicholas Georgescu-Roegen, the conven-
tional view of economic process as circular and time-
less ignores the increase in entropy accompanying
every human endeavor. Georgescu-Roegen sees
economic activity as turning low-entropy inputs into
high-entropy outputs.

All species depend on the sun as their ultimate source of
low entropy except man, who has learned also to exploit the
terrestrial stores of low entropy such as minerals and fossil
fuels. Life feeds on low entropy and so does economic life.
Objects of economic value such as fruit, cloth, china, lum-
ber, and copper, are highly ordered, low-entropy structures.
For low entropy is the taproot of economic scarcity.

He observes that economic development as presently
practiced is based on rapacious consumption of our
terrestrial dowry of low-entropy materials and con-
cludes that this can not be indefinitely sustained.
Using the entropy-conscious economic views of
Georgescu-Roegen, Herman Daly [13] argues persua-
sively against the fallacy of perpetual growth and even
suggests that growth itself may be illusory when mea-
sured by the GNP. This closely watched economic in-
dicator includes, besides the value of goods produced
and services rendered, all costs associated with pro-
duction, including pollution control costs. As the
richer and more accessible resources are preferen-
tially consumed, the remaining lower-quality re-
sources require higher production costs and thus a
rise in GNP may reflect increased cost of production
rather than increased level of production. This is the
entropic factor largely ignored by conventional
economics. As an alternative to growth, Daly has out-
lined in some detail the workings of a steady-state
economy.
The work of Georgescu-Roegen has also inspired
Jeremy Rifkin's Entropy: A New World View [14] in
which the entropy law is credited with singlehandedly
undermining the Newtonian-Cartesian mechanistic
world view. Rifkin believes that a new world view
based on the entropy law would expose the fallacy in


our present obsession with growth and would allow us
to meet the future with awareness and acceptance of
nature's constraints. This change in outlook would af-
fect practically every area of human endeavor ranging
from technology to religion and could move us toward
the long-sought but ever-distant utopian dream.
There is no doubt that Rifkin has overstated his case
in a single-minded and over-simplified manner, but
exaggeration is a forensic device which often becomes
acceptable when used in passion for a worthy cause.
Because the inculcation of an entropy-conscious world
view is considered by many to be such a cause,
perhaps Rifkin should not be judged too harshly for
his excesses.

LITERATURE AND ART
Metaphoric entropic themes abound in modern fic-
tion [15] as well as in serious science fiction [16]. Most
applications are implicit, although several writers
explicitly use the term entropy in their work. These
writers are predominately American and include John
Barth, Donald Barthelme, Saul Bellow, Stanley Elkin,
Norman Mailer, Walker Percy, Thomas Pynchon, and
John Updike. As expected of fertile imaginations, one
finds many variations of the entropic theme whether
used explicitly or implicitly. The theme has been
applied to either individuals or entire societies, some-
times within the context of a closed system, with ulti-
mate states as extreme as chaos and stagnation.
Norbert Wiener's famous book on cybernetics [17]
first appeared in 1950 and was undoutedly instrumen-
tal in popularizing the application of the entropy con-
cept to the many aspects of the human condition as
found in modern literature. Moreover, Wiener's iden-
tification of entropy as a measure of the information
content of messages stimulated writers to examine the
very act of writing [18]. In Wiener's words

Messages are themselves a form of pattern and organiza-
tion. Indeed, it is possible to treat sets of messages as having
an entropy like sets of states of the external world. Just as
entropy is a measure of disorganization, the information
carried by a set of messages is a measure of organization. In
fact, it is possible to interpret the information carried by a
message as essentially the negative of its entropy, and the
negative logarithm of its probability. That is, the more prob-
able the message, the less information it gives. Cliches, for
example, are less illuminating than great poems.

The idea behind this passage is easily grasped
when applied to the transmission of factual data (e.g.,
a seven-digit telephone number) or a coded message
(e.g., one if by land, two if by sea) but quickly becomes
fuzzy when considering the transmission of subtle or


SPRING 1988










One is left wondering why the entropy concept, which offers so little insight, has been so widely employed
outside its narrow scientific context and why any serious artist or writer would even consider the application
of scientific principles to the very act of creating. There does indeed appear to be an entropy mystique .


abstract ideas or the more vital aspects of communica-
tion inherent in the sender and receiver. This becomes
apparent when the idea is pushed to its limit. The
writer, in an attempt to counter the entropic effect of
banality, looks for unexpected or improbable ways of
using words and thus runs the risk of verbal chaos.
The dilemma is articulated by Lewicki [12]:

In order to avoid entropy, a writer must therefore walk a
narrow path between the danger of producing probable mes-
sages of low informational value (such as, for example, new
versions of old themes, written in a conventional manner),
and the risk of turning out incomprehensible bodies of words
that would seem disorderly to the reader.

Rudolph Arnheim [19] has recognized the con-
tradictions in information theory as applied to art:

Here order is described as the carrier of information, be-
cause information is defined as the opposite of entropy, and
entropy is a measure of disorder. To transmit information
means to induce order. This sounds reasonable enough.
Next, since entropy grows with the probability of a state of
affairs, information does the opposite: it increases with its
improbability. The less likely an event is to happen, the more
information does its occurrence represent. This again seems
reasonable. Now what sort of sequence of events will be least
predictable and therefore carry a maximum of information.
Obviously a totally disordered one, since when we are con-
fronted with chaos we can never predict what will happen
next. The conclusion is that total disorder provides a maxi-
mum of information; and since information is measured by
order, a maximum of order is conveyed by a maximum of dis-
order.

Arnheim attributes the contradiction to a failure
to distinguish order from structure:

Any predictable regularity is termed redundant by the in-
formation theorist because he is committed to economy:
every statement must be limited to what is needed. He shares
this commitment with scientists and artists; its meaning,
however, depends on whether one chops up patterns into
elementary bits or whether one treats them as structures. A
straight line reduced to a sequence of dots for the purpose of
piecemeal analysis or transmission can be highly redundant;
in the drawing ofa geometrician, engineer, or artist it is not.

In explaining the creative process underlying a
work of art, Arnheim sees an interplay between the
anabolic tendency which establishes the structural
theme and the entropic principle which is manifested
both as a catabolic destruction of patterns, or disor-


during, and a simplification or ordering due to tension
reduction. Here we see the entropy principle invoked
metaphorically and identified simultaneously with two
opposing tendencies: ordering and disordering.

COMMENTARY
Of all the properties of matter, entropy is most
difficult to conceptualize and seems the most con-
trived. It cannot be evaluated directly from an experi-
mental measurement, but must be calculated from a
somewhat arbitrary computational path and thus its
existence and evaluation are closely tied to human ac-
tivity. It is tainted with a human scent. Perhaps this
is a reason this most anthropomorphic of all the con-
cepts of science has fascinated the human mind.
Despite its mystique, entropy actually provides
very little insight into the mysteries of nature. This
can be appreciated by recalling the familiar textbook
derivations. A Carnot cycle is used to define entropy
and demonstrate that it is a state property. Next, an
isolated system undergoing a spontaneous process is
considered and a simple argument shows that an in-
crease in entropy results. Ramifications of this result
are extremely useful for the detailed calculations
made by engineers and scientists, but in what way
has it improved our general understanding of nature?
Are not the following statements equivalent and
equally enlightening?

A. Every spontaneous process tends toward a condition of
deterioration or stagnation.
B. Entropy is the measure of deterioration or stagnation.
Every spontaneous process results in an increase in en-
tropy.

Statement A is a direct statement of experience.
Statement B is expressed in the language of an empir-
ically based science and is therefore merely an indirect
but more formalized statement of experience.
The idea of things running down, which can now
be proclaimed as a decree of science, has been found
to be a dominant theme in ancient myths [20]. This
condition also has been lamented by Hesiod, the Greek
poet of the 8th century B.C., who told of the five
descending ages of Man beginning with a pristine age
of gold and ending in his own worldly age of iron.
Also, somewhat later, we find St. Paul [21] referring
to a coming time of glory when "creation itself will be
set free from its bondage to decay." These may be


CHEMICAL ENGINEERING EDUCATION










regarded as generalizations of statement A and ap-
pear more naive than the B-based concept of thermal
death. Yet, even though we favor statement B today,
let us not be deceived into thinking it more insightful
than statement A.
With regard to fixing the direction of time, it is
also instructive to consider two alternative state-
ments.
A. In the usual sense of before and after, spontaneous pro-
cesses proceed in the forward direction of time.
B. All spontaneous processes are accompanied by an in-
crease in entropy. The forward direction of time is the
direction in which the entropy increases.
While the practical person would give no advan-
tage to statement B, many philosophers prefer some-
thing similar to it because it appears to have elimi-
nated some of the subjective element that clings to
statement A. Their object is to define time without
reference to the human mind [22]. However, in this
respect an obvious problem arises in the actual deter-
mination of entropy. This requires a human mind to
devise a reversible path between the initial and final
states which is essential to the calculation of the en-
tropy change. There is no entropy meter and the in-
voking of entropy, which is itself a construct of the
human mind, does little to remove the human scent
from the concept of time.
To place the question of time in perspective we
should recognize that it is a construct of the conscious
mind. The subconscious mind, the Freudian id, has no
awareness of time-a fact known to science and easily
verified personally by recalling the particulars of our
dreams. What we have done is to construct from the
regular rhythms of the universe (e.g., the motion of
the earth about its axis or about the sun) a lifeless,
linear time scale upon which we can place in monotonic
order the events of our physical world. This seems so
natural that it comes as something of a shock to learn
that it is possible to have a valid and effective world
view that does not include the concept of time. An
outstanding example of this is the language of the
Hopi indians of the American southwest which con-
tains no reference to "time" either explicitly or im-
plicitly. This language is capable of accounting for and
describing correctly, in a pragmatic or operational
sense, all observable phenomena in the universe with-
out the mental construct of time [23].
The avowed aim of science is to establish the basis
of physical reality. However, two different interpreta-
tions exist: external and internal. The external ver-
sion is the older, traditional view in which the human
mind objectively probes and observes nature and
thereby discovers natural laws. Recent developments


in physics, however, have cast doubt on the concept
of an objective observer independent of the system
observed [24], and this has led to the view that science
merely creates a set of interlocking laws which pro-
vides a consistent description of nature. This descrip-
tion usually involves the use of abstractions and men-
tal constructs far removed from our everyday experi-
ence and may thus be termed a view of physical reality
internal to the mind [25]. The previous discussion of
time and entropy is obviously an argument for the
internal view.
One is now left wondering why the entropy con-
cept, which offers so little insight, has been so widely
employed outside its narrow scientific context and
why any serious artist or writer would even consider
the application of scientific principles to the very act
of creating. There does indeed appear to be an entropy
mystique, although the reasons for it will not be
explored. Perhaps the following verse is appropriate.

ENTROPY
For thermodynamics we feel a sense of awe
And marvel at the power of those early minds
Who showed that mundane matters lead to
natural law
And boldly stated that the universe unwinds.
The second law is never denied its due
In science, belles-lettres, and philosophy.
It pales the onward-and-upward view,
For no one is consoled by entropy.
While nothing temporal eludes its iron rule
And most would take decay's decree as true,
The sage's bane can be the builder's tool
As entropy shows the best that we can do.
This useful concept prompts a primal groan:
A dread we've only named but always known.

REFERENCES
1. Snow, C. P., The Two Cultures: And a Second Look, The
New American Library, New York, 1964.
2. See, for example, Weinberg, S., The First Three Minutes,
Fontana, 1978.
3. Inge, W. R., God and the Astronomers, Longmans, Green &
Co., London, 1934.
4. Address to the Pontificial Academy of Sciences, Rome 1951,
reprinted in English in Bulletin of the Atomic Scientists, 8,
143 (1952).
5. Bazarov, I. P., Thermodynamics, pg 74, Macmillan, New
York, 1964.
6. Davies, Paul, God and the New Physics, Ch. 3, Simon and
Schuster, Inc., New York, 1983.
7. Brush, S., The Kind of Motion We Call Heat, North Holland
Pub. Co., Amsterdam, 1976.
8. For a review of this problem, see Hollinger, H. B. and M. J.
Continued on page 102.


SPRING 1988











classroom


DISCRETE-EVENT SIMULATION IN

CHEMICAL ENGINEERING


DANIEL J. SCHULTHEISZ and
JUDE T. SOMMERFELD
Georgia Institute of Technology
Atlanta, GA 30332-0100

R ECENT YEARS HAVE witnessed a rapid and dra-
matic change in the nature of the chemical pro-
cess industries in the developed countries of the
world. Specifically, there has been an intense revival
of commercial interest in batch chemical processes,
such as those employed in the manufacture of fine and
specialty chemicals, at the expense of traditional con-
tinuous steady-state processes for the manufacture of
commodity chemicals. One large British chemical com-
pany reports that specialty chemicals manufactured
by batch processing contributed over 30% to their
total profits in 1983 as opposed to 18% in 1977 [1].
Certainly, one of the primary driving forces for this
change has been the recent commissioning of many
world-scale commodity chemicals plants in various de-
veloping countries.

SIMULATION SYSTEMS
Concomitant with these industry changes, signifi-
cant developments have occurred in the modeling and
simulation of chemical processes. To be sure, usage
(including academic) and development of conventional
steady-state process simulators continue at an active
level. Thus, the FLOWTRAN system [2] developed
by the Monsanto Company was made available to
chemical engineering schools in 1973 and has been ex-
tensively employed for educational purposes ever
since [3]. Subsequently, newer steady-state process
simulation systems such as PROCESS, ASPEN
PLUS and DESIGN II became available to academic
users.

Discrete-event simulators were originally
developed as numerical aids to solve complex
queuing theory problems which were not amenable
to analytical solution. Such problems occur
routinely in the field of industrial engineering .

Copyright ChE Division ASEE 1988


Daniel J. Schultheisz received his bachelor's degree in chemical
engineering from the University of Pennsylvania and is currently com-
pleting the requirements for his master's degree in chemical engineer-
ing at Georgia Tech. For his thesis, he is preparing an instructor's
manual of exercises in chemical engineering using GPSS, to be distri-
buted and marketed by the CACHE Corporation. (L)
Jude T. Sommerfeld is professor and associate director of the School
of Chemical Engineering at Georgia Tech. He received his BChE degree
from the University of Detroit and his MSE and PhD degrees, also in
chemical engineering, from the University of Michigan. His 25 years
of industrial and academic experience have been primarily in the area
of computer-aided design, and he has published over seventy articles
in this and other areas. (R)

We have also witnessed the development and ap-
plication of various simulators for batch chemical pro-
cesses in recent years. These developments have in-
cluded both discrete-event and combined (discrete +
continuous) systems, as employed in the industrial en-
gineering field. There is an unfortunate confusion in
terminology here: the industrial engineering interpre-
tation of the term 'continuous' is not the same as that
associated with chemical engineering usage, namely,
steady-state operation. Rather, the industrial en-
gineering meaning of continuous should be construed
by chemical engineers as dynamic or unsteady-state.
The progenitor of discrete-event simulation sys-
tems is GPSS [4], which dates back to 1959 and is still
used extensively in many manufacturing sectors. Be-
cause of its easy use, availability, reliability, and effi-
cient operation (integer arithmetic only in many ver-
sions), GPSS is a very effective tool if only discrete
simulation capability is required. Other popular dis-


CHEMICAL ENGINEERING EDUCATION











TABLE 1
Recent Applications of GPSS to Chemical
and Allied Processes


APPLICATION
Two batch reactors in parallel followed by a batch
still
DDT manufacture
Chocolate manufacturing
Sequence of batch distillation columns
Large-scale poliomyelitis vaccine production
Choline chloride manufacture
Polyvinyl chloride (PVC) manufacture
Sequencing batch reactors (SBR) for wastewater
treatment


crete-event simulation systems include S
(more prevalent in Europe) and SIMSCG
general, however, there are not many pub
cations of discrete-event simulation system
chemical processes. Morris [7], for exam
scribed a very simple application of GPS
process comprised of two reactors in para
by a still. Other recent applications of GP
ical and allied batch (or semi-continuous) p
listed in Table 1.

DISCRETE-EVENT SIMULATION
Discrete-event simulators were or
veloped as numerical aids to solve comp
theory problems which were not amenable


TABLE 2
Terminology in Usage of the FLOWTR
GPSS Simulators


ITEMS
Precoded functional
subroutines
Arguments of functional
subroutines
Items moving through
the model
Numerical characteristics
of moving items

Output quantities from
subroutines (other
than moving items)


FLOWTRAN
(Steady-State
Process
Simulator)
Blocks

Parameters

Streams

Properties ( e.g.,
temperature,
composition)
Retention vector
contents


(


[REF.]

[7]
[8]
[9]
[10]
[11]
[12]
[13]

[14]


IMULA [5]
RIPT [6]. In
ilished appli-
ems to batch
iple, has de-
S to a batch
llel followed
'SS to chem-
irocesses are


Despite their considerable differences in origin
and application, there are noteworthy similarities
among the various types of simulators ... For example,
FLOWTRAN and GPSS have a number of
preceded functional subroutines.


cal solution. Such problems occur routinely in the field
of industrial engineering and typical example applica-
tions include machine shops, customer service sta-
tions, and transportation networks.
Most discrete simulation systems have stochastic
capabilities for the scheduling of time events. To sup-
port this function, most such systems also have one or
more built-in random-number generators. Output
from the latter is used to sample event times (or dura-
tions between time events) from various probability
distributions. In GPSS, the only easy-to-use, built-in
distribution is the uniform or rectangular distribution.
Thus, for example, a service time can take the form,
A B, where A represents the mean value and B is
the half-width, in appropriate time units, of the distri-
bution.

SIMILARITIES IN SIMULATORS


Despite their considerable differences in origin and
application, there are noteworthy similarities among
the various types of simulators described above. For
example, FLOWTRAN and GPSS have a number of
iginally de- preceded functional subroutines (generally written in
flex queuing FORTRAN). In both FLOWTRAN and GPSS, these
le to analyti- functional subroutines are known as blocks.
There is a number of other similarities between
these two systems, obscured only by the technical jar-
gon employed. In conventional steady-state chemical
AN and process simulators such as FLOWTRAN and PRO-
CESS, the items which move from one block to
another in the model are known as streams. Each in-
GPSS dividual stream has a set of properties (composition,
Discrete-Event temperature, pressure) associated with it, which are
Simulator) typically modified as the stream passes through a
Blocks block. In analogous fashion, the items which proceed
from block to block in a GPSS model are known as
Operands transactions. Also in analogy with stream properties
in a steady-state process simulator, GPSS transac-
Transactions tions have associated with them various parameters
(such as priority level or lifetime in the model) which
Parameters can be modified by the passage of the transaction
through certain blocks. In a GPSS model of a batch
Standard chemical plant, for example, transactions could repre-
numerical sent batches of material proceeding through the pro-
attributes cess. Table 2 summarizes these similarities and ter-
minology for the FLOWTRAN and GPSS simulators.


SPRING 1988










GPSS PROCESSOR
There are about thirty-five different blocks in
GPSS, roughly the same number as in the FLOW-
TRAN system. A listing of these GPSS blocks is given
in Table 3. It is common to construct block diagrams
in the development of GPSS models. In contrast with
FLOWTRAN where each block in such diagrams is
represented more or less by a rectangle, each differ-
ent functional block in a GPSS block diagram has its
own distinctive shape (see Schriber [4]).
Some of the GPSS blocks listed in Table 3 are quite
complicated and would typically be used only by more
sophisticated analysts. There are others, however,
which would be common to any GPSS model. Thus,
GENERATE blocks are used to provide transactions
to a model, much like a chemical engineer inputs feed
streams to a FLOWTRAN model. Conversely, a
transaction is removed from a GPSS model by a TER-
MINATE block.
There is a block named SPLIT in both FLOW-
TRAN and GPSS, but there is one fundamental differ-
ence between the two. In the FLOWTRAN system,
the sum of each extensive property over all of the
output streams from the block equals that property
for the incoming stream. In discrete-event simulation
with GPSS, however, the SPLIT block really per-
forms a cloning operation. That is, one or more identi-
cal offspring transactions are created from the single
parent transaction (which retains its existence) enter-
ing the block.


GPSS OUTPUT


As with the FLOWTRAN system which provides
a summary table of the streams passing through the
model and output results from each of the blocks in
the model, GPSS automatically prints out a variety of
output statistics at the conclusion of a simulation.
These statistics pertain primarily to the various


TABLE 3
Listing of GPSS Blocks


ADVANCE
ASSEMBLE
ASSIGN
BUFFER
DEPART
ENTER
FAVAIL
FUNAVAIL
GATE
GATHER
GENERATE


LEAVE
LINK
LOGIC
LOOP
MARK
MATCH
MSAVEVALUE
PREEMPT
PRINT
PRIORITY
QUEUE


RELEASE
RETURN
SAVEVALUE
SEIZE
SELECT
SPLIT
TABULATE
TERMINATE
TEST
TRANSFER
UNLINK


facilities, queues, and storage in the model.
Thus, from an inspection of the facility output
statistics from a GPSS simulation, an analyst might
find that the average holding time per transaction for
a given facility is considerably greater than the user-
supplied average service time for that facility. In a
chemical engineering application, for example, this
could indicate that a reactor, after finishing processing
of a batch (transaction), often cannot discharge the
batch because of an unavailable downstream facility.
The latter might correspond to a storage tank which
is full or another processing unit (e.g., still, centrifuge,
dryer) which is engaged. The regular occurrence of
such a situation would normally be accompanied by an
average utilization (fraction of total time busy) ap-
proaching unity for the original upstream facility and
would suggest the existence of some downstream
bottleneck. The existence of similar bottleneck situa-
tions can also be deduced from the output statistics
for GPSS storage. The productivity (number of
batches produced) of the modelled process is, of
course, related to the number of transactions passing
through the GPSS model.

EXAMPLE APPLICATION

Let us consider a very simple application of GPSS
to the modeling of a batch chemical process. This
example is an adaptation of a problem (number
2.41.14) presented by Schriber [4]; the process flow
diagram for this example is presented in Figure 1.



CUSTOMER STILL
ORDERS ------ O PRODUCT
QUEUE STORAGE
QUEUE TANK
THREE
IDENTICAL
REACTORS
FIGURE 1. Sketch of batch process for example problem.


Thus, a small, single-product batch chemical plant
has three identical reactors in parallel, followed by a
single storage tank and a batch still. Customer orders
(batches) to be filled (which begin with processing in
the reactor) occur every 115 30 minutes, uniformly
distributed. The reaction time in a given reactor is
335 60 minutes, and the distillation time in the still
is 110 25 minutes, both times uniformly distributed.
The holding capacity of the storage tank is exactly one
batch. Hence, the storage tank must be empty for a
given reactor to discharge its batch; if not, the reactor
cannot begin processing a new batch until the storage


CHEMICAL ENGINEERING EDUCATION


I

































FIGURE 2. GPSS block diagram for example problem.


tank becomes empty. The simulation is to be run for
100 batches. The model should have the capability to
collect waiting line statistics for the queue im-
mediately upstream of the reactor.
The GPSS block diagram for this example model
is shown in Figure 2. Note the distinctive shapes for
each of the blocks employed. The first executable
block is the GENERATE block, which creates trans-
actions representing customer orders (batches). These
transactions immediately queue up and attempt to
capture an available reactor via the ENTER block.
After capturing a reactor, a batch leaves the reactor
queue through the DEPART REACQ block, and is
processed in the ADVANCE 335,60 block. The batch
must first be able to enter the storage tank (ENTER
TANK block) before it releases its reactor in the
LEAVE REACT block. The batch then attempts to
capture the single still facility in the SEIZE STILL
block. Having accomplished such, the batch leaves the
storage tank, is processed in the still, releases the
latter, and finally leaves the model through the TER-
MINATE block. Selected output statistics from this
simulation are summarized in Table 4.
From Table 4, one sees that the batch still was in
use 91.1% of the time, and the average holding (pro-
cessing) time per batch was 108 minutes. The average
contents in the queue upstream of the reactors was
0.44 batch, and the average waiting time for all
batches, including ones which experienced no waiting,
in this queue was 50.5 minutes. The three reactors
were in use 95.2% of the time, and the average holding


time for a batch in a reactor was 329 minutes. Simi-
larly the storage tank (with a capacity of one batch)
was full 41.4% of the time, and the average holding
time therein was 48.6 minutes. Although not pre-
sented in Table 4, the total simulation time, to com-
pletely process 100 batches, was 11,967 minutes.
One can easily explore proposed modifications to
this process. Thus, one more reactor could be added
in an effort to increase productivity. One might find
as a result, however, a significant increase in the aver-
age reactor holding time beyond the nominal average
reaction time of 335 minutes. In this case, one could
explore increasing the intermediate storage capacity
(TANK) and/or improving the downstream distillation
operation.

SUMMARY
This article has attempted to serve as a brief intro-
ductory tutorial on discrete-event simulation, with
emphasis on chemical engineering applications. For
some simple batch process applications, only discrete
simulation capability is required. More complex appli-
cations would require usage of a combined (discrete
plus dynamic) simulation system, but knowledge of
the essential features of discrete-event simulation re-
mains useful background in such cases.

REFERENCES
1. Preston, M. L., and G. W. H. Frank, "A New Tool for Batch
Process Engineers--ICI's BatchMASTER," Plant/Opera-
tions Progr., 4, 217 (1985).
2. Seader, J. D., W. D. Seider, and A. C. Pauls, FLOWTRAN


TABLE 4
Selected GPSS Output from Example Simulation
of a Batch Chemical Process


Facility Statistics
(batch still, STILL):
Average utilization
Average holding time per batch, min
Queue Statistics
(reactor queue, REACQ):
Maximum contents
Average contents
Average waiting time (all batches), min
Storage Statistics:
Reactors (REACT):
Average utilization
Average holding time per batch, min
Storage Tank (TANK):
Average utilization
Average holding time per batch, min


0.911
108


3
0.44
50.5


0.952
329

0.414
48.6


SPRING 1988










Simulation-An Introduction, 2nd Ed., Ulrich's Bookstore,
Ann Arbor, MI, 1977.
3. Clark, J. P., and J. T. Sommerfeld, "Use of FLOWTRAN
Simulation in Education," Chem. Eng. Ed., 10, No. 2, 90
(1976).
4. Schriber, T. J., Simulation Using GPSS, Wiley, New York,
1974.
5. Franta, W. R., The Process View of Simulation, Elsevier
North-Holland, New York, 1977.
6. Wyman, F. P., Simulation Modeling: A Guide to Using
SIMSCRIPT, Wiley, New York, 1970.
7. Morris, R. C., "Simulating Batch Processes," Chem. Eng.,
90, No. 10, 77 (1983).
8. Blaylock, C. R., C. O. Morgan, and J. T. Sommerfeld, "GPSS
Simulation of DDT Manufacture," Proc. 14th IASTED Intl.
Conf. on Appl. Sim. and Mod., Vancouver, June, 1986, p.
314.
9. Passariello, I., and J. T. Sommerfeld, "GPSS Simulation of
Chocolate Manufacturing" in Tools for the Simulation Profes-
sion, Soc. for Computer Simulation, San Diego, 1987, p. 1.
10. Barnette, D. T., andJ. T. Sommerfeld, "Discrete-Event Simu-
lation of a Sequence of Multi-Component Batch Distillation
Columns," Comp. & Chem. Eng., 11, 395 (1987).
11. Kenvin, J. C., and J. T. Sommerfeld, "Discrete-Event Simula-
tion of Large-Scale Poliomyelitis Vaccine Production," Process
Biochem., 22, 74 (1987).
12. Bales, W. J., J. R. Johnson, and J. T. Sommerfeld, "Use of a
Queuing Simulator in Design of a Batch Chemical Production
System," Prodn. and Inventory Mgmt., accepted for publica-
tion (1987).
13. Adebekun, A. K., Z. Q. Song, and J. T. Sommerfeld, "GPSS
Simulation of PVC Manufacture," Polymer Process Eng., 5,
145 (1987).
14. Glenn, S. L., R. T. Norris, Jr., and J. T. Sommerfeld, "Dis-
crete-Event Simulation of Sequencing Batch Reactors (SBR)
for Wastewater Treatment," J. Water Poll. Cont. Fedn., sub-
mitted for publication (1987). El


ENTROPY
Continued from page 97.
Zenzen, The Nature of Irreversibility, D. Reidel Pub. Co.,
Dordrecht, Holland, 1985.
9. Teilhard de Chardin, P., The Phenomenon of Man, Harper
and Row, New York, 1975. A good summary of Teilhard's
arguments is presented in Barrow, J. D. and F. J. Tipler, The
Anthropic Cosmological Principle, Ch. 3, Clarendon Press,
Oxford, 1986.
10. Adams, Henry, The Degradation of Democratic Dogma, Re-
print of the 1919 ed., Harper Torchbooks, New York, 1969.
11. Levi-Strauss, C., Tristes Tropiques, translated by J. and D.
Weightman, Jonathan Cape, London, 1973.
12. Georgescu-Roegen, N., The Entropy Law and the Economic
Process, Harvard University Press, Cambridge, 1971.
13. Daly, Herman E., Steady-State Economics, W. H. Freeman
& Co., San Francisco, 1977.
14. Rifldn, J., Entropy: A New World View, Bantam Books, New
York, 1980.
15. Tanner, Tony, City of Words, Chap. 6, Harper and Row, New
York, 1971, and Lewicki, Zbigniew, The Bang and the
Whimper, Greenwood Press, Westport, CT., 1984.
16. Greenland, Colin, The Entropy Exhibition, Routledge and
Kegan Paul, London, 1983.


17. Wiener, Norbert, Human Use of Human Beings, Houghton-
Mifflin, Boston, 1950.
18. Whether entropy can be linked to the information in a message
is still an undecided question. The views of Brillouin and the
Denbighs typify the pro and con, respectively. Brillouin, L.,
Scientific Uncertainty and Information, Academic Press,
New York, 1964. Denbigh, K. G. and J. S. Denbigh, Entropy
in Relation to Incomplete Knowledge, Cambridge University
Press, Cambridge, 1985.
19. Arnheim, Rudolf, Entropy and Art, University of California
Press, Berkeley, 1971.
20. de Santillana, G., and H. von Dechend, Hamlet's Mill, Gambit
Inc., Boston, 1969.
21. Romans 8:21.
22. For lucid and concise discussion of the philosophical problem
of time, see Denbigh, K. G., Three Concepts of Time,
Springer-Verlag, Berlin, 1981.
23. Whorf, B. L., Language, Thought and Reality, M.I.T. Press,
Cambridge, 1956.
24. See, for example, Capra, F., The Tao of Physics, Bantam
Books, New York, 1975.
25. Both the external and internal views are discussed by Morris
Kline in Mathematics and the Search for Knowledge, Oxford
University Press, Oxford, 1985. The internal view is examined
by Roger S. Jones in Physics as Metaphor, New American
Library, New York, 1982. O



EDUCATOR: Bailey
Continued from page 61.

pitchers helps my students know each other better,"
Jay notes. "It also helps me maintain a friendly and
open relationship with my group that's important in
our work together."
Research by Bailey and his students was recog-
nized by the Curtis W. McGraw Research Award of
the American Society of Engineering Education in
1983, by Jay's election to the National Academy of
Engineering in 1986, and by the AIChE Professional
Progress Award in 1987.
Bailey does have interests outside of the lab.
Everyone who knows him remarks on his devotion to
Sean, his 18-year-old son, who's now a freshman at
the University of Colorado, Boulder. Jay's an avid
amateur musician-the guitar is his instrument-and
he loves active sports such as tennis, racquetball, and
bicycling. He and Arnold also love to travel. Says
Bailey, "We went to Malaysia and Indonesia last sum-
mer and just wandered around for four weeks for ab-
solutely no professional reason whatsoever. It was
wonderful."
Frances Arnold sums up Jay Bailey's influence on
his profession in the following way, "Jay stands out in
the field as a pioneer in new techniques in the 8,000-
year-old discipline of biochemical engineering. You
won't find many new products coming out of his lab,
but you will find many new ideas." [


CHEMICAL ENGINEERING EDUCATION





















book reviews

CATALYST SUPPORTS AND
SUPPORTED CATALYSTS
by A. B. Stiles
Published by Butterworths,
80 Montvale Ave., Stoneham, MA 02180;
270 pages, $54.95 (1987)
Reviewed by
John B. Butt
Northwestern University
The title of this book is interesting enough since
most of those who deal in catalysts, particularly of
supported metals, often have an uneasy feeling that
the "support" dispersivee phase, contact phase, car-
rier, holder-i.e., any number of names) has never
been given enough attention. This book is a good start
in trying to rectify this situation, and Dr. Stiles has
collected a good group of reviews concerned with this.
I particularly enjoyed the first five chapters, devoted
to alumina, oxide supports other than alumina, acti-
vated carbon, and the associated information on their
preparation and properties. The surface chemistry in-
volved in catalyst-support interaction is probably not
as extensively dealt with here as in other sources, but
the overall treatment considered together with prepa-
ration techniques is quite a satisfying and useful one.
Chapter 7, on organic polymers, also falls into this
category.
The remainder of the book sort of strays from the
announced title. Khoobiar has done a good job in
Chapter 9 of "Spillover," and while significant oppos-
ing points are ignored, this is still a good review. Less
satisfying are chapters on the "Commercial Applica-
tion of Molecular Sieve Catalysts" and "Multifunction-
ing Catalysts." This is all old stuff, it seems rather
qualitative, and it strays far from tne announced title
of the volume.
The book is not very well proofread, as illustrated
on page fourteen as well as many other places in the
text. This makes me wonder how good the numbers


in the many tables and illustrations are. The publisher
should be more careful.
In spite of these reservations, I would say that
this is a book worth having. Get it, and learn about
supports. O
MASS TRANSFER WITH CHEMICAL
REACTION IN MULTIPHASE SYSTEMS
Vol I: Two-Phase Systems (679 pages)
Vol II: Three-Phase Systems (399 pages)
Edited by E. Alper; Martinus Nijhoff Publishers,
The Hague, Netherlands, 1983. $140
Reviewed by
Arvind Varma
University of Notre Dame
This two-volume book constitutes the proceedings
of a NATO Advanced Study Institute held in Turkey
in 1981. It includes thirty papers, primarily of a re-
view type, by twelve invited lecturers, and nine other
contributions. Various topics in the area of mass
transfer with chemical reaction in gas-liquid, liquid-
liquid, and gas-liquid-solid systems are covered. These
topics arise in the context of either separation pro-
cesses or reaction engineering. Some of the papers
treat the general problem of multiphase contacting
and reactor design. Others deal with the modeling of
specific types of contractors or reactors, and include
methods for obtaining or estimating physicochemical
and other data. Finally, some papers deal with a spe-
cific application, e.g., facilitated transport, bioreac-
tors, or reactors for coal conversion technology.
The invited lecturers are experts in the area
(mostly from Europe) who have written other reviews
as well. The material is somewhat dated by now, and
other more recent reviews and books have appeared
in print. Nevertheless, these volumes constitute a rich
source of information for this relatively narrow but
important area, and they should prove quite useful to
those involved with multiphase chemically reacting
systems.
The volumes were printed from camera-ready
copy. For this type of production, the cost of the book
is high. O


SPRING 1988


REQUEST FOR FALL ISSUE PAPERS

Each year Chemical Engineering Education publishes a special fall issue devoted to grad-
uate education. This issue consists of 1) articles on graduate courses and research, written by pro-
fessors at various universities, and 2) ads placed by chemical engineering departments describ-
ing their graduate programs. Anyone interested in contributing to the editorial content of the 1988
fall issue should write to the editor, indicating the subject of the contribution and the tentative date
it can be submitted. Deadline is June 1st.










Classroom


LEVELS OF SIMPLIFICATION

THE USE OF ASSUMPTIONS, RESTRICTIONS, AND CONSTRAINTS

IN ENGINEERING ANALYSIS


STEPHEN WHITAKER
University of California
Davis, CA 95616

The Navier-Stokes equations

p [ t+ V-VV = Vp + pg + V2 (1)

are exceedingly difficult to solve in their general form.
Thus there is great motivation to search for plausible
simplifications. One of these simplifications takes the
form: convective inertial effects are negligible. This
allows us to extract the linear version of Eq. (1) which
is given by

p- = p + pg + iv (2)

One could express this simplification as an equation,
and there is some advantage in identifying it as a
Level I assumption and expressing the idea as


Level I:


pvYv = 0


This type of statement indicates precisely what is
being done in a mathematical sense, but it provides
no basis for the action. For an engineer, it is more
attractive to make a statement of the type: convective
inertial effects are small compared to viscous ef-
fects. This leads to a Level II restriction of the form
Level II: pv.7v << pV2V (4)
In writing inequalities of this type it is understood
that the comparison is being made between the abso,
lute values of the terms under consideration.
Equation (4) has very definite advantages over Eq.
(3) since a comment concerning the physics of the pro-
cess under consideration has been made. While Eq.
(4) tells the reader what must occur in order that Eq.
(2) be valid, it does not indicate, in terms of the pa-
rameters of the problem, when it will occur. In order
to determine this, one must be able to estimate the
magnitude of the terms in Eq. (4). The treatment here
Copyright ChE Division ASEE 1988


Steve Whitaker received his undergraduate degree in chemical
engineering from the University of California at Berkeley and his PhD
from the University of Delaware. He is the author of three books: In-
troduction to Fluid Mechanics, Elementory Heat Transfer Analysis, and
Fundamental Principles of Heat Transfer, and he is the co-editor (with
Alberto Cassano) of Concepts and Design of Chemical Reactors. His
research deals with problems of multiphase transport phenomena, and
he has taught at U.C. Davis, Northwestern University, and the Univer-
sity of Houston.

will be brief since the details are given elsewhere [1].
We begin by expressing the velocity in terms of its
magnitude and a unit tangent vector
v = vX (5)
so that the convective inertial terms take the form
vVv = vX)Vv (6)
Since X is a unit tangent vector to a streamline, we
have
vv = v (7)
where s is the arclength measured along a streamline.
The derivative in Eq. (7) is estimated as [2, Sec. 2.9]
dv Av (8)
ds Lp

in which Av represents the change in v that occurs
along a streamline over the inertial length L,. Use of
Eq. (8) in Eq. (7) leads to an estimate of the convective
inertial terms given by


CHEMICAL ENGINEERING EDUCATION










PV7v = 0(Pv (9)

It should be clear that a successful use of this estimate
requires a reasonably good knowledge of the flow
field. The viscous terms in Eq. (1) can be expressed as

v = 2 + av a2v (10)
ax2 ay2 3z2
and the associated order of magnitude is given by


AV

L
Y


Here Avix represents the change of v that occurs in
the x-direction over the distance Lx, and the meaning
of Avl, and AvIl is analogous for the y and z-directions.
We represent the largest of the three terms on the
right hand side of Eq. (11) as Av/L2 and our estimate
of the viscous terms takes the form

v2 = OA (12)

Here LP is referred to as the viscous length. For many
cases the value of Av in Eq. (12) is comparable to the
value in Eq. (9) and this allows us to substitute Eqs.
(9) and (12) into the inequality given by Eq. (4) in
order to obtain
pvL2
v << 1 (13)

Traditionally the Reynolds number is defined in terms
of a length that is comparable to L,. Thus we use
pvL
Re = -p (14)

so that Eq. (13) takes the form

Level III: Re (j<< 1 (15)

Obviously this Level III constraint has a great deal
more utility than the Level I assumption given by Eq.
(3) for it allows one to decide a priori whether the
analysis is applicable to a particular problem. When
simplifying the Navier-Stokes equations on the basis
of Eq. (15), one must remember Birkhoffs warning
concerning the plausible intuitive hypothesis that
"small causes produce small effects" [3].
While the route to Eq. (15) is straightforward, it
is important to keep in mind that it is a scalar con-
straint associated with the magnitude of vectors and
it must be used with care. In addition, it is crucial to
understand that Eq. (15) has nothing to do with di-


My thoughts concerning the various levels of
simplification began to develop several years ago,
and while the origin remains diffuse, I might place it
in the early stages of an undergraduate
heat transfer course.


mensional analysis, but is based entirely on the pro-
cess of estimating the derivatives of the velocity that
appear in Eq. (4). When the flow is turbulent, Eq. (1)
must be time-averaged and Eq. (15) then applies to
the time-averaged inertial and viscous terms. Es-
timating the Reynolds stress, pv'*Vv', is more difficult
than estimating p*.VV since a knowledge of the mag-
nitude of v' and the turbulent length scale is re-
quired.
Often it is difficult to develop Level III constraints
for complex problems (think about the "perfectly
mixed" stirred tank reactor), and one must settle for
the type of simplification indicated by Eq. (3) in order
to meet deadlines and complete required course mate-
rial. From my point of view, the clearly stated Level
I assumption is an acceptable simplification for it tells
you what is being done and it reminds you that Level
II restrictions and Level III constraints are waiting
to be found. In addition, it should remind you that the
analysis has an unspecified range of validity and that
experiments and further analysis are in order.

SCENE
My thoughts concerning the various levels of
simplification began to develop several years ago, and
while the origin remains diffuse, I might place it in
the early stages of an undergraduate heat transfer
course. Because the subject under consideration was
heat conduction, I began a lecture with v = 0 and
quickly discarded radiant energy transport to arrive
at


P aT-
pc -


(16)


Since the assigned chapter and homework problems
dealt with steady, one-dimensional heat conduction,
we quickly moved to the following boundary value
problem:

dxk[ ) (17)


B.C.I:
B.C.2:


T=To x =
T = T x=L


With the comment that "we can treat the thermal con-
ductivity as constant," I was on the verge of present-


SPRING 1988


AV|
v2 = 0 -x L
2
x










ing the classical result given by

T = + (Tr1- T (20)

However, there was a flaw in my development. The
title of the chapter under consideration indicated that
we were to study the subject of steady, one-dimen-
sional heat conduction, but it said nothing about the
thermal conductivity being constant. One of the sages
from the back row spotted the opening, and the trad-
itional train of events was disrupted by the observa-
tion that "nothing is truly constant." Delighted to find
that a portion of the back row was awake, I pursued
Eq. (17) a bit further to arrive at

0-- = d + I k (d 2 (21)
Cdx2J k dx)
Since nonlinearities can be eliminated with impunity
in an undergraduate class, I was willing to asume that
ak/aT was zero and move on to the desired result
given by Eq. (20). However, the back row was warm-
ing to the task, and one of its occupants persisted
with, "But nothing is really zero is it?" A reviewer of
this article suggested that I should have counter-
attacked with the Kirchhoff transformation [4, Sec.
2.16] so that Eqs. (17) through (19) could be expressed
as


Sd2U
0 = -


from the tradition. Students who are tempted to ques-
tion the existence of the perfectly mixed stirred tank
reactor are afraid that the instructor might plunge
into a discourse on viscous dissipation, the Kol-
mogoroff length scale, and Damk6hler numbers.* The
terms of the treaty between students and faculty have
been hammered out over the years, and by and large
they work reasonably well. For example, can you im-
agine the difficulties of a study of fluid statics if it
were preceded by the Level III constraints associated
with
av
P- << pg pV'VV << pg ~2V << pg (26)

It is better to have a chapter entitled "Fluid Statics"
so that the deck is cleared for an exploration of the
pressure fields and forces associated with


0 = Vp + pg


Still, the question was posed from the back row,
and it deserved an answer. Furthermore, it seemed
to me that Eqs. (22) through (25) were most certainly
not the answer, for the question was, in reality: How
can you justify the simplification of Eq. (21) to arrive


Sd2T
0 =
dx2


(22) Clearly the second term in Eq. (21) cannot be neg-
lected on the basis of


B.C. 1:
B.C. 2:


U=0 x= 0
U =U x= L


Here the transformed temperature is given by
=T
U= k(11) dn (25)
ko
ln=To
in which ko is the thermal conductivity at the temper-
ature To. This approach would have avoided making
the assumption that k was constant, but it would have
delayed our arrival at Eq. (20) and the physical insight
that can be gained from that result. While Eqs. (22)
through (25) can provide an "exact" solution to the
problem posed by Eqs. (17) through (19), we usually
seek "approximate" solutions and often the approxi-
mations that we make are forced on our students by
the title of the chapter and the name of the textbook.
In engineering education there is a conspiracy
among students and faculty to base their assumptions
on the title of the chapter currently under considera-
tion. It is a game that is played with well established
rules and most often both parties are loath to depart


1 3k dT) << d2T
kaTJ d J dx2


(29)


but surely conditions must exist for which the vari-
ation of the thermal conductivity is "small enough" so
that Eq. (21) could be replaced by Eq. (28). This raises
the question of "small relative to what?" and the fol-
lowing problem was devised to explore this question
and to help students understand what is meant by
quasi-steady.

SAMPLE PROBLEM
We consider the boundary value problem given by

pc f) = k (a- + (k W (30)
p? F5 Ti .3X2. F [T xJ FaX J


I.C T = Ti t = 0
B.C. 1: T = T. + (T T )g(t) x = 0


B.C. 2:


T = T x=L


*It is bad enough that the material would not be available in the
text, but what is worse is that it would not be covered on the final!


CHEMICAL ENGINEERING EDUCATION









along with a second integration leads to


g(t)= 0 t = 0
g(t) = 1 t = t*


(34a)
(34b)


The classic test piece in a study of separation of vari-
ables is associated with t* -> 0+ and (ak/aT) -> 0, but
in this case we should simply think of t* as some
characteristic time associated with the boundary con-
dition at x = 0. Everyone knows that if t* is "large
enough" and if the variation of k with T is "small
enough," the solution to this boundary value problem
will yield the linear temperature profiles associated
with Eq. (28). The Level I assumptions related to
these conditions are


Level Ia:

Level Ib:


(35a)

(35b)


(DT)= 0

(- =
[FaTJ


and one should be careful to identify the first of these
as the quasi-steady assumption.
Our objective at this point is to develop the Level
II and Level III restrictions and constraints that are
associated with Eqs. (35a,b). Thus we seek to deter-
mine what is "large enough" and what is "small
enough." If you have an idea that a satisfactory solu-
tion to Eqs. (30) through (34) might be given by*

T = T + (To T1)g(t) 1 ( (36)

the possibility can be explored by decomposing the
temperature into the result represented by Eq. (36)
and whatever else is left. One method of doing this is
to arrange Eq. (30) as


aT2 1 f aT2
TX2 a FtJ k J 3xj


(37)


and to form the indefinite integral in order to obtain

aaT = r lo -T 1 rk T2 d (38)

x = o

Use of the definition

o = T 1 ak) (T2 (39)
S- atat k[aTJ [a (39)

*This solution is obtained by using Eqs. (35a,b) to reduce Eq. (30)
to the form given by Eq. (28). When Eq. (28) is solved subject to
Eqs. (32) and (33), the solution given by Eq. (36) results.


T(x,t) = T =
x=0


(aT) + F
+ x=O j + 0 d
Tn=o =o


The boundary conditions given by Eqs. (32) and (33)
can be used to evaluate the two constants in Eq. (40)
and the general solution is given by

T(x,t) = T1 + (To T)g(t) 1 !]


+ 'nddn -) nd=gd
n=o 0 =o T-=0 =o


One should keep in mind that this is an exact rep-
resentation for the temperature, but it is only useful
when the integrals are negligible. The integrals in Eq.
(41) can be estimated as


=x =nr 2
1nddn = 0() -
1=o 0=o


ndtdn = O(Q) -
n=o 0 =o


(42a)



(42b)


and use of these estimates in Eq. (41) leads to

T(x,t) = T + (To T)g(t) 1 (3j


+L 1 ( 0() (43)

We are now in a position to state that the solution for
T(x,t) is qiven by

T(xt) = T + (T T)g(t) 1 [J1 (44)

provided that the following inequality is satisfied


S- L)g(t) 2
(TO T )g(t) >> --0(9)


This result allows us to replace the Level I assump-
tions given by Eqs. (35a,b) with the following Level
II restrictions


SPRING 1988


Here g(t) is a function such that










Level IIa: (To T)g(t) > 0 (lt (46a)

L .2 fak)_ 3T .T 2
Level IIb: (T T)g(t) >> 0 k- T J (46b)

Here we are beginning to see how "long" one must
wait before the solution becomes quasi-steady, and
how "small" the variation of the thermal conductivity
must be in order that the last term in Eq. (30) can be
discarded.
In order to proceed further, we must be willing to
estimate the derivatives that appear in Eqs. (46a,b),
and in this development we will be satisfied with the
rather crude estimates given by [2, Sec. 2.9].
T I o TL (T T )g(t) t*
aT = x=L -(To 1*
-- = 0 = 0 t t t
(47a)


aT I 0 =-T x= = 0 L (47b)
ax L L

This aspect of the problem could be considered more
carefully by introducing the thermal boundary layer
thickness; however, we are interested in knowing
under what circumstances Eq. (44) is valid and the
estimates given by Eqs. (47a,b) are consistent with
that objective. When Eqs. (47a,b) are used in Eqs.
(46a,b) we obtain the Level III constraints given by

Level liIa: >> 1 t t (48a)
L2
Level IIIb: ktaJ(To Tl)g(t) < 1 (48b)

The first of these clearly indicates that the process
will be quasi-steady when t* is large compared to
L2/a and an exact solution of the boundary value prob-
lem will indicate that this is a conservative constraint.
Since g(t) has an upperbound of one, Eq. (48b) can be
replaced by

Level IIIb: _k To T1) << 1 (48c)

While the results given by Eqs. (48a,b,c) are some-
thing that "everyone knows," not everyone knows
how to arrive at these constraints without solving the
full boundary value problem and exploring special
cases. In addition, the identification of various levels
of simplification is an important concept to bring to
the attention of students, for it allows us to move
quickly to certain simple engineering solutions while


reminding us of our obligation to be more thorough
when time permits or necessity demands. Following
up on our obligations is sometimes easy to do. For
example, in the typical heat transfer course transient
processes are always studied, and when exact solu-
tions are available it is easy to remind students of
prior constraints that were developed on the basis of
order of magnitude analysis. In the study of transient
heat conduction in a flat plate one finds that Eq. (48a)
can be replaced with at/L2 1, thus providing a clear
indication that the original estimation was overly se-
vere. To support the result given by Eq. (48c), a
homework problem associated with Eqs. (22) through
(25) does rather nicely. The process of following order
of magnitude estimates with exact solutions is an at-
tractive method of encouraging students to develop
their own assumptions, restrictions and constraints.
As they gain confidence in this process, chapter titles
become guidelines for the voyage rather than con-
straints for the next exam.
REFERENCES
1. Whitaker, S., 1982, "The laws of continuum physics for single-
phase, single-component systems," Chapter 1.1 in Handbook of
Multiphase Systems, edited by G. Hetsroni, Hemisphere Pub.
Co., New York.
2. Whitaker, S., 1983, Fundatmental Principles of Heat Trans-
fer, R. E. Krieger Pub. Co., Malabar, Florida.
3. Birkhoff, G., 1960, Hydrodynamics, A Study in Logic, Fact,
and Similitude, Princeton University Press, Princeton, NJ.
4. Carslaw, H. S., and J. C. Jaeger, 1959, Conduction of Heat in
Solids, Oxford Press, London. O

books received
Gaseous Detonations: Their Nature, Effects and Control, by
M.A. Nettleton. Chapman & Hall, Methuen, Inc., 29 West
35th St., New York, NY 10001; (1987) 255 pages, $72
Radiation and Combined Heat Transfer in Channels, by M.
Tamonis (edited by Zukauskas and Karni). Hemisphere Pub-
lishing Co., 79 Madison Ave., New York, NY 10016; (1987) 239
pages, $69.95
Particulate and Multiphase Processes: Vol. 1, General Partic-
ulate Phenomena; Vol. 2, Contamination Analysis and Con-
trol; Vol. 3, Colloidal and Interfacial Phenomena, by Ariman
and Veziroglu; Hemisphere Publishing Corp., 79 Madison
Ave., New York, NY 10016 (1987); 932; 760; 544 pages, $133;
$131;$131
Batch Process Automation: Theory and Practice, by Howard
P. Rosenof and Asish Ghosh. Van Nostrand Reinhold Co., 115
Fifth Ave., New York, NY 10003 (1987); 336 pages
Vapor Cloud Dispersion Models, by Steven R. Hanna and
Peter J. Drivas. AIChE, 345 East 47 St., New York, NY 10017
(1987); 177 pages, $40 members, $75 others
Handbook of Thermodynamic High Temperature Process
Data, by A.L. Suris. Hemisphere Publishing Co., 79 Madison
Ave., New York, NY 10016 (1987); 601 pages, $139.95


CHEMICAL ENGINEERING EDUCATION















ACKNOWLEDGMENTS


Departmental Sponsors

The following 156 departments contributed to the support of CEE in 1988 with 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
California Institute of Technology
California State Polytechnic University
California State University, Long Beach
University of California (Berkeley)
University of California (Davis).
University of California (Los Angeles)
University of California (Santa Barbara)
University of California at San Diego
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 Florida
Florida State University
Florida Institute of Technology
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 of Technology
Louisiana State University
Louisiana Technical University
University of Louisville
University of Lowell
University of Maine
Manhattan College
University of Maryland
University of Massachusetts
Massachusetts Institute of Technology
McGill University
McMaster University
McNeese State University
University of Michigan
Michigan State University
Michigan Technical University
University of Minnesota
University of Missouri (Columbia)
University of Missouri (Rolla)
Monash University
Montana State University
University of Nebraska
University of New Hampshire
New Jersey Institute of Tech.
University of New Mexico
New Mexico State University
University of New South Wales
City University of New York
Polytechnic Institute of New York
State University of N.Y. at Buffalo
North Carolina A&T State University
North Carolina State University
University of North Dakota
Northeastern University
Northwestern University
University of Notre Dame
Nova Scotia Technical College
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


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 University
University of Saskatchewan
University of South Alabama
University of South Carolina
South Dakota School of Mines
University of South Florida
University of Southern California
University of Southwestern Louisiana
Stanford University
Stevens Institute of Technology
University of Sydney
Syracuse University
Tennessee Technological University
University of Tennessee
Texas A&M University
University of Texas at Austin
Texas Technological University
University of Toledo
Tri-State University
Tufts University
Tulane University
University of Tulsa
Tuskegee Institute
University of Utah
Vanderbilt University
Villanova University
University of Virginia
Virginia Polytechnic Institute
Washington State University
University of Washington
Washington University
University of Waterloo
Wayne State University
West Virginia College of Grad Studies
West Virginia Institute Technology
West Virginia University
University of Western Ontario
Widener University
University of Windsor
University of Wisconsin (Madison)
Worcester Polytechnic Institute
University of Wyoming
Yale University
Youngstown State University


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




P I


Union Carbide



Union Carbide Corporation is one of the nation's major
industrial companies. We employ more than 50,000
people at 700 plants and facilities in 37 countries, and
market our products in about 130 countries.

Over the past few years, we have become a more
streamlined, simplified and focused organization,
structured to respond effectively to rapidly changing
economic and market conditions.


Streamlined
for Success



Our principal businesses include chemicals and plastics,
industrial gases, and carbon products.

Union Carbide is one of the world's leading producers
of ethylene oxide/glycol and polyethylene. We have
the widest range of solvents of any U.S. chemical
company and are the largest domestic supplier of
industrial gases. We're also the world's largest producer
of carbon and graphite electrodes for steel-making.


A.- 3


Union
Carbide
Corporation


Department of
University
Relations


39 Old
Ridgebury
Road


Danbury, CT
06817-0001








An Equal
Opportunity
Employer


?1




Full Text

PAGE 1

z 0 j::: < u ::, 0 w c., z a2 w w z <5 z w 0 u.. u z < u a2 w < u.. 0 z 0 en > c c., z a2 w w z c., z w _, < u J: u VOLUME XXII NUMBER 2 James E. Bailey of Caltech The Mystique of Entropy KYLE Flow Sheet is Process Language FEHR A Course on Presenting Technical Talks FELDER Discrete-Event Simulation in Chemical Engineering SCHULTHEISZ, SOMMERFELD Safety and Loss Prevention in the Undergraduate Curriculum CROWL, LOUVAR A Pilot-Scale Heat Recovery System for Computer Process Control CALLAGHAN, LEE, NEWELL Engineering Management: Minimizing Functional Handicaps of Graduates ELLINGTON SPRING 1988 Levels of Simplification: Using Assumptions, Restrictions, and Constraints in Engineering Analysis WHITAKER and ChE at the /LLINOIS INSTITUTE OF TECHNOLOGY

PAGE 2

We wish to acknowledge and thank ... 3M FOUNDATION .. for supporting CHEMICAL ENGINEERING EDUCATION with a donation of funds.

PAGE 3

EDITORIAL AND BUSINESS ADDRESS Department of Chemical Engineering University of Florida Gainesville, Florida 32611 Editor: Ray Fahien (904) 392 -0857 Consulting Editor: Mack Tyner Managing Editor: Carole C. Yocum (904) 392-0861 Publications Board and Regional Advertising Representatives: Chairman: Gary Poehlein Georgia Institute of Technology Past Chairmen: Klaus D. Timmerhaus University of Colorado Lee C. Eagleton Pennsylvania State University Members SOUTH: R ichard Felder North Carolina State University Jack R Hopper Lamar University Donald R. Paul University of Texas Ja mes Fair University of Texas CENTRAL: J S. Dranoff Northwestern University WEST: Fr ederick H Shair California Institute of Technology Alexis T B ell University of California, Berkeley NORTHEAST : Angelo J. Perna New Jersey Institute of Technology Stuart W Churchill University of Pennsylvania Raymond Baddour M.I.T. NORTHWEST: Charles Sleicher University of Washington CANADA: Lesl ie W. Shemilt McMaster University LIBRARY REPRESENTATIVE Thomas W. Weber State University of New York SPRING 1988 Chemical ~ngineering Education VOLUME XXII NUMBER 2 SPRING 1988 Educator 58 James E. Bailey of Caltech, Robert Finn Department 62 Illinois Institute of Technology, Satish J Parulekar Laboratory 68 A Pilot-Scale Heat Recovery System for Computer Process Control Teaching and Research, P. J. Callaghan, P. L. Lee R. B. Newell 74 Safety and Loss Prevention in the Under graduate Curriculum: A Dual Perspective, Daniel A. Crowl, Joseph F Louvar 80 Engineering Management: A Course to Minimize Functional Handicaps of Graduates, Rex T. Ellington 84 A Course on Presenting Technical Talks, Richard M. Felder Classroom 88 Flow Sheet is Process Language, Manfred Fehr 98 Discrete-Event Simulation in Chemical Engineering, Daniel J. Schultheisz, Jude T. Sommerfeld 104 Levels of Simplification: The Use of Assumptions, Restrictions, and Constraints in Engineering Analysis, Stephen Whitaker Lecture 92 The Mystique of Entropy, B. G. Kyle 71 Letter to the Editor 72 Memoriam: James J. Christensen 73 Memoriam: Roland Andrew Ragatz 79, 86, 90, 91, 103 Book Reviews 108 Books Received C HEMICAL ENGINEERING EDUCATION (ISSN 0009-2479) is published quarterly b y Chemical Engineering Division, Am e rican Society for Engine e ring Education and is edite d at the University of Florida. Correspondence regarding editorial matter circulat ion and changes of address s hould be sent to GEE, Chemical Engineering Department, University of Florida, Gainesville, FL 32611. Advertising mate rial may be sent directly to E 0. Painter Printing Co., P 0. Box 877, DeLeon Springs, FL 32028. Copyright @ 1988 bf the Chemical Engineerin~ Divi s ion, American Society for Engineering Education. The statements and opiruons expressed in this penodical are those of the writers and not necessarily those of the ChE Division, ASEE, which body assu~es n o r~sponsibility for. them. Defective copies replaced if notif!ed ~th 120 days of publication. Wnte for information on s ubscnpti o n costa and for back copy cost and availa bility POSTMASTER: Send address changes to GEE, C hemical Engineering Department, University of Florida, Gainesville, FL 32611. 57

PAGE 4

ti Na educator I JAMES E. BAILEY of Caltech ROBERT FINN California Institute of Technology Pasadena, CA 91125 H IS COLLEAGUES describe James E. (Jay) Bailey as a man open to new ideas. Trained as a classi cal chemical engineer (Bailey did his doctoral work on chemical reactor theory, optimization theory, and non linear mathematics) he soon developed an interest in the relatively young field of biochemical engineering. Lately, Jay has turned even more biological in his ap proach, pioneering a brand new field he calls metabolic engineering. His graduate students, who have joined in this odyssey, are well prepared to combine the tech niques of molecular biology with more traditional en gineering methods in the chemical engineering prob lem-solving armamentarium. Jay ascribes many of the twists and turns of this intellectual odyssey to fortunate encounters with bright students and interesting colleagues. One of these chance encounters took place at the University of Houston in 1972 when Jay, then on the UH faculty, happened to meet biochemical engineer David Ollis. (Ollis was at Princeton University at that time; he is now Distinguished Professor of Chemical Engineering at North Carolina State University.) As Jay recalls it, "I was really impressed by the breadth of Dave s knowledge in biochemical engineering. I didn't know what a protein was in those days, but after a 30 -min ute conversation and a couple of phone calls, we de cided to write a textbook." The result was Biochemical Engin eeri ng Funda mentals, which has by now become the required text in almost every biochemical engineering course in the United States and in much of the rest of the world. It is now in its seco nd edition. "Writing the first edition of this text was a tremendous learning experience," Jay recalls. "I didn't know much biochemistry, micro biology, or biochemical engineering at that time. I read texts and review articles, and I went to Elmer Gaden's short course. All this convinced me that we had to emphasize the background material in biochemistry, physiology, and genetics that diffen : mCopyright ChE D ivision ASEE 1988 58 tiates biochemical engineering from chemical en gineering. A lot of other texts assume that you know what a protein is, what a bacterium is, what a gene is. We decided to define those terms, and others very explicitly. I think that my entry into the field as a beginner made the text suitable for other beginners. The second edition prepared after I had been heavily involved in biochemical engineering research for more than a decade, provided an opportunity to inject some personal perspectives into the text, in addition to im proving its organization and updating its content." Jay's own education as a chemical engineer was typical enough. He did both his undergraduate and graduate work at Rice University, earning his doctor ate working on reactor theory and optimization with Fritz Horn. After a two-year stint with Shell Develop ment in Emeryville, California, Jay established his own laboratory in the Department of Chemical En gineering at the University of Houston. It was there that he first let himself be nudged in the direction of biochemical engineering. "In the late 1960s and the early 1970s, the National Science Foundation sponsored a major effort in enCHEMICAL ENGINEERING EDUCATION

PAGE 5

Trained as a classical chemical engineer, [Jay] soon developed an interest in the relatively young field of biochemical engineering. Lately, he has turned even more biological in his approach, pioneering a brand new field he calls metabolic engineering. His graduate students, who have joined in this odyssey .. combine the techniques of molecular biology with more traditional engineering methods .. zyme engineering," Jay notes. I was doing computa tional and theoretical work on immobilized enzyme catalysis and on some diffusion-reaction problems with unusual properties. Two of my graduate students at UH had interesting backgrounds-Mike Cho had a background in biochemistry, and Jila Fazel-Madjlessi had a background in microbial fermentation. They helped me start experimental programs in those areas." With Cho, for example, Bailey made major theoretical and practical advances in the area of im mobilized enzymes. Their work involved the first ap plication of the carbodiimide-mediated covalent at tachment of enzymes to activated carbon. Using this approach they constructed hybrid catalysts that com bined an enzyme's activity with the catalytic action of the support surface. Specifically, they looked at the oxidation of glucose by the enzyme glucose oxidase. A byproduct of this reaction is hydrogen peroxide which tends to deactivate the enzyme. Bailey and Cho dis covered that activated carbon decomposes the peroxide and also assists in the supply of oxygen to the enzyme sites within the porous catalyst. They also discovered however, that loading too much en zyme onto the support compromises its efficiency in deactivating peroxide, and they were able to charac terize the tradeoff between higher enzyme loading and peroxide decomposition in some detail. Jay's experimental work on immobilized enzymes continues to this day. One of his most significant dis coveries was accomplished in collaboration with Ph.D. student Doug Clark (now an assistant professor and NSF Presidential Young Investigator at the Univer sity of California, Berkeley). Clark and Bailey com bined classical activity studies with electron paramagnetic spin resonance measurements of spin labeled immobilized enzymes. Using this technique, they were able to establish direct evidence of active site modification caused by immobilization and to show the relationship between the degree of modification and the activity of the enzyme. Bailey's collaboration with Clark characterizes Jay's philosophy of graduate education. He firmly be lieves that graduate students should participate heav ily in their research projects and in formulating strategies for solving their problems He observes, "Graduate students are here to learn how to do re search-not just how to do experiments or calculaSPRING 1988 Jay en;oys weekly group seminars to discuss progress and problems in the lab. tions, but how to identify a particular problem within an important general area. I encourage them to con sider the method of accomplishing a solution and its feasibility as well as the ultimate impact that the so lution will have. Besides, since I've been fortunate to work with many excellent students it would be foolish of me to presume that I have all the necessary ideas and knowledge. I depend on my students to contribute substantially to all aspects of their projects. On many occasions, including Doug Clark s work, my students have transformed my vague suggestions into highly original and productive research. Jay's work on periodic processes and biochemical engineering during nine years on the faculty of the University of Houston was recognized in 1979 by the Allan P. Colburn Award of the American Institute of Chemical Engineers. He came to Caltech in 1980. Ac cording to John Seinfeld a close friend of Bailey's and Caltech s Louis E. Nohl Professor and Professor of Chemical Engineering, "I first met Jay about twelve years ago at an AI ChE meeting and we became friends. Around 1979 I realized that Jay might con sider moving. He was quickly establishing himself as one of the leading people in biochemical engineering in the United States." Jay has been good for Caltech and the intimacy and informality of the Caltech environment have ap parently been very good for him as well. "The biggest qualitative change in my work came after I arrived at Caltech. Here my students and I can learn from and 59

PAGE 6

interact with some of the top people in biochemistry, molecular biology, and cell biology, not to mention chemical engineering. Cross-disciplinary collaboration happens almost automatically at Caltech." It was at Caltech that the techniques of genetic engineering began influencing Jay's work. The revolu tion in recombinant DNA technology that began in the late 1970s has been widely seen as the biologist's show. But Bailey realized that if these techniques were to realize their commercial potential, chemical engineers would have to get intimately involved. He also realized that modeling concepts from trad itional chemical reaction engineering could be effec tively applied in a new contextr-viewing a single cell A sophisticated flow cytometer-ce/1 sorter, here operated by Dane Wittrup and Elaine Mei/hoc, has supported many PhD pro;ects in Bailey's laboratory. as a complex chemical reactor and transport ap paratus. Working in Jay's group, Sun Bok Lee (now at the Korea Advanced Institute for Science and Technology) and Steve Peretti (currently an assistant professor at North Carolina State University) formu lated powerful new mathematical models that included genetic level regulation. Another consequence of Jay's interest in genetic engineering was his collaboration with Caltech's Judith Campbell, associate professor of chemistry and biology, who studies basic mechanisms of DNA repli cation in yeast. Genetic engineers often use the tech nique of inserting recombinant "plasmids" into yeast cells to get them to produce desired proteins in quan tity. Plasmids are relatively small, circular segments of DNA that contain regulatory genes and genes that code for proteins. There is one important problem with foreign plasmids, however: As the recombinant yeast cells divide, they tend to lose their plasmids generation by generation, gradually becoming less and less productive. Bailey and Campbell have made a concerted effort to understand this process using flow cytometry mea surements to provide unprecedented detail on single60 cell plasmid content in recombinant populations. In flow cytometry individual cells are sent past a laser beam, one by one, but very rapidly. In a process that takes about one millisecond per cell, flow cytometry permits determinations of cell size, shape, internal structure, and the concentration of specific compounds in individual cells. Many graduate students and post doctoral fellows have contributed to a growing body of flow cytometry research in Jay's lab, and several of these collaborators, including Friedrich Srienc at Min nesota and Jin-Ho Seo at Purdue, are now extending these methods in their research labs. Some of Jay's work in progress involves sharpen ing the focus of the genetic engineering studies to more subtle and interesting problems. For example, he's gotten interested in the details of what exactly happens to a protein after synthesis. There are essen tially three possibilities: The protein can be degraded, it can be exported from the cell, or it can remain in the cell in solution or in an aggregate. If genetic en gineering is to fulfil its economic promise, these alter natives must be understood and the factors that con trol protein targeting must be better characterized in quantitative terms. Typically, genetic engineers concern themselves with attempts to make cells produce certain proteins in quantity. But chemical engineers are well aware of the fact that proteins are not the only compounds of economic importance made by cells. Realizing this, Jay's group is concentrating on a brand new field that Jay calls metabolic engineering. "Metabolic engineer ing involves the application of genetic engineering techniques to alter or enhance the metabolism of cells," Jay explains. "The goal is to overproduce spe cific amino acids, acetone, vitamins, certain biopoly mers, or other important chemicals made in cells. Metabolic engineering requires a lot better under standing of what a cell is doing. "First, we must understand all the pathways avail able to an organism for producing the target com pound and determine what additional pathways we could make possible by introducing new enzyme ac tivities into that organism. Then, it's important to judge which regulatory points along the pathway are most important, since these determine the rate and the selectivity for the target compound. After deter mining the most promising target enzymes, we must clone the gene for that enzyme, attach the gene to a suitable vector, and insert it into the organism. Then, of course, we must investigate the effects of our gene tic manipulations on the production of the target com pound and its intermediates." Jay's group has made significant advances in all CHEMICAL ENGINEERING EDUCATION

PAGE 7

steps of this process. In terms of understanding metabolic pathways in detail, Alex Seressiotis, a former student of his who's now at Columbia Univer sity, developed a computer program called MPS (Metabolic Pathway Synthesis). The program contains a database system that stores enzyme and substance descriptions. "MPS can be used on a qualitative basis to examine the effects of adding or deleting enzyme activities to or from the cellular environment, to clas sify pathways with respect to cellular objectives, and to extract information about metabolic regulation," Jay says. In an illustration of the power of MPS, Seressiotis and Bailey had it consider the conversion of pyruvate to the amino acid L-alanine. MPS came up with a route to L-alanine that does not incorporate the enzyme alanine aminotransferase, which is com monly assumed to be a required step for alanine biosynthesis. Jay's lab has the distinction of being one of the first chemical engineering labs in the world with full cloning facilities. "The traditional way of producing strains of an organism with desired properties was to use a mutagenesis program," notes Bailey. This in volves exposing the cells to radiation or other muta gens and assaying the resulting organisms for the property in question. "But this is a very random, time consuming, and sloppy technique. By using recombin ant DNA technology we can make the whole process far more rational. Furthermore, students with cloning experience are more equipped to consider genetic as well as process solutions to engineering problems. And they'll be far better prepared to interact produc tively with molecular biologists in the future." To investigate the effects of genetic manipulations made in his lab, J&y's group has recently developed novel data analysis methods for use with his new, 300 megahertz, wide-bore NMR. These methods allow im portant measurements to be accomplished simultane ously from a group of living cells. In a single experi ment, Jay's students can estimate intracellular pH, the concentration of several key sugar phosphate metabolic intermediates, and the concentrations of adenosine diand triphosphate (ADP and ATP). At present, Bailey's lab is working on three pro jects in metabolic engineering. In the first of these projects, the lab is attempting to enhance the uptake of the sugar hexose in the yeast Saccharomyces cere visiae Hexose uptake is the rate-limiting step for growth in this economically important organism, which is used for ethanol production and in the genet ically engineered production of other compounds. The second project involves engineering the bacterium Es cherichia coli to produce ethanol by inserting genes SPRING 1988 ----------------Joy explored the crater ol Mt. Bromo in Java, Indonesia, in April 1987. for pyruvate decarboxylase and alcohol dehydro genase. In a third project, the group is trying to im prove production of ATP (the cell's energy currency) in E. coli. Not all chemical engineers agree with the direction Bailey's research has taken. According to John Sein feld, there are still a few diehards out there who ex press skepticism about chemical engineers who get involved with essentially biological techniques. But, Seinfeld says, there are fewer and fewer naysayers. "I think biochemical engineering is going to mature. People are recognizing that it will take its place as a firm field in chemical engineering and, in a sense, set tle down. It is difficult for biochemical engineering to be as rigorous as some other fields of chemical en gineering that are closer to physics. That level of rigor is being supplied by people like Jay Bailey. There will be more biology in the education of future chemical engineers as a result of his work." The education of chemical engineers is a subject that's important to Bailey. He's particularly proud of the fact that many of his former graduate students have become faculty members at important univer sities around the world. Frances Arnold, assistant professor of chemical engineering at Caltech and Bailey's wife says, "Jay takes a personal interest in each of his students. He's always willing to sit down with them and tell them the 'facts of life' of the chem ical engineering profession. His students show a lot of loyalty as a result of his interest in them." Important institutions to Jay's group of students and postdocs are Friday afternoon group seminars and, following that, the traditional "Ho-Ho" in the Rathskeller of the Athenaeum, Caltech's faculty club. "Gathering outside the lab for conversation and a few Continued on page 102 61

PAGE 8

department Chemical engineering offices and laboratories ate located in Perlstein Hall. ILLINOIS INSTITUTE OF TECHNOLOGY SATISH J. PARULEKAR Illinois Institute of Technology Chicago, IL 60616 T HE DEPARTMENT OF Chemical Engineering at the Illinois Institute of Technology (IIT), formally established in 1904, is one of the oldest chemical en gineering departments in the nation. The develop ment of the department in the pre-1971 period was previously described in this journal [1]. Dramatic changes have occurred in the department since that time in terms of faculty, undergraduate and graduate programs, and research. All current chemical en gineering faculty, with the exception of one member, were added after 1974. Since then, there has been a significant expansion in research activities and re search areas pursued, with the current faculty having expertise in traditional as well as emerging areas in Copyrigh t C kE Division AS E E 1988 62 chemical engineering. This is also reflected in the un dergraduate and graduate curricula, which have changed significantly to meet the changing needs of the prof e ssion. IIT's main campus, internationally recognized as an architectural landmark and located three miles south of Chicago' s downtown area and a mile west of Lake Michigan, was designed in the 1940s by Ludwig Mies van der Rohe, one of this century s most influen tial architects. The current university population on the main campus includes about 6500 undergraduate and graduate students and about 500 faculty. The other campuses of IIT are located in downtown Chicago and in the western suburbs of Chicago in DuPage County. Chicago, one of the largest cities in the world and a national and int e rnational center of busine s s and indu s try offers students an exception ally wide variet y of professional and cultural re source s The close proximity of several chemical, food, and allied industry research centers and production C HEMI C AL ENGINEERING EDUCATION

PAGE 9

Since engineering is largely a team effort, we believe that development of the individual's ability to work effectively as part of a team is important. To accomplish this, the laboratory courses and the design courses involve teams of students ... laboratory sections are small and a high level of personal contact between students and instructor is maintained ... plants provides for the UT community an invaluable source of close interaction with industry. THE DEPARTMENT The chemical engineering offices and laboratories are located in Perlstein Hall. Presently, about 150 full time undergraduate students and 60 full-time graduate students are enrolled in the department. In addition, there are about 40 part-time undergraduate students and 50 part-time graduate students. Three graduate degrees are offered by the Chemical En gineering Department: Master of Science (MS), Mas ter of Chemical Engineering (MChE), and Doctor of Philosophy (PhD). The department annually awards nearly 40 Bachelor of Science, 10 MS, 10 MChE, and 8 PhD degrees. Over the years, the department has graduated many competent engineers, a number of whom have obtained significant national prominence in their pro fessional careers. Several of our alumni have held na tional office in the American Institute of Chemical En gineers. Although a majority of our alumni have pur sued professional careers in industry, we are proud that, over the years, a significant number have joined chemical engineering faculties at major institutions. RESEARCH The research areas pursued in the pre-1971 period were chemical reaction engineering, separation pro cesses, thermodynamics, and transport and interfacial phenomena [1]. In recent years, many additional areas of research have been initiated and strengthened. They include biochemical engineering, biomedical en gineering, combustion, energy technology, enhanced gas and oil recovery, and process dynamics and con trol. Interdisciplinary research has been a strong tradition at UT and several joint initiatives with other departments have evolved to address problems in the critical emerging areas. Some of the faculty have cooperative research activities with the Institute of Gas Technology, UT Research Institute, and Argonne National Laboratory. All such interactions provide an intellectually stimulating environment for the graduate students. The development of UT's very strong research and educational center in food technology at the Moffett Technical Center (located in the Chicago area) will be completed this summer. SPRING 1988 Some of the chemical engineering faculty will be closely associated with this center. It is interesting to look at the current faculty in chronological order of joining the chemical engineer ing department at UT in order to trace the growth of our program in recent years. Darsh Wasan, currently acting Dean of the Ar mour College of Engineering, is the oldest faculty member in terms of service, having joined UT in 1964. He served as department chairman from 1971 to 1987. His research deals with interfacial and colloidal phe nomena, enhanced oil recovery, and separation pro cesses. Darsh and his research team are developing experimental techniques to accurately measure sur face and interfacial rheological properties. This re search has significant utility in control of a variety of dispersed phase systems and in study of the mecha nisms of oil bank formation and propagation in chem ical flooding processes involving surfactants and al kaline agents, foams, emulsions, and polymers for mo bility control. Research in separations is concentrated on separation of fine particles from non-aqueous media based on electrokinetic phenomena, emulsification/de mulsification processes and thin liquid film phenome na. Darsh is a recipient of the Western Electric Fund Award of the ASEE, the Hausner Award of the Fine Particle Society, and the Special Creativity Award of the National Science Foundation. The mid-1970s saw the addition of two of the cur rent faculty members, Rob Selman and Dimitri Gidas pow. Rob Selman, currently acting chairman of the department, joined the faculty in 1975. His research interests are electrochemistry and electrochemical en gineering with a special emphasis on high-tempera ture processes. Rob and his students are currently investigating the formation, growth, and corrosion of dendritic metal deposits, a problem common in the charging of zinc-based aqueous batteries which are being developed for electric vehicles and load leveling. Other current research projects deal with fundamen tal aspects of molten carbonate fuel cells and develop ment of porous-electrode models for these fuel cells. The characterization of micro-emulsions by AC impe dance is being investigated to assess the effectiveness of electrochemical oxidation as a means of breaking micro-emulsions. Other research projects under Rob's direction deal with molten salt processes involving 63

PAGE 10

carbon cathodes and amorphous metal deposition from molten salts. Dimitri Gidaspow joined the faculty in 1977. Dimitri's research interests are in the areas of mathematical modeling and analysis of various energy conversion processes. His current research involves studies of the hydrodynamic theories of fluidization and gas-solids transport. Generalizations of N avier Stokes' equations are being solved using supercom puters to predict phenomena such as cluster forma tions in risers and bubble motion in fluidized beds. In collaboration with Darsh Wasan, he also conducts re search in the areas of separation of colloidal particle s and electrostatic desulfurization of coal. Dimitri has been recognized for his research through the Donald Q. Kern Award of the AIChE and the Special Creativ ity Award of the National Science Foundation. The bulk of the present faculty members arrived on the scene in the 1980s. Richard Beissinger joined the chemical engineering department in 1981. Taha Alkhamis and Richard Beissinger worlcing with a Weissenberg rheogoniometer for blood viscosity mea surement. Richard's research is concerned with transport phe nomena in biological systems. His current research activities are in blood-artificial surface interactions biorheology, and development of artificial red blood cells. He also conducts research in the development of implantable drug infusion systems and in phar macokinetics. Some of his current research projects are: augmented mass transport in sheared suspen sions; macromolecular adsorption to solid surfaces; liposome-encapsulated hemoglobin and hemoglobin-in oil-in-water multiple emulsion droplets as red blood substitutes; and the effects of red blood cells on platelet adhesion and aggregation in laminar shear 64 Jill Weldon and Selim Senkan discussing experiments with a I/at flame burner in the combustion research lab oratory. flow. Rich is also directing development of the inter disciplinary Polymer Science and Engineering pro gram. Ali Cinar, who arrived in 1982, conducts research in process control and dynamics. Control strategies for multivariable chemical processes are being de veloped and tested using experimental systems and real-time microcomputers. One of his research pro jects deals with developing methods for selecting robust operating configurations and improved control strategies using a pilot plant consisting of two cataly tic tubular reactor beds with internal heat exchange and a feed-effluent heat exchanger. Ali and some of his graduate students are also investigating forced periodic operation of chemical reactors using the vi brational control approach. Their work is focused on stabilization of exothermic continuous stirred tank reactors and tubular packed bed reactors as well as on improvement in selectivity and yield of complex reactions. Another of Ali s research projects deals with development of expert systems for fault-tolerant computer control of complex processes, such as multi bed autothermal reactors. Selim Senkan joined IIT in 1982. His research is in the areas of combustion and high-temperature chemical reaction engineering as applied to problems in energy conversion, propulsion, and environmental protection. One of his research projects deals with ex perimental and theoretical investigation of oxidation and pyrolysis of chlorinated hydrocarbons (CHCs) using model compounds. In the experimental pro gram, stationary flames of selected CHCs are probed for the determination of species and species profiles CHEMI C AL ENGINEERING EDUCATION

PAGE 11

using supersonic jet sampling coupled with on-line molecular beam mass spectroscopy (MBMS). An out come of this research is the recently patented, economic process for conversion of methane into acetylene and ethylene. Because of low NOx pollutant emissions, catalytic combustion is an attractive alter native to flame combustion when very high tempera tures are not required. Selim and his group are also involved in the study of the impact of halogens on catalyst development, reaction rates, and selectivity in catalytic combustion processes. William Weigand joined the IIT faculty in 1983. His research interests are biochemical engineering and process control. Bill and his research group are involved in kinetic modeling of cell growth and prod uct formation for microorganisms which produce pri mary and secondary metabolites. The optimal operat ing procedures are then derived and examined with the aid of computer simulations and experiments car ried out with highly-instrumented computer-coupled fermentors. Bill is also interested in development of new sensors and the use of estimation techniques which permit optimal operation in the absence of com plete variable measurement. In the area of process control, techniques for control in the presence of mod eling error, for changing system dynamics, and for nonlinear interacting systems are being developed. Hamid Arastoopour and Satish Parulekar joined the department in the fall of 1985. Hamid Aras toopour has research interests in the areas of multi phase flow, flow in porous media and unconventional gas reserves, and fossil fuel conversion processes. Current research projects being investigated by Hamid and his graduate students are hydrodynamic analysis of pneumatic conveying of solids, numerical analysis of single and multiphase flow in unconven tional reserves, analysis and measurement of the agglomeration of sticky particles and heat transfer in fluidized beds, and analysis and measurement of tran sient gas and condensate flow in gas transmission and distribution systems. Satish Parulekar conducts research in biochemi cal engineering and chemical reaction engineering. His research in biochemical engineering deals with production of extracellular and intracellular enzymes / proteins by recombinant and wild-type microor ganisms and production of biochemicals using im mobilized cell reactors. The research with recombin ant organisms is aimed at gaining a fundamental un derstanding of host-plasmid interactions in these species and the effect of these interactions on plasmid stability, cell growth and product gene expression. The research with wild-type microorganisms is diSPRING 1988 rected towards understanding the mechanism of syn thesis of extracellular enzymes such as amylases and proteases. The research dealing with immobilized cell reactors is focused on study of alteration in cellular metabolism due to immobilization and implications of such alteration in design of these reactors. His re search in chemical reaction engineering is concerned with identification of optimal reactor structures for Prasad Davuluri and Hamid Arastoopour studying tran sient gas transport with a Laser Doppler velocimeter. complex reaction networks and experimental and theoretical investigation of forced periodic operation of continuous flow ( CSTR and tubular) reactors. Henry Linden joined the faculty in the spring of 1987 as the F. W. Gunsaulus Distinguished Professor of Chemical Engineering. Henry has had an excep tionally illustrious career in research in fuel technol ogy at both the In s titute of Gas Technology and the Ga s Research Institute and was president of each of these institutions for several years. His distinguished career has led to membership in the National Academy of Engineering, as well as to numerous hon ors and award s such as the Distinguished Service Award from the American Gas Association. Having Henry on board will give us the benefit of his wide ranging experience and insight in energy-related re search. 65

PAGE 12

UNDERGRADUATE PROGRAM The mission of the undergraduate program in chemical engineering at IIT is to prepare our students for their professional careers and to enable them to develop the technology of the future. There is enough flexibility in the program for each student to tailor an individualized curriculum to satisfy his or her particu lar scientific or technical interests. The first two years are devoted to the fundamental sciences, mathema tics, and engineering sciences and are particularly concerned with the development of professional skills. In addition to developing engineering competence the program examines the economic and societal implica tions of chemical engineering. The required courses are material and energy balances, and unit opera tions-I in the sophomore year; thermodynamics, trans port phenomena, unit operations-II, and chemical en gineering laboratory-I in the junior year; and chemical reaction engineering, process control, chemical pro cess design (two semesters), and chemical engineering laboratory-II in the senior year. In addition, there are e lective courses in bioen gineering, computer applications in chemical en gineering, colloidal and interfacial phenomena, elec trochemistry energy technology, food technology microelectronics fabrication, and polymer processing. Students interested in gaining professional specializa tion may include some of these courses in their cur riculum to earn a specialized minor in one of the fol lowing areas: biotechnology computers in chemical engineering, energy technology, food technology, and polymer engineering. Since engineering is largely a team effort, we be lieve that development of the individual's ability to work effectively as part of a team is important. To accomplish this, the laboratory courses and the design courses involve teams of students. Since individual in struction is so important to students' growth, labora tory sections are small and a high level of personal contact between student and instructor is maintained. Some of our best students also choose to work on inde pendent research projects during their junior and / or senior years which prepare them for graduate re search. GRADUATE PROGRAM The coursework for MS and MChE degrees must include at least four of the following six core courses : chemical reaction engineering, fluid mechanics, heat transfer, mass transfer, process control, and ther modynamics. For a PhD degree, coursework in all six areas is required so that the students will be equipped 66 At IIT we are developing courses to integrate these emerging areas into our curricula while, at the same time, leaving the emphasis on science and engineering fundamentals unchanged. to apply advanced principles from the entire spectrum of chemical engineering irrespective of their research specialization. A student pursuing an MS must com plete eight credit hours in research and thesis work. The MChE is a professionally-oriented degree pro gram which permits a concentration in engineering practice. The requirements are the same as those for the MS degree, except that additional courses and/or a project replace the eight credit hours of thesis. Every prospective PhD candidate must take a qualify ing examination to determine fitness and aptitude for further graduate study. Research qualities are judged during the oral comprehensive examination over the student's dissertation proposal, taken some time after admission to candidacy. After successfully clearing this examination, the candidate pursues the selected research program in consultation with the research adviser and advisory committee. In addition to the core courses, IIT offers a wide variety of elective graduate courses which in the past three years have included biochemical engineering, catalysis, computational techniques, electrochemical engineering, polymer processing, reservoir engineer ing, separation processes and transport phenomena in living systems. Many of the graduate courses, particu larly the core courses, are televised through the IITN network. Remote centers for reception of these tele casts are located near several of the industrial centers within a fifty-mile radius from the IIT campus for the benefit of part-time students. THE FUTURE As the chemical engineering profession changes and adapts to the technology needs of the next decade and the next century, so must chemical engineering education. In the coming years, the curriculum con tent will be modified to accommodate new technology areas in chemical engineering. At IIT, we are develop ing courses to integrate these emerging areas into our curricula while, at the same time, leaving the em phasis on science and engineering fundamentals un changed. As in the past, thorough training of the un dergraduate as well as graduate students will continue to receive the highest priority. REFERENCES 1. Kintner R. C ., D. T. Wasan, Chem Eng. Ed., 5, 108 (1971). CHEMICAL ENGINEERING EDUCATION

PAGE 13

AIChE JOURNAL Morton M Denn Editor Published Monthly The AIChE JOURNAL, a monthly publication of the American Institute of Chemical Engineers, is devoted to fundamental research and developments having immediate or potential value in chemical engineering. Increased to 12 issues in 1985, this journal now provides more comprehensive coverage of ongoing research and developing technologies in chemical engineering. As a permanent repository of innovative processes, efficient analyses of data and current theoretical ideas, it serves as an excellent reference source for researchers. Original papers are reviewed by a board of peer scientists and engineers. Contributors come from industry, government and univerity research groups in the U.S. and throughout the world. Range is broad and includes such varied topics as: heat transfer, mass transfer, fluidization, kinetics, adsorption, process dynamics, membrane technology, separation processes, solid-liquid systems, reactor technology, electrofaction, polymer processing, filtration, crystallization. Coverage impacts upon chemical, environmental and biotechnological engineering. 1988 Annual Subscription Prices: AIChE Members $40 Others $275 Foreign Extra: $13 Subscriptions are on a calendar-year basis Send Orders to: AIChE Subscription Dept. A, 345 East 47 Street, New York NY 10017. Prepayment in U.S currency is required (check, VISA, MasterCard, international money order, or bank draft drawn on a New York bank). Members may order only one subscription at member price and must indicate membership number when ordering. Credit card customers: Please indicate "VISA" or "MasterCard," include card number, expiration date, printed name of cardholder and signature. AMERICAN INSTITUTE OF CHEMICAL ENGINEERS

PAGE 14

[eJ ;j a laboratory A PILOT-SCALE HEAT RECOVERY SYSTEM FOR COMPUTER PROCESS CONTROL TEACHING AND RESEARCH P. J. CALLAGHAN, P. L. LEE and R. B. NEWELL University of Queensland St. Lucia, Qld, Australia 4067 I N THE LAST two decades efficient energy utilization and process integration have become increasingly important to chemical plant economics. Plant design for these factors has meant higher levels of process interaction, often with positive feedback, which re quire improved control. A typical example is the heat recovery circuit in reactor ~r distillation systems. Process equipment used in the laboratory to inves tigate the control problems created by tighter process integration is more convenient. This can be sum marized as Lower cost of pilot-scale equipment and Loss of production and or product quality are not issues during process experimentation. However, instrumentation and control equipment is "off the shelf'' as is found in industry and hence is a significant cost factor. The aim in building process control rigs for undergraduate teaching and graduate FIGURE 1. Schematic diagram of pilot-scale heat recov ery circuit.
PAGE 15

PILOT-SCALE HEAT RECOVERY CIRCUIT A schematic diagram of the laboratory rig is shown in Figure 1. Mains water is preheated in a pre-heat plate heat exchanger, HEL After preheating it flows through a steam heated double pipe heat exchanger, HE2. After the double pipe exchanger the water splits either to be used in the preheat plate heat exchanger or to bypass the heat exchanger. When the streams rejoin they are fed to a constant head drain. A control valve, V3, is installed in the bypass line and may be used to regulate the bypass flowrate. Equipment de tails are given in Table 1. There are two manipulated variables in the sys tem: a control valve in the steam line, Vl, and a con trol valve, V2, in the hot line after the double pipe exchanger leading to the preheat plate exchanger. They can be used to control any of the dependent vari ables, but typically the final mixed temperature, TT3, P J. Callaghan received his BEChem and his MEngSci degrees from the University of Queensland in 1984 and 1986. His Master's programme centered an the development and use of a predictive con trol algorithm far the heat exchange system described in this paper. He is currently employed by Akoa Ltd. as a control engineer (Not pictured) P. L. Lee received his BEChem from RMIT in Melbourne and his PhD from Monash University in 1980 He worked in the design and commis sioning of computer control systems far both continuous and batch plants far three years before coming to Queensland. His early research was on the control of the unstable steady state in an exothermic CSTR. His interests include multivariable self-tuning and adaptive control of fermentation and heat recovery systems. (L} R. B. Newell received his BScApp and BEChem from Queensland and his PhD from the University of Alberta. He also has a DipEd in Tertiary Education from Monash He joined the staff at Monash Univer sity in 1974 and moved ta Queensland in 1980 His early research was in the multivariable control of a pilot plant evaporator, unstable steady state control in a CSTR, and multilevel hierarchical optimization Current interests include optimization of the Australian ail refinery and transportation system, combined fuzzy and deterministic control, and selftuning and adaptive control of heat recovery systems (R) SPRING 1988 and the temperature out of the double pipe exchanger, TT2, are used. A standard differential-pressure bypass control loop is usually implemented to maintain the pressure drop across the plate heat exchanger con stant. PROCESS-CONTROL SYSTEM INTERFACE An interface box between the process instrumen tation and the control system is located on the rig. This box displays all rig variables, both manipulated and controlled, i.e., the steam pressure, the hot side water flowrate to the plate heat exchanger, and the three temperature measurements and the valve posi tions. The interface box is supplied with 240V AC power and supplies 24 V DC for instruments and trans ducers. Standard signals are 0-5V for control signals and 1-5V for measurements. The interface box also has the facility to switch the manipulated variables from remote to local, by interrupting the signal from the control system and supplying a manual signal. Figure 2 shows the responses in the exit tempera72 71 70 .. 1 ~)1 :11 1 86 L .. . ._____i __ _._ __ 1_ _.__..1 e 2e oeo .. 87 .. 88 100 129 1 0 160 188 200 Sample (al Sam ple (bl FIGURE 2. Step test responses for TT2 due to a 10% step (a) up and (b) down in VI after sample 20. 69

PAGE 16

ture to 10 % step changes up and down in the steam. The sample time during the collection of this data was five seconds. Note that both magnitude and shape of responses differ, illustrating non-linear behaviour. Also the need for filtering is apparent from this raw data. CONTROL SYSTEM EQUIPMENT Basic Data Collection and Regulatory Control The control system used to monitor and control the pilot-scale apparatus is arranged as shown in Fig ure 3. This equipment is arranged in a hierarchical manner, consisting of two layers: a primary data col lection and regulatory control level, and an advanced control level. IB M Superv ACS Level iaory ~BAILEY NETWORK 90 Regulatory Level PROCESS FIGURE 3. Control system equipment. The primary data collection and regulatory control functions are performed using a Bailey Network 90 distributed control system. The heart of the system is a "multi-function controller," which is a micro-proces sor based controller. It performs all of the regulatory control functions and coordinates the operation of the analog-to-digital input and digital-to-analog output cards. Operator interface is provided by means of a color CRT display. This display is capable of showing schematic representations of the pilot-scale equipment with measurement data displayed, controller faceplate displays, trend plots of mea surement inputs, and com binations of the above types of displays. 70 Advanced Control Function The advanced control functions such as imple menting multivariable predictive controllers are per formed using the IBM Advanced Control System (ACS) previously described in this journal by Koppel and Sullivan [1]. However, this application uses ACS in a real-time environment. The University of Queensland installation consists of an IBM 4341-2 mainframe with 8 Mbytes of mem ory, 12 terminals and 2 printers in two clusters line printer, 1.8 Gbytes of disc storage, a tape drive and a Series / I minicomputer with direct process I/0 and connections to the Bailey Network 90 system UNDERGRADUATE COMPUTER-BASED PROCESS CONTROL INSTRUCTION The experimental rig has a number of interesting characteristics that make it particularly suitable for teaching real-time process control: Non-linear behaviour. This is shown in Figure 2, which illustrates the different time constants for a step increase and a step decrease in the steam supply v alve Vl. The time constants are approximately 96 and 45 seconds respec tively. Interacting behaviour. This is illustrated by e x amination of the relative gain array for this process shown in Table 2. The inherent noise in the measurements, again illustrated in Figure 2. The following teaching experiments utilize this rig: 1. Examination of digital filtering. Students are asked to examine the effects of different filter bandwidths and different filter types (exponential, Butterworth union). This short experiment is used in teaching real-time computing and would be ex pected to take one hour of laboratory time. 2. A cascade loop controls the double pipe exit temperature with a slave loop controlling the pre s sure in the steam jacket via the steam valve. This ex periment only utilizes the Bailey Network 90 system and students are expected to complete this experi ment in one and a half hours. 3. Feedforward control. A feedforward controller compensates for disturbances in the feed flowrate by adjusting both the steam and water valves. This ex periment, combined with a model identification exer cise usually takes about three hours of laboratory time. 4. Decoupling control. Two feedback loops or two interacting feedback loops are used as a basis for inCHEMICAL ENGINEERING EDUCATION

PAGE 17

traducing the concept of process interactions. The students are then expected to design and implement a decoupled controller using lead-lag compensators. This experiment normally takes six to nine hours to complete. 5. Adaptive control. An adaptive single loop con troller adjusts the steam valve to control the exit temperature. This experiment, used in a fourth-year elective control course, makes use of the ACS system and would normally involve fifteen hours of labora tory time. POSTGRADUATE PROCESS CONTROL RESEARCH Current research utilizing the rig is on a predictive control design technique described by Maurath, et al. [2]. The basic predictive control algorithm uses a model to predict the output for a number of future moves due to the previous inputs. An error is formed by subtracting the prediction of the output from the setpoint. This error vector is then used to calculate a change in the manipulated variables. The design tech nique being tested is to use a singular value decompo sition to condition the Dynamic Matrix [2] of step re sponses. Results from this study are reported else where [3]. Future work on the rig will include fuzzy identifi cation and control. A model-based controller designed around a fuzzy model of the process will be used to control the rig. Fuzzy identification techniques will be used off-line to derive an initial model, and on-line to provide continuous updating of the model. The on-line use of fuzzy identification turns the model-based con troller into a fuzzy adaptive controller [ 4, 5, 6). A reactor heat recovery system is a possible exten sion to the current rig. This is achievable by using an exponential function on the exit temperature of the double pipe exchanger to move the steam valve. This allows for simulation of variable heats of reaction without the danger and cost ofreactants. In this mode only one manipulated variable is possible, i.e. the water valve (V2). CONCLUSIONS AND SIGNIFICANCE The pilot-scale heat recovery system is a valuable teaching and research tool. It is sufficiently flexible to Vl V2 SPRING 1988 TABLE 2 Relative Gain Array TT2 0.22 0.78 TT3 0.78 0.22 demonstrate basic principles and yet sufficiently com plex to demonstrate common process control problems such as nonlinearities and interactions between vari ables. The process control system equipment allows stu dents to obtain experience in real time computing, at several process control functional levels. REFERENCES 1. Koppel, L. B. and G. R Sullivan "Use of IBM 's Advanced Control System in Undergraduate Process Control Education," Chem Eng. Educ XX, No. 2, pp. 70-7 3, 106-07 1986 2. Maurath, P.R., D. E. Seborg, and D. A Mellichamp, "Predic tive Controller Design by Principal Components Analysis," Proc. Acc 1985, pp. 1059-1065. 3. Callaghan, P J., M.Eng.Sc Dissertation, University of Queensland, 1986. 4. Graham, B. P. and R. B. Newell, "Computer Control of Dif ficult Processes," Chemeca 84, Vol 2, pp. 743-749 (1984). 5. Graham, B. P., P. L Lee and R. B. Newell, "Simulation of Muitivariable Control of a Cement Kiln ," Chem. Eng R es. Des., 63, Nov. 1985. 6. Graham, B. P., R. B. Newell, G. P. Le Page, and L. C. Stonehouse, "Industrial Applications of Fuzzy Identification and Control," Chemeca 86, pp. 221-226 (1986). D (ejij:,j letters UPDATED REFERENCES Dear Editor: In a recent paper, "A Course in Mass Transfer with Chemical Reaction," Chem. Eng. Ed., 21, No. 4, 164 (Fall 1987), W. J. Decoursey refers to Gas Liquid Reactions by P. V. Danck werts as a source in English for the seminal pa pers by S. Hatta which were published in Japanese [Technol. Repts. Tokoku University, 8, 1 (1929); 10, 613,630 (1931); and 11, 365, (1932)]. Complete translations of the latter articles have been published in International Chemical Engi neering, 18, 443-475 (1978). The readers of this article may also be interested to know that a translation of the closely related pioneering paper by G. Damkohler, "The Influence of Diffusion, Fluid Flow and Heat Transport on the Yield in Chemical Reactors," Der Chemie-Ingenieur, 3, Part I (1937), has recently been published in International Chemical Engineering, 28, 132-198 (1988). Stuart W. Churchill University of Pennsylvania 71

PAGE 18

In Memoriam James J. Christensen 1931-1987 Dr. James J. Christensen's life was devoted to ser vice: service to wife and family; service to his profes sion in chemical engineering and to the field of ther modynamics; service to his university, the Brigham Young University and to his students, both graduate and undergraduate; and service to his church and com munity. This life of service was brought to a close by a sudden heart attack on September 5, 1987. Jim's worldwide recognition as an outstanding thermodynamicist and a developer and builder of state-of-the-art calorimeters gave him and his wife a great opportunity to travel around the world. An out standing feature of the family room of their home is a large map with the continents shown in bas-relief cut from plywood. The map is studded with pins, indicat ing the many places they have gone where they had the privilege of meeting new people and of teaching his methods to many. Jim and Virginia shared each other's life in this way as well as in taking dance clas ses together and in participating in the great books program over the last 28 years. He was dedicated to his four sons and a daughter and spent many hours with them in such activities as hiking. One of his sons followed him into the chemical engineering profession. Although Jim loved his professional career, he loved service to his family equally well. Jim's service to the profession in the field of calorimetry and thermodynamics is unparalleled. He received a BS degree in 1953 and an MS degree in 1955 in chemical engineering from the University of Utah. He then went on to complete his formal educa tion with a doctor's degree from Carnegie-Mellon in 1958. Part of his research work involved heat transfer to coils which later showed up in the design of the first-class calorimeters which he built. These 72 calorimeters have been developed commercially, are sold around the world, and are a standard for making precise calorimetry measurements. One of Jim's early ambitions was to build a calorimeter precise enough to measure the germination of a single seed but his designs greatly exceeded this. The early designs were crude compared with the later designs, but still re quired correction s for the energy input from a stirrer motor. His latest design was a calorimeter which would measure the heat of mixing at high pressure and high temperature. Professor Christensen was also a prolific author He has written or edited twelve books and co-au thored fifteen major review articles. He has also pub lished over 250 research articles in a wide variety of journals. He was instrumental in the founding of inter national conferences dealing with the chemistry of macro-cyclic molecules. He has been recognized by his peers, who selected him for the Sigma Xi annual lec ture, a Blue Key lecture, and the Distinguished Fac ulty Annual Lecture at Brigham Young University. He also received the Maiser Outstanding Research Award and the Huffman Memorial Award a s an out standing thermochemist. Last August he was asked by the Chemical Engineering Division of the Amer ican Society of Engineering Education to give the 3M lecture which is based on outstanding research and contribution to the profession. (This lecture i s sched uled for publication in a future issue of GEE). Jim had a genuine love for his students, both graduate and undergraduate, and they returned that love through extra effort. He was recognized three times by the students as being the outstanding lec turer in the department and he was recognized by the college as the outstanding teacher with the univer sity's Maesar Outstanding Teaching Award (which is the counterpart of the outstanding research award.) Jim took great delight in collecting toys which demonstrated the principles of thermodynamics and used them continually to interest the student s They included the familiar bird dunking its beak into water, rotating pinwheels, and his latest acquisition a motor which worked by dipping one end in hot water. Jim s favorite courses were thermodynamics and creativity, and he exhibited creativity in teaching. He taught the students new ways of approaching old problems. He was dynamic in hi s presentations and gave an ani mated demonstration of the energy content of gas molecule s by running back and forth twirling and jumping up and down colliding with student s and walls. Almost every year during the student banquet, this little demonstration was repeated by the students as a sign of their regard. His annual backpacking trip C HEMI C AL ENGINEERING EDU C ATION

PAGE 19

to the Uintah mountains with the graduate students was a highlight of each year. He also served his community and church, the Church of Jesus Christ of Latter-Day Saints, in many ways. He was active in the education of his children and in the processes by which parents can contribute (P. T.A., etc ). He participated in great book discus sion groups, he was a cub master for eight years, and demonstrated his love of teaching by teaching adult priesthood groups. Jim's life of personal service has concluded. But his influence will continue to be felt in the lives of his family, his students, and his friends for many years to come. Dee H. Barker Roland Andrew Ragatz 1898-1987 .. w~ 1 Will' ~ Roland Andrew Ragatz, Professor Emeritus of Chemical Engineering at the University of Wisconsin, was born in Prairie du Sac, Wisconsin, in 1898 and died in Madison, Wisconsin, on May 30, 1987, at t he age of 88. He is survived by his wife Nancy, who he married in 1930, a son Andrew (Ellen), a daughter Karen Roberts (Burnell), and seven grandchildren. Professor Ragatz graduated from Prairie du Sac High School in 1915. He completed his studies for a BS degree (in the then new discipline of chemical en gineering) in 1920 at the University of Wiscon s in, Madison, and was immediately appointed instructor in chemical engineering. He remained on the teaching staff at Wisconsin until his retirement in 1969, except for a one-year leave of absence in 1929-30. While serv ing as a full time instructor, Professor Ragatz earned his MS degree in 1923 and his PhD degree in 1931 also in chemical engineering. He served as depart ment chairman during three periods totalling sixteen SPRING 1988 ----------~years, and as associate chairman for five years there after. Professor Ragatz' 49-year period of service to the University is one of the longest on record. He contri buted significantly to his college and department and to his students during the important early period when the developing field of chemical engineering was emerging from an empirical discipline and evolving into a more scientifically based study. Basically de voted to teaching, Roland's areas of interest followed the needs of the department and its students in pre paring them for industrial careers in a variety of often traditional manufacturing enterprises. In 1920 he began the development of introductory and advanced metallography courses including laboratory studies related to the microscopic structures of metals, with emphasis on mat e rials of construction. When instruc tion in this area wa s transferred to the Department of Mining and Metallurgy in 194 8, Professor Ragatz turned his interest in materials to developing courses in plastics, again including a laboratory program. This early attention to materials science established the foundation on which the department's current strong program is based. Starting in 1935, Professor Ragatz joined the late Olaf A. Hougen (and, during the 1940s, Kenneth M Watson) in giving courses in material and energy bal ances and in thermodynamics for chemical engineers and participated in rewriting the corresponding parts (I and II) of the text Chem i cal Process Pri n ciples, which long has been a standard work in chemical en gineering. During his extended service as department chair man Roland alternated with Olaf Hougen, and, in the words of Professor Hougen, "he assembled a staff well balanced in special talents for teaching and research, for undergraduate and graduate instruction, for vari ety in engineering and scientific interests, and with balance in laboratory and classroom instruction." Roland also served long and efficiently as chairman of the College of Engineering committee on fellow ships and scholar s hips, a s well as on numerous other committees. He w as a long-time member of the ASEE serving on it s C ouncil a s chairman of its Chemical Engineering Division, and as organizer of the important Summer School for Teachers of Chemi cal Engineering (at Madison in 1948). Professor Ragatz exemplified the ideals of the fac ulty of the University of Wisconsin: scholarship, re search, public service, and above all, teaching. Roger J. Altpeter Wayne K. Neill Charles C. Watson 73

PAGE 20

[eJ b #I curriculum SAFETY AND LOSS PREVENTION IN THE UNDERGRADUATE CURRICULUM A Dual Perspective DANIEL A. CROWL Wayne State University Detroit, MI 48202 and JOSEPH F. LOUVAR BASF Corporation Wyandotte, MI 48192 R ECENT CHEMICAL PLANT accidents in India and in Switzerland have increased public and indus try-wide concern and awareness of safety and loss pre vention. This is in spite of the fact that the chemical process industries are safer than most other industries [1]. However, due to the potential for a serious plant accident, both the public and industry recognize that safety must be increased. The AICHE is taking considerable steps to im prove safety. First, it has formed a Center for Chem ical Process Safety with the charge to "address the concerns about the handling of toxic or reactive mate rials and the safety of plant operating procedures" (2). Second, it has formed the Design Institute for Emergency Relief Systems (DIERS) User Group to continue the cooperative industrial activities to extend the DIERS technology (3). And third, a Task Force on Safety and Health in the Undergraduate Cur riculum has been formed under the Safety and Health Division of the AICHE with the major objective to "identify the key concepts of loss prevention, safety, and health which should be considered essential for accreditation of the curriculum by 1990" [ 4). As a re sult of this new awareness and concern for safety, it is apparent that safety and loss prevention will be come a part of the future undergraduate chemical en gineering curriculum. Most undergraduate chemical engineering cur ricula in the United States contain little in the way of safety and loss prevention. In fact, university laboratories are, typically serious safety offenders. C opyright ChE Di vision ASEE 19 88 74 Great Britain has had an ambitious safety and loss prevention program in their undergraduate cur riculum for some years now. This is a result of a sub stantial chemical process accident that occurred at Flixborough, England, in 1974 (1]. All British chemi cal, engineering curricula are presently required to contain a significant amount of safety related content for accreditation. This is achieved through a combina tion of dedicated courses or by demonstrating a cer tain fraction of safety related content in the existing courses. This article will provide a dual perspective on the need for more emphasis on safety in our chemical en gineering undergraduate curricula. Both the academic Daniel A. Crowl is an associate professor of chemical engineering at Wayne State University He received his BS degree in fuel science from the Pennsylvania State University and his MS and PhD in chem ical engineering from the University of Illinois. After graduation he spent two years as a senior process control engineer for St Regis Paper Company before returning to academic work His present research in terests are in chemical process safety process simulat i on and the ap plication of CAD techniques to process design. (L) Joseph F. Louvar is director of the Chemical Engineering Depart ment (Research and Development) of BASF Corporation Wyandotte, Ml. A member of AICHE s Loss Prevention Committee he also teaches experimental statistics at Wayne State University He received a PhD in chemical engineering from Wayne State University, an MSChE from Carnegie-Mellon University and a BSChE from the University of Mis souri (Rolla). (R) CHEMICAL ENGINEERING EDUCATION

PAGE 21

This article provides[s] a dual perspective on the need for more emphasis on safety in our ... undergraduate curricula. Both the academic and industrial viewpoints [are] presented. We ... also discuss a unique safety related course development effort being undertaken at Wayne State and Michigan Tech Universities ... and industrial viewpoints will be presented. We will also discuss a unique safety related course develop ment effort being undertaken at Wayne State Univer sity and Michigan Technological University, with sub stantial technical assistance from BASF Corporation. This program is being supported by the National Sci ence Foundation under its University/Industry/Gov ernment (UIG) program. THE UNIVERSITY PERSPECTIVE Dan Crowl In the summer of 1986 I had an opportunity to spend a few months at BASF Corporation working on a computer simulation project. One day I was visited by Joe Louvar, Director of Chemical Engineering who asked me if I was interested in working on a few safety related projects. In total ignorance I replied, "You mean hard hats and safety shoes?" Joe went on to explain some of the more fundamental aspects of safety, including reactor dynamics, two-phase flow during reactor venting, gas dispersion models, and so forth. It had never occurred to me that safety had a fundamental aspect! I had always assumed safety was simply a set of rules developed as a result of practical experience. During the remainder of my stay at BASF I ob served how safety was practiced in an industrial envi ronment. It became apparent to me that practicing chemical engineers spend a considerable amount of time on safety related activities. In spite of all my years of academic training, I felt woefully inadequate with respect to safety. I could indeed understand pieces of the fundamental components, but using those pieces in the practical application and management of safety seemed a mystery to me. Despite a continuing feeling of inadequacy, I was now aware of safety. I began to look at my past academic experiences and at the experiences we are providing for our present students. I found little in the way of safety instruction. Many of you will argue that the academic experience is designed to provide only a fundamental knowledge, with practical applica tion being the responsibility of industry. But safety is a systems science involving the application of a broad range of fundamental skills strongly coupled with practical application. Why not adjust the fundamental skills taught to our undergraduates to strengthen the SPRING 1988 safety aspect? Can the academic community continue to neglect an area that is already very important to the industrial community? As an undergraduate I learned that process design is motivated by savings in material usage and capital investment. In the 1970's energy also became an im portant driving force in design. Now, safety consider ations are becoming just as important. The academic community is in a difficult position with respect to safety. First, most faculty have little safety experience. This is either a result of little indus trial experience or of participation in industry during a time when safety was a lower priority and was prac ticed in a different fashion than it is today. Second, we have inadequate equipment and facilities for dem onstrating safety. Finally, our curricula are already at their practical maximums. Where can we find room to add a safety component? One possibility is to add safety instruction throughout all of the chemical engineering courses. Safety is practiced in every aspect of the industrial experience, so why not practice it throughout the en tire undergraduate course sequence? I agree with this approach. However, a final capstone course is neces sary to culminate the course sequence. One can teach reactor safety (runaway reactors) in the kinetics and reactor design course, but a capstone course is essen tial for the teaching of safety reviews, hazard identifi cation, risk assessment, and other topics that are special to safety. I recently attended the International Symposium on Preventing Major Chemical Accidents, organized by the Center for Chemical Process Safety of the AICHE This symposium was held in Washington, DC, during the week of February 3, 1987. Of the over four hundred participants less than five were from academia. This was surprising since the papers pre sented were as fundamental and research-oriented as those presented at academic conferences. I believe that safety can become an important academic area if faculty become aware of 1) the potentially fundamen tal nature of safety, and 2) the opportunities for re search and funding. Safety is an excellent area for industry / academic research collaboration since most of the industrial proprietary barriers dissolve. Safety is an essential part of the industrial experi ence, more now than some of the traditional funda mental chemical engineering areas. I believe it is time 75

PAGE 22

to include safety as an important part of the under graduate chemical engineering experience as well. THE INDUSTRIAL PERSPECTIVE Joe Louvar All of my safety knowledge has been acquired in an industrial setting. It came in bits and pieces, and it took several years to really appreciate the signifi cance, the subtlety, and the technical complexity of chemical process safety. In hindsight, I recognize that during those formative years I made many serious safety errors. It was only the result of pure chance that I escaped having serious accidents and/or in juries. Using today's standards, most of my past errors would have been identified and corrected by supervi sion or by the safety review process which is used very effectively within industry. Unfortunately I am convinced that the self-motivation necessary to take on the responsibility for safety (responsibility of every engineer), and to learn the technology of safety, still requires years within an industrial setting. From a management standpoint I have also recog nized that supervision has a serious handicap when working with fresh graduates. Although they have an TABLE 1 Course Outline 1. Introduction 4. Spill models 5. Inhalation exposure: spills or from vessels A. Course syllabus 6. Inhalation exposure: filling containers B. Introduction C. Accident and loss statistics 3. Applications for safety D. The accident process A. Introduction E. Three significant disasters B. Designing for safety F. Personal and management responsibilities 1. Intrinsic design G. Legislative responsibilities 2. Extrinsic design H. Employee and community "right to know" 3. Maintenance 2. Fundamentals of Safety A. Toxicology, industrial hygiene and exposure control 1. Toxicology History Dose vs response How the body responds to exposure Determining safe working levels 2. Industrial hygiene and exposure control Types of exposure Methods for exposure control Administrative and engineering methods Personal protection C. Engineering to Prevent Fires and Explosions 1. The Fire Triangle 2. Passive Protection Methods Types of environments Eliminating sources of ignition Atmospheric control, including inerting and purging Plant siting 3. Active Protection Methods Shutoff and check valves Combustible gas monitors Emergency material transfer Sprinkler systems Water vs. foams for fire control B. Fires 1. Flammability of vapors and liquids 2. Minimum oxygen concentration 3. Inerting 4. Autoignition temperature 5. Auto-oxidation 6. Adiabatic compression 7. Effects of sprays and mists C. Explosions 1. Definitions: Explosion, detonation deflagration, confined and unconfined vapor explosion, BLEVE 2. The nature of the explosion process 3. Effects of explosions 4. Calculations relating to explosions D. Engineering to Prevent Toxic Release 1. Relief systems 2. Design and selection of relief valves 3. Flares, vents and scrubbers E. Process hazards identification and risk assessment 1. Hazards identification Hazards surveys, including the Dow Index HAZOP FMECA 2. Risk Assessment Event trees Fault trees Consequences analysis Limitations to risk assessment D. Liquid and Vapor releases 3. Safety reviews 1. Gaussian distribution 2. Gaussian plume model 4. Accident Investigations 3. Gaussian puff model 5 Case histories 76 C HEMICAL ENGINEERING EDUCATION

PAGE 23

excellent foundation in the traditional chemical en gineering fundamentals, the concepts of safety are foreign to them. In fact, they perceive safety to be unimportant because the subject is hardly mentioned during their schooling. In some universities, safety practices are totally neglected. Consequently, indus try starts with graduates who need a basic improve1. Introduction TABLE 2 Video Session 1 INTRODUCTION 2. Brief description of five video sessions 3. BASF Corporation's safety program a. Policy b. Committments c. Training d. Activities e. Audits 4. Introduction to safety terminology and principles a. XP vs. non-XP b. Relief vs. rupture disc c. Runaway reaction d. Two phase flow e. Design Institute for Emergency Relief Systems (DIERS) TABLE 3 Video Session 2 BASIC SAFETY EQUIPMENT AND PROCEDURES 1. Introduction 2. Personal protection equipment a. Motivation b. Face and eye protection: glasses, goggles, face shield c. Clothing: aprons, gloves, suits d. Respirators: dust, cartridge, canister, air-line e. Miscellaneous: hats and shoes 3. Sources of toxicity and safety information a. Material Safety Data Sheets (MSDS) b. Vendor information 4. Safety procedures a. Hot work permits b. Lock Tag Try c. Vessel entry d. Grounding and bonding e. Fail safe f. Safety reviews 5. Safety features a. Sprinklers b. Alarms c. Showers d. Color indicator tubes e. Extinguishers SPRING 1988 ment in attitude prior to addressing the more complex features of safety. The chemical industry is entering an era where processes will be even more complex. To prepare our future engineers for this era, I believe we must begin to educate them on the principles of safety. This edu cation must begin at the university, with an emphasis equal to heat transfer, mass transfer, ther modynamics, etc., and should be given during the same period as the core courses. This would be an effective way of emphasizing the importance of safety. I believe that there should be a three-hour course during the third year which is dedicated to safety. Some educators believe that safety should be a part of every course, and I agree. But the more technically complex areas of safety need more time and attention than could possibly be allotted in core courses. The following subjects could only be adequately covered in a separate safety course: Dispersion studies Relief valve sizing (including 2-phase flow) Safety reviews and hazard identification Flammability of chemicals Toxicity of chemicals The benefits of teaching safety in the university exceed those mentioned above. An important spin-off of safety courses could be the development of more interest in initiating safety-oriented research. In a re view of university research (PhD dissertations), it is apparent that there is very little research devoted to safety. Topics which could be fruitfully addressed by universities include: Advanced adaptive reactor control methods for hazardous reactions Expert systems for monitoring reactor and/or plant safety Expert systems to improve the reliability of fault analysis Advanced relief methods for runaway two-phase flow sys tems From my perspective, the chemical industry will continue to stress faster reactions, more complex reaction systems, and a more complex utilization of investments via multiple product reactor systems. The success of these complex processes will depend upon our ability to design modern systems with the safety features demanded by both industry and the public. To meet these challenges of the future, we must give more attention, concern, and respect to safety. 77

PAGE 24

NEW COURSE DEVELOPMENT We have developed a senior level course entitled "Safety in the Chemical Process Industries." This development effort was supported by the National Science Foundation under their UIG program. The unique feature of the course was that it included five two-hour lectures, broadcast live from the chemical pilot plant facilities at BASF Corporation in Wyan dotte, MI. The format supported interactive question ing between the students in the classroom and the practicing chemical engineering professionals in the pilot plant. The broadcasts were uplinked to a satellite for downlinking at both Wayne State University in Detroit and Michigan Technological University in the upper peninsula. The course was taught simultane ously at both locations during the fall of 1987. The course provided a rare opportunity for stu dents to see safety being practiced by professional chemical engineers in an actual chemical plant envi ronment. Demonstrations were provided using actual equipment and safety situations that could never be shown in the classroom or laboratory. The video lec tures were in addition to a series of25 one and one-half hour classroom lectures. These lectures presented the fundamental and theoretical features of safety, sub jects requiring problem solving and discussion within the classroom environment. TABLE 4 Video Session 3 INSPECTION OF LABORATORY AND PROCESS AREA 1. Introduction 2. Inspection concepts 3. Review film on Stop Observe Act Report (SOAR) 4. Inspection of laboratory (several examples of poor safety practices will be staged) a. Storage of solvents b. Storage of glass equipment c. Safety equipment and operation d. Principles of using hoods 5. Inspection of Process Development (PD) area (several ex78 amples of poor safety practices will be staged) a. Reliefs and rupture discs b. Nitrogen vented in room c. Non-XP in XP room d. Poor grounds e. Belts not guarded f. Bad drum vent g. Incorrect tools h. Poor ventilation i. No double block and bleed (show correct configuration via glass system) j. No hot-work permit A total of four industrial and seven academic con tributors were involve with the effort. The BASF Corporation industrial participants were: G. W. Boicourt, M. Capraro, J. F. Louvar, and J. Strick land. Their effort was directed mostly towards the video lecture material and scripts. J. F. Lou var also contributed towards the development of lecture mate rial. The academic participants were D. A. Crowl and R. W. Powitz from Wayne State University and M. Banks, B. A. Barna, E. R. Fisher, N. K. Kim, and D. G. Leddy from Michigan Technological University. The major focus of the academic group was toward TABLE 5 Video Session 4 EXPERIMENTS FOR SAFETY 1. Introduction 2. Vent Sizing Package (VSP) for sizing reliefs a. Show features of equipment b. Show types of data collected c Illustrate specific tests for various objectives d. Introduce relief valve sizing concepts 3. Characterizing dusts a. Deflagration index b. Illustrate features of equipment c. Describe test procedures d Describe data from tests e. Introduce relief sizing concepts 4. Flammability limits a. Illustrate features of equipment b. Describe test procedures c. Describe data from tests d. Introduce relief sizing concepts 1. Introduction TABLE 6 Video Session 5 SAFETY REVIEWS 2. Informal safety review in laboratory a. Describe phosgenation system b. Analyze procedures relevant to safety c. Progressively make improvements via team dialogue concept 3. Formal safety review a. Introduction b. Safety Review Meeting c. Equipment inspection d. Wrap-up meeting and development of Action Plan 4. Wrap-up of video sessions a. Summary remarks b. Open questions and answers CHEMICAL ENGINEERING EDUCATION

PAGE 25

development of the lecture materials. The overall pro ject was managed by D. A. Crowl. The course outline is shown in Table 1. It is divided into two major parts. The first part presents the fun damentals of safety and includes discussions of tox icology, fire, explosion, and toxic release. The second part deals with using those fundamentals in practice and includes a discussion of "designing for safety" and using various safety review procedures (such as hazards and operability studies). The course also in cludes a discussion on case histories and accident in vestigations. The outlines for the five video lectures are shown in Tables 2 through 6. Except for video session 4, the videos are not dependent on the lecture material. The emphasis of the videos was to show the students how safety is practiced on real process equipment. The fourth video lecture on "Experiments for Safety" re quired some fundamental lecture material prior to broadcast. kib a book reviews INTRODUCTION TO POLYMER VISCOELASTICITY, Second Edition by John J. Aklonis, William J. MacKnight John Wiley & Sons, Somerset, NJ 08873; $39.95 (1983) Reviewed by Albert Co University of Maine This book introduces various fundamental concepts in studying the viscoelastic behavior of polymers, with an emphasis on the molecular approach. The book con sists of nine chapters. Chapter 1 introduces the reader to several exper iments that display the viscoelastic nature of poly mers. In Chapter 2, viscoelastic material properties in transient and oscillatory experiments are defined and are illustrated clearly with simple experiments. The Boltzmann superposition principle is stated; its applications in relating the creep compliance and the stress relaxation modulus and in relating transient and oscillatory properties are demonstrated. In Chapter 3 the regions of viscoelastic behavior are described and the effects of molecular weight, crystallinity, and plasticizing agents are explained. The concept of time-temperature superposition, the master curves, and the WLF equation are then pre sented. In Chapter 4, the phenomenon of glass trans ition is examined, and explanations based on free vol ume, thermodynamics, and kinetic theories are pre sented. The effects of structural parameters on glass SPRING 1988 SUMMARY This paper has presented both the industrial and university perspectives regarding the need for teach ing safety in the chemical engineering undergraduate curriculum. We have also presented one approach to teaching safety and loss prevention. As a result of NSF support we had a unique opportunity to bring the students into an operating chemical pilot plant, through the use of live TV. We hope that this approach, and others, will im prove the engineers of the future and result in safer chemical process plants. REFERENCES 1. Lees, F. P ., Loss Pr evention in the Process Industries, Butter worths, Boston (1986). 2. Anonymous, AICHE En vironmenta l Division Newsletter, February, 1986. 3 Fisher, H. G., "DIERS Research Program on Emergency Re liefSystems," Chemical Engineering Progress, August (1985). 4 Glass, Arnold J., Letter of February 2, 1986. D transition temperature and the relaxation occurring in the glassy state are rationalized in terms of molecu lar motion and chain mobility. In preparing the reader for subsequent chapters, the statistics of a polymer chain are reviewed in Chapter 5. In Chapter 6, various treatments of rubber elastic ity and the structural factors that affect rubber elas ticity are discussed. In Chapter 7, the behavior of typ ical mechanical models is analyzed and the Rouse Zimm molecular theories for polymer solutions are discussed. Extensions of these molecular theories to bulk polymers are then considered and the reptation theories are briefly described. In Chapters 8 and 9, the phenomena of dielectric relaxation and chemical stress relaxation are examined, respectively. Throughout the book, the mathematical treat ments are maintained at a level comfortable for under graduates. Advanced mathematics required for the discussion of a subject matter are elaborated in the corresponding appendices. The problems at the end of each chapter range from simple calcu lations to ad vanced problems requiring a certain degree of mathematical sophistication. Readers will find the so lutions located at the end of the book to be helpful. Overall, this book is an excellent introduction to polymer viscoelasticity. However, the treatise is re stricted to amorphous polymers. The treatment on crystalline polymers is very limited, and topics such as solution behavior, melt rheology, and birefringence are not covered. Nevertheless, it is a good choice as a textbook for one of a series of courses on polymer viscoelasticity. D 79

PAGE 26

[eJ b a curriculum ENGINEERING MANAGEMENT A Course to Minimize Functional Handicaps of Graduates REX T. ELLINGTON University of Oklahoma Norman, OK 73019 T HE INDUSTRIAL ECONOMY in the u. S. has un dergone significant changes since 1982. Some of the changes have been initiated by recent intensive overseas competition, some by the drop in oil prices, and some by mergers. In the 'right sizing,' or downsiz ing, that has taken place in most companies in this decade in order to maintain competitiveness whole layers of management have been expunged. Other changes resulted from technological developments in the late 1970s and in the first half of this decade. De velopments in computer workstations and software have added such power to the individual engineer that companies in chemical processing and the engineering companies serving them will never again need as many engineers of any discipline for the same volume of business as they needed in the 1975 era. As a conse quence, there has been a reduction in hiring by most Rex T Ellington has been professor of chemical engineering at the University of Oklahoma for five years after nine years at IGT (Illinois Tech) and nearly thirty years i n i ndustry He received his BS (ChE) from the Un iversity Colorado and the MGT MS (ChE) and PhD from IIT He held r esearch and general management pos i t ions in Sinclair Oil At lant ic Richfield and Transco Energy and had project management po sitions in Sinclair, Transco, Fluor Eng ineers and Constructors and O cci dental. Along the way he had a privat e consulting practice for a year His teaching has concentrated on senior design, process contro l and engineering management 80 companies, with the result that many students have not been able to find jobs even months after gradua tion. Competition for the available jobs has become intense and many young engineers have had to go into fields where they may never be able to use their hard-won training. The purpose of this paper is to review a course, "Engineering Management," that has been developed at the University of Oklahoma to im prove the students' ability to get jobs and to perform better in any company. Important technical consequences of the industrial changes which were described in a recent article in this journal [1] agree with my own observations as a senior manager that A smaller percentage of all engineers, especially in larger companies, will be able to move into management. All engineers will have to be more competent technically and to remain competent longer before they can become managers. While they practice their specialty, engineers will have to function more effectively in a company environment to have job security. Engineers will need far greater organizationand people skills than past graduates possessed. The needs represented by the last item have been voiced by every industrial advisory group that I have known or been a part of in the last twenty years. As soon as recruiters identify a lack of understand ing of the industrial environment and an obvious lack of people-skills in interviewees, they lose interest. These graduates may never get an engineering job, or they may be shuffled aside at an early date. As recently as April 1987, senior managers advising our department told about students who were essentially unemployable, no matter what their GP A, because of their obvious lack of people-skills. The student needs to understand how the industrial world operates and should develop people-skills before leaving college so he/she can function effectively immediately after being hired. Costs are simply too high for businesses Copyright ChE Division ASEE 1988 CHEMICAL ENGINEERING EDUCATION

PAGE 27

to tolerate a one-year shakedown period for new en gineers, as they did in the past. Another educational need is based on the convic tion of most senior managers that learning to manage is a lifetime job. This learning process must be started very early by engineers in order to meet the greater competition for managerial positions or they will be left behind, and business majors will skim off an even larger share of the top jobs in the future. CHANGING THE CURRICULUM C. T. Sciance [1] cites a significant need to aug ment the technical content of the chemical engineering curriculum, but it could be even more important to improve people-skills and knowledge of how in~ustry functions. Many students gravitate toward engmeer ing because their natural aptitudes lie more in math and science and less in communication and interper sonal relations. Throughout their engineering study this uneveness in skills probably increases. Students are shocked to find, while being interviewed or upon being hired, that the practic~ of engineering is a very people-oriented profession. They soon realize they are woefully unprepared in many areas that could affect career achievement even more than their calculation ability. A serious gap in their education soon becomes apparent. The effects of the new employment environment have only lately reached the universities in the form of reduced job opportunites, but few changes have been made in the educational program in response to the needs of those now graduating and to the next generation of students. Why has the gap in organiza tional know-how not been filled? Many professors have no industrial experience, or it was of such short duration, or at such a junior level that they do not know what is really involved and are uncomfortable trying to teach such material. They may agree with the need but they lack the tools to deal with it. An effort has been underway at the University of Oklahoma to develop a local solution to the problem for students willing to devote three semester hours to the effort and to draft a text which could be used by other fac~lties. Further, we hope that the text might ultimately become a desktop reference which will be consulted more frequently than many technical refer ences. COURSE SPECIFICS Ideas for this course have gestated for more than a decade through thousands of goal-settings, perfor mance r;views, hirings, firings, and many mistakes SPRING 1988 The course then becomes one of high student-participation. Basic value systems are discussed to help the student understand why others react to life's inputs differently than the student himself does. by the author. They came from fellow executives of all ages who urged that students be made aware of the things that cause so much difficulty in industry and handicap so many careers. Time and time again, the root causes of those difficulties are poor communi cation and lack of interpersonal relation ability, in adequate knowledge of company workings, or dis torted concepts of management. Lately, new ideas have come as feedback from recent graduates. This is a course in what life is really like "out there." Many of the rough edges have been ground off. Attendance has been interdisciplinary, and the first results have been good. Feedback and requests from earlier graduates have resulted in the develop ment of a text. Graduates have already reported on more comfort in working in team situations a greater ability to determine supervisors needs by quickly de termining how an organization works, and less fear of moving into project work immediately. These results suggest that the effort is moving in the right direction. The first draft of the text was developed to meet classroom needs and it is outlined in Table 1. The first part of the course is very important for the student and new graduate. It starts with a brief perspective of the entire course and then presents a view of the types of behavior and functioning expected of all em ployees. The characteristics of effective employees and good supervisors are discussed. There are numer ous "horrible examples" to awaken the student to things that can happen which develop a bad reputa tion. The intent is to open the student s mind to the concepts that lead to being an effective person. The course then becomes one of high student-par ticipation. Basic v alue s ystems are discussed to help the s tudent under s tand why others react to life's in puts differently than the student himself does In the main students s eek the company of peers and cliques with whom they a re comfortable. It is often a shock for them to realize that they may have to get along and work with people who see life differently. The value system is a convenient mechanism for identify ing differences in reaction to one' s environment. Next, attention turns to getting a jo~the right job for the individual. The student is taught how to take full, personal charge of job-finding. Careful as sessment of the s tudent 's attributes leads to the de81

PAGE 28

velopment of lists of job opportunities, location of target companies, investment of significant time on company research, and improved communication in resumes and interviews. Effort is expended to heighten the student's awareness of his/her impact on recruiters, interviewers, and other people during plant trips. Then, important things that the newly employed engineer should consider in order to get the best pos sible start on the job are reviewed. These are obvious truisms, but they are often overlooked or ignored. It is necessary to develop a different frame of mind in the students in order to attack the deficiency in communication ability which is mentioned by nearly all supervisors. Some of the most important areas concern the study of personal ethics, business ethics, technical and intel lectual property of a company, liability problems (to give the student perspective regarding what must be protected), and obligations during and after employ ment. Then they examine one problem all of them will encounter some day in some way: whistle blowing Moral obligations are analyzed in detail. It is necessary to develop a different frame of mind in the students in order to attack the deficiency in communication ability which is mentioned by nearly all supervisors. Communication theory is examined briefly to learn why some of the best intentions go awry in speaking and writing. Throughout the course there are opportunities to make brief formal presenta tions, and each is critiqued by the class for its organi zation and delivery. At least one presentation is re peated several times in order to force the individual to revise and improve it. Written assignments are re quired frequently. The student is taught to observe audience awareness, whatever the occasion. The true function of management is examined next and it proves to be a revelation for many students. As a result, some students are less anxious to become managers, but at the same time they develop a better appreciation of the needs of supervisors and mana gers. To help the student understand his/her new en vironment, subjects such as types of company organi zation, job descriptions, job evaluations, goal setting for individuals, performance reviews, and merit budgeting are taken up. Leadership types are examined in some detail so that young employees can better characterize the behavior of supervisors and work with individuals more effectively. Equal employ ment opportunity requirements are carefully dis82 cussed to help people avoid problems and/or to insure that their rights are observed. The added awareness provided by these subjects makes the student a better employee. People who have had the course seem to get a faster start on the job and to have an edge on the competition-befitting the adage, "If at first you don't succeed, you may never." They are more aware of the importance of fitting into group activities, have greater empathy for a supervisor's needs, and are able to develop a per sonal program for advancement at an early date. The rate of return for the time and money invested in this part of the course appears to be very high. MODERN TECHNOLOGY EMPHASIZED The middle section of the course deals with modern methods of modeling, planning, and analysis. Tech niques and sophistication vary widely from company to company, so the treatment provides a look at the utility of these tools without taking a course in opera tions research or a business course in modeling. Each method is examined from a management rather than a program writer's viewpoint, to show how the analyst and manager must work together and how model com puter outputs or "decisions" are used. General economics and spreadsheets are reviewed quickly to provide the basis for economic decisions. Reading and analysis of company financial reports is stressed as an ability that every young engineer should have. The linear programming, which is one of the most widely used optimization aids-from refinery operation to vegetable planting-is taken through graphical inter pretation, matrix methods, analysis of computer out puts, and sensitivity studies. Study of integer pro gramming provides the basis for analyzing supply, transportation, and other networks. Among these are the critical path and PERT techniques. Next, the stu dent is presented with the structural basis for making decisions on a more rational basis. Decision theory considers the different paths electable in solving a problem and deals with the probable cost or conse quence of picking each. Several types of business examples are used. Reliability analysis is a field which the Japanese have adopted more widely than we have in the U.S. for a number of years. Failure, repairability, main tainability, and availability are taken up in terms of process plant units and design for reliability This leads to consideration of risk analysis, fault trees, and failure mode and effect analysis. The fact that the rigor of the methods clarifies thinking is immediately clear. Results of these analyses do no good in the files, CHEMICAL ENGINEERING EDUCATION

PAGE 29

however, and emphasis is placed on continued think ing by those involved in operations in order to prevent future catastrophes such as those which have occurred in the last decade. By this time, students realize the limitations of deterministic inputs to models. Distributions are con sidered for input values and Monte Carlo methods are applied to such things as cost estimates and se quential operation simulations. This section of the course provides a new awareness and perspective of tools most engineering students have not seen before. With this awareness comes the ability to help super visors through the use of tools lacking in other en gineers and the ability to sharpshoot with further study when the time comes. PROJECT MANAGEMENT TOOLS TAUGHT Since so many tasks in business and engineering are of a project nature the third part of the course treats management of projects from beginning to end. Few young engineers know how projects get started ; therefore, design bases, requests for proposals, pro posals, contracts and agreements are carefully reTABLE 1 Table of Contents PART I. Functioning with People in Organizations 1. Introduction 2. Value Systems 3. Getting the Correct Job 4. Ethics Technical Property and Legal Issues 5. Getting Along on the Job 6. Communication 7. Functions of Management 8 Company Organization 9. Managing People PART II. Planning and Analysis Methods 10. General Economics 11. Linear Programming 12. Integer Programming and Networks 13. Decision Theory 1 4. Reliability Fault Trees, FMEA 15. Probabalistic and Simulation Models PART III. Work Management (Project Basis) Glossary 16. Project Initiation 17 Project Organization/Kickoff 18. Engineering and Detailed Design 19. Equipment and Material Procurement 20. Time and Money control 21. Quality Assurance 22. Construction and Manufacture 23. Project Completion SPRING 1988 viewed and discussed to show that projects don't "just happen. A typical project is organized and engineer ing and detailed design activities are defined. Project activities and individual engineering discipline ac tivities are related to total project progress curves, with emphasis on the interrelationship of activities. Probability of success on any project is related to the amount of planning, to the handling of details, and to the accuracy of measurements that go into it. All pro curement activities (spec sheets, inquiries, bids, bid tabs, vendor selection, expediting, inspection, vendor data acquisition, and transport) are put into time per spective. Time and money control discussions involve activity definition, loading of effort on activities, crit ical path analysis, time and cost estimates, and cost control. Emphasis is placed on the need to measure progress in the same terms as the original estimate for accurate progress reporting. Product and service quality assurance needed for better reliability, maximum productivity, and lower costs in the face of overseas competition, provide background for treat ment of construction or manufacturing and project completion. At the end of this section, many students under stand the unusual characteristics of good project en gineers. Although they may not want to enter this specialty, they can function much more effectively on any project and realize the benefit of a project-man age-approach to their own lives. CONCLUSIONS Obviously, many topics are treated quickly, and not all are treated every semester. But the objective is perspective and development of the "big picture" viewpoint. With this the young engineer knows where to dig when the need for details of a given method or topic arises. Over the years, the author has found that the engineer who has this perspective is most often the one who defines the problem correctly and comes up with the needed solution at the right time, i .e ., is the most promotable individual. The chapter listing for the text is given in Table 1. Comments and ideas from educators or indus trialists are always welcome. As a veteran project manager the author s primary interest is improving the product and serving the customer (the student and the employer). REFERENCES 1. Sciance C. T. Chemical Engineerin g in the Future, Chem E ng Ed X XI No 1 Wint e r 19 8 7 83

PAGE 30

(eJ b I curriculum A COURSE ON PRESENTING TECHNICAL TALKS RICHARD M. FELDER North Carolina State University Raleigh, NC 27695 I N THE ONGOING debate about what should be put in and taken out of the chemical engineering cur riculum, one of the few points of universal agreement is that oral communication skills are essential for all practicing engineers and not enough is done to develop them in most engineering curricula. Like many other departments we used to have a senior seminar course in which each student presented a single 30-45 minute talk sometime during the seme s ter. The results were not particularly impressive: the students whom you would have expected to give good talks gave them; the other talks ranged from poor to average; evaluations were superficial; and the one shot nature of the talks provided little opportunity for individual improvement. Also, the worst of the talks, being as long as 45 minutes, were excruciating experi ences for everyone involved. When Harold Hopfenberg became Department Head eight years ago one of the first changes he in troduced was a complete reorganization of the seminar course. Under the new system the senior class is di vided into groups of six to eight students and a facult y member is assigned to each group. The groups meet once a week for most of the semester. Each faculty member runs the seminar in any way he or she chooses; the next section describes a structure that has worked particularly well for the author. COURSE STRUCTURE Each student prepares and delivers two talks dur ing the course. Talks in the first round are each fifteen ... one of the first changes [Harold Hopfenberg] introduced was a complete reorganization of the seminar course. Under the new system the senior class is divided into groups of six to eight students and a faculty member is assigned to each group. Copyright C hE Di vision AS E E 1988 84 Richard M. Felder i s a professor af ChE a t N C. State wh e re he has been sin c e 1969 He re c e i ved his BChE at C i ty College a f C.U N Y. and his PhD from Princ e ton He ha s worked at th e A.E R E ., Harwell and Brookhaven National Laborat o ry and ha s pr e sent e d co urs es o n c h e m ical engineer i ng principles reactor de s ign pr oc ess opt i m iz a t i on a nd radioisotope applicat io ns to various Am e ric a n and foreign i ndustri es and institutions He i s c oauthor of the te x t Elementary Principles af Chemical Processes (Wiley 1986) minutes long for a 50-minute clas s period and twenty minutes long for a 75-minute period. Talk s in the sec ond round are ten minutes long. The first meeting is organizational. The instructor explains the course structure and hands out and re views a list of suggestions for good presentations (to be given subsequently in this paper). He then lists all of the dates on which the course will meet and re quests volunteers for the first two presentations to be given two weeks from the current date. If there are volunteers the rest of the calendar can usually be filled in on a voluntary basis; if not, a lottery i s u s ed to assign presentation dates. In the second period the instructor delivers an il lustrative seminar. He begins by explaining that he has been giving technical talks for a long time and the students should not expect to be able to duplicate his skill at this stage of the game. He adds that even experienced speakers can find room for improvement, and he requests that the students take notes on things that might be improved in the presentation. He then C HEMI C AL ENGINEERING EDU C ATION

PAGE 31

----He provides no real introduction but launches directly into a long monologue read verbatim from notes. He shows a crudely drawn flow chart with no units or streams labeled; several data plots with no apparent relationship to anything; and one transparency that looks like a facsimile of the Dead Sea Scrolls ... no eyecontact is made; and gum is chewed continuously and ostentatiously. proceeds to give the worst talk he can possibly give. He provides no real introduction but launches directly into a long monologue read verbatim from notes. He shows a crudely drawn flow chart with no units or streams labeled; several data plots with no apparent relationship to anything; and one transparency that looks like a facsimile of the Dead Sea Scrolls, filled from top to bottom with handwritten equations in tiny print. The talk is crammed with undefined technical jargon; no eye contact is ever made; and gum is chewed continuously and ostentatiously. The speaker concludes by fumbling around with his transparencies and then muttering, ''Well, I guess that's about it." The talk lasts for approximately eight minutes and by the end of it the students have all caught on to what is happening. The class then brainstorms the things that were wrong about the talk and what should have been done differently. The instructor sub sequently gives a coherent version of the same talk complete with introduction, body, summary, and in telligible transparencies, and the class briefly discus ses the things that made a difference relative to the first presentation. At least one week before students are scheduled to present their talk they are required to submit a topic. They may talk about published papers or about work they did on projects or summer jobs; the only ground rule is that the talks should be reasonably heavy in technical content at a level appropriate for chemical engineering seniors. The instructor reviews the chosen topics and tells the students either to go ahead with them or to find alternatives with more technical content. During the week before the presen tation the students write, duplicate, and hand out seminar announcements and abstracts to the instruc tor and to all group members. A typical class session begins with the presenta tion of the first talk. During the talk the class mem bers jot down questions about the content and com ments on the presentation. A five-minute questioning period follows the talk. The students and instructor then fill out a checklist rating various aspects of the talk (introduction, body, and summary; level of the material presented; use of time; quality of transparen cies; clarity of presentation; speaking style; strong and weak points of the talk). Finally, the students and the instructor each present brief oral critiques. The SPRING 1988 checklists are given to the speaker to review and are handed in to the instructor at the following class ses sion. The procedure is then repeated for the second speaker. Since the course was last offered the department has acquired a videotaping facility. In the future, each talk will be taped and the students will be requested to view their presentations and critique themselves. SUGGESTIONS TO SPEAKERS Preparation and Organization Know the technical background of your audience and gear your talk to that level. Do not use a lot of unexplained technical jargon unless you are sure the audience already knows what it means, and don't explain what a material balance is to a group of chemical engineering seniors. Make sure your talk has a distinct introduction (outline what you are going to say and why it might be important or interesting to your audience) a body, and a summary (repeat what you particularly want your audience to retain from the talk). Use overhead transparencies or slides to present main points and provide explanatory details in the talk. Trans parencies work well for informal seminars, and you can easily make them yourself. Slides are more difficult to pro duce, but they are often more convenient for short tightly timed presentations and they are required at some national and regional conferences. Never present a large body of information orally without summarizing its main points on a transparency. Do not present more than about eight lines on a single transparency. Transparencies crowded with information are useless. It should take about two minutes to go through a single transparency in the talk. Use large type on transparencies--a label maker or the Orator ball on a Selectric typewriter or a word processing program with variable type size. Ordinary size type doesn't look good. If you handwrite the transparency, use large block lettering with horizontal guidelines to keep your lines straight. Never use script unless you're Octave Leven spiel. If you show a process flow chart, make sure the units and streams are labeled. A bunch of unlabeled boxes and lines with arrows is worthless to the audience. Try to avoid complex equations, which can rarely be explained intelligibly in the amount of time available. If you are talking about a mathematical model, focus on what it does (input and output variables, assumptions) and provide at most, qualitative summaries of the mathemati cal and computational details. (If listeners want more de85

PAGE 32

tails they can ask you for them in the question period.) If you show data plots be sure the axes are clearly labeled. Rehearse your talk several times with a friend or in front of your mirror, and make sure the time it takes is within one minute of the time allotted for the talk. Running long can be a di saster in a formal presentation and running short may not win you any friends if you're at a meeting where consecutive talks are scheduled at set times. Presentation Never read directly from prepared text-there is nothing more deadly to an audience. Make frequent eye contact with your audience throughout the talk. Do not stare at your notes or at the screen. Sou nd enthusiastic about your subject, or at least in terested in it. Do not speak in a monotone. Gesture occa sionally. If you seem bored by your material you can be guaranteed your audience will follow your lead. Make sure your watch is visible and check it occasionally to see how the time is runnning. If you see you are running short or long, try to adjust the speed of your presentation to compensate. DISCUSSION The improvements in the student presentations as the semester progresses are clear and frequently dramatic. Almost invariably poor speakers become adequate, adequate speakers become good, and good speakers become better. During the past six years a student from our department has won the regional AIChE student chapter paper award competition three times. We can't prove it, but we are convinced that the seminar course has a lot to do with this re cord. The oral critiques are a valuable and interesting part of the course. The natural student tendency is to be excessively polite, to avoid criticizing harshly lest they themselves come in for the same treatment when it's their turn to speak. As a result, in the first few sessions the principal burden of criticism falls on the instructor. However, as the semester pro gresses the student criticisms become more and more germane and incisive, although courtesy is always ap propriately retained. (We are Southern here, after all.) By the end of the semester the instructor is al most redundant: the points he is prepared to make in his critique are usually made first by the students. Requiring each student to give a fifteenor twenty-minute talk and subsequently a ten-minute talk seems to work very well. It is usually difficult (even for seasoned professionals) to present a signifi cant body of technical material in twenty minutes; 86 having to do so provides the students with excellent practice in preparing technical seminars such as those at national AIChE meetings. Cutting the material down to ten minutes presents a whole different set of problems, as the students quickly discover. The latter exercise is good preparation for, say, company staff meetings at which many people must summarize their work in a relatively short time. Finally, it is critically important for the course in structor to remember that the students taking the course are particularly vulnerable: they are nervous about public speaking in general and they are espe cially not used to being publicly critiqued. If the criti cism is destructive or unduly harsh, or seemingly arbi trary and unfair, the course has the potential of doing much more harm than good. However, as long as the instructor establishes firm ground rules about criti cism and takes the lead himself in creating a suppor tive environment, the course can be among the most positive and rewarding educational experiences the students experience in their academic careers. 0 [ej;j ?I book reviews CATALYST DESIGN: PROGRESS AND PERSPECTIVES by L.L. Hegedus, A.T. Bell, N Y. Chen, W.O. Haag, J. Wei, R. Aris, M Boudart, B C. Gates, and G.B. Somorjai John Wiley & Sons, Som e rset, NJ 08873; 2 88 pages, $47.50 (1987) Reviewed by R. J. Gorte University of Pennsylvania While there are a number of books on catalysis, it is very difficult to find a book which gives a balanced presentation of the many topics in this field The prob lem is that everyone working in catalysis has a differ ent view of what the subject is and what is important. People working in surface physics view catalysts as adsorption on single crystals in ultra-high vacuum, mathematical modellers view it as concentration and temperature gradients across a catalyst pellet, and traditional workers in catalysis view it as the turnover number or selectivity for a reaction carried out over a fixed bed. While not written specifically as a textbook, Catalyst Design: Progress and P e rspecti v es has tried to give an overview of work carried out by CHEMICAL ENGINEERING EDUCATION

PAGE 33

all types of catalyst researchers by bringing together leaders from several of the important areas in the field and having each write a brief review of the important aspects of their particular area. The book itself is a series of short review articles, each written by a different author. The first chapter, written by L.L. Hegedus, very briefly discusses the continued importance of heterogeneous catalysis to in dustrial practice and lists the applications which uti lize the largest quantities of catalysts. A microscopic viewpoint of catalysis on single crystal, metal surfaces is presented by G.A. Somorjai in Chapter 2. The work cited is mainly from Professor Somorjai's own research and discusses the results of reaction and adsorption studies on single crystals, in cluding topics such as the importance of crystallo graphic structure for reactions on metals and the influ ence of surface modifiers on several example reac tions. It should be noted, however that some of the conclusions reached in this chapter are still controver sial within the surface science community. The third chapter provides a discussion of sup ported, organometallic clusters by B.C. Gates. The chapter begins with a review of catalysis by transition metal clusters and continues with a discussion of work carried out to anchor these compounds to a support. This second part reviews the synthesis of supported complexes and concentrates on the spectroscopic tech niques which have been utilized in characterizing these catalysts. Following the section on synthesis and characterization is a discussion of the catalytic properties for several example catalytic systems. Chapter 4, by A. T. Bell, is a review of supported metal catalysis, with an emphasis on the effect that the support can have on the metal. The chapter re views a wide range of topics, including support acid ity, preparation procedures for introducing metals onto a support, and the influence that a support can have on a metal's adsorption and reaction properties. Most of this last section involves a discussion of the unusual properties which can be observed with titania supported metals. It should be noted that Professor Bell presents certain conclusions concerning the role of titania which are still being debated in the litera ture. A discussion of reaction kinetics and the design of catalytic cycles is given in Chapter 5 by M. Boudart. Since most reactions involve several elementary steps, Professor Boudart suggests ways for logically designing catalysts assuming that the intermediate SPRING 1988 steps can be selectively altered by judicious choice of catalyst or operating conditions. W.O. Haag and N.Y. Chen have written a review of acid catalysis by zeolites in Chapter 6. Their chap ter starts by describing what zeolites are, followed by a discussion of zeolite properties including sorption behavior, diffusional phenomena, and catalytic activ ity. The chapter includes a concise introduction to preparation methods for zeolites, to techniques for characterization of the acid sites, and to methods for changing zeolite acidity. Following this introduction, the role of zeolites in several commercial processes is described with a particular emphasis on the impor tance of molecular shape selectivity in those proces ses. The chapter ends with an overview of the design principles which were incorporated into the develop ment of the first zeolite hydrocarbon cracking catalysts. The section on mathematical modelling of trans port properties in catalysis, Chapter 7, was written by R. Aris. The chapter begins with the history be hind calculations of catalyst effectiveness and follows with a tutorial on how to determine the influence of catalyst geometry reaction kinetics, and other factors on the observed reaction rates. The chapter includes a short section on methods for controlling the distribu tion of catalytic activity within a catalyst pellet and concludes with a discussion of rate multiplicities and stabilities. The final chapter, written by J. Wei, presents the design considerations used for hydrodemetallation catalysts. The chapter begins by introducing the reader to the complex structure of metal-containing molecules which are present in petroleum. The rest of the chapter reviews the problems associated with de hydrodemetallation in the presence of hydrodesulfuri zation and discusses the principles used to design catalysts which have a high activity for long periods of time. Over all, this book provides a good review of a wide range of topics in heterogeneous catalysis. While the book could be used as a text for a course in catalysis, it would be necessary to provide supplemen tary materials to provide background on the different techniques which are discussed. As with any book on topics for which research is ongoing, one should not consider any of the chapters as being the final word. However, each section does provide a good beginning for the interested reader. There is clearly a need for a book of this type. D 87

PAGE 34

[eJ N =I classroom FLOW SHEET IS PROCESS LANGUAGE MANFRED FEHR Federal University at Uberlandia 38400 Uberlandia MG Brazil T E XTBOOKS WRI'ITEN FOR chemical engineering core courses do not generally provide links be tween the theory presented and industrial practice of process engineering. Many surprises are waiting for students in the plant environment. We have found that this situation can be considerably improved by incorporating flow sheet conception into the lecture program, mainly in unit operations and design courses. The flow sheet forces the student to consider a particular piece of equipment in its true industrial context where it ties in with utilities and basic control strategy. The method not only motivates students to participate creatively, but it also generates questions on process engineering logic that are a welcome en richment of class activity. This paper will illustrate with examples how students can be induced to ask these questions and to find answers to them. FIRST HEAT EXCHANGER EXAMPLE To illustrate the calculations for heat transfer to jacketed tank reactors, we went through an example solved in a well-known textbook [1]. Knowing the heat Manfred Fehr received his BS (1967) from the Universite Laval, his MS (1969) from the University of Alberta and his PhD (1977) from Universite Laval. He is a registered professional engineer in Alberta Quebec, Sao Paulo and Minas Gerais After short industrial e x periences in Canada and Germany, he lectured at various South American uni versities and is now engineering consultant and professor at the Fed eral University, Uberlandia, Brazil. His research interests are in the area of local energy sources 88 220 F team u palg FIGURE 1. Steam supply to reactor jacket. Above: book approach. Below : flow sheet approach. load and the available surface, the steam temperature required in the jacket is found to be 220F. The sub ject matter of the book goes this far, but thinking that this was not a real challenge to my students, I went on to speak some process language: our plant has two steam headers, 50 psig and 150 psig, and a condensate return header at 20 psig. I asked the students to pre pare a flow sheet with enough details to show exactly how steam flows from which of the supply headers through the jacket and back into the return header. The jacketed reactor then acquires its true identity as part of the processing unit. This is process language. Some quite common "nonsense situations" were sub mitted from which a lot could be learned about the logic of process engineering. Can you imagine a pump at the jacket exit that simply pulls the steam through from the supply header? Figure 1 shows a workable solution. It also depicts the difference between a book approach and a flow sheet approach to the problem. t, Copyright ChE Diviswn ASEE 19 88 CHEMICAL ENGINEERING EDUCATION

PAGE 35

SECOND HEAT EXCHANGER EXAMPLE When studying the thermal design of shell and tube exchangers, we went through an example solved in another well-known textbook [2] which states that "atmospheric pressure steam condenses on the shell side." Nobody objected to this as long as we were busy :figuring the number of tubes and passes. The surprise came when I again asked for the flow sheet. Only then did it become clear that book language is not process language Even the simplest steam trap requires a pressure drop to function, which means that in addition to the equipment shown in Figure 1, it will be necessary to pull a vacuum on the condensate collection tank in order to make the steam flow at all. What is more, this vacuum has to be monitored if a pressure of 1.00 atm. abs. is to prevail in the shell. The flow sheet question helped integrate the ex changer into the processing environment, and it be came evident that atmospheric pressure was a poor choice for the shell. EVAPORATOR EXAMPLE The typical textbook evaporator calculation [3] will state the presure in the last effect to have a certain value, such as 100 mm Hg absolute or 26 in Hg vac uum. All calculations are based on this number, al though no effort is made to explain its source. In addi tion, our school is located 3500 feet above sea level where 26 in Hg is considered perfect vacuum. This is another excellent situation to be explored with pro cess language. I got surprised looks when I asked, ''Would you please supply a flow sheet?" and, ''What would you do if some day you were asked to operate at 23 in Hg vacuum?" The problem statement in the book did not require that the student think this far ahead. Many misconceptions about vacuum equipment can be dispelled by simple confrontation of sense and nonsense on the flow sheet. Can you imagine a pump that sucks out all the vapor produced in the last effect? The surprised looks disappeared as the answers slowly took shape. A workable solution is presented in Figure 2. Again, the figure compares book approach to flow sheet approach to show the enrichment ob tained through process language. DISTILLATION PRESSURE EXAMPLE An uncommon situation appeared in the study of the conversion of fermented beer into 96 volume % ethanol in two distillation towers. The equilibrium conditions are such that no reflux is required in the first column. The overhead vapor containing 24 mol % SPRING 1988 The method not only motivates students to participate creatively, but it also generates questions on process engineering logic that are a welcome enrichment of class activity. This paper will illustrate ... how students can be induced to ask these questions and to find answers ... ,---II Ill H 'f'NIUIM FIGURE 2. Condensate collection from evaporator. Above: book approach. Below: flow sheet approach. ethanol is fed directly to the appropriate plate of the second column. On paper, the two columns looked like any column depicted in textbooks, with the number of plates determined by the method of McCabe-Thiele and the feed plates chosen accordingly. Mass and heat balances closed with the correct amount of open steam at the bottoms and the correct amount of cooling water in the second column condenser. There was but one little problem: At the start on deciding the operating pressure, there was no reason for it to be much different from atmospheric. A distilling com pany was consulted and the information obtained that the first column runs at 3 psig, which confirmed our reasoning Consequently, the design proceeded with this pressure. When the time came to provide a flow sheet, there was no variable available for manipula tion to control the first column pressure. Textbooks were consulted but no answer was found. On combin ing the flow sheets of the two towers and after a brain storming session the fact slowly emerged that the first column pressure in this scheme is not an inde89

PAGE 36

pendent variable as it is in standard reflux columns discussed in the books. The second column pressure has to be controlled at the condensate drum, and the first column pressure would then find its own equilib rium value, much like it does in a double effect evaporator. Had the problem been terminated at the textbook stage, this "discovery" would have been lost. CONCLUSIONS Flow sheet is process language. Process language is exciting. Process language transforms book exam ples into process plant examples. Process language resorts to the use of utilities to vitalize unit operations teaching. Process language is a stimulating teaching tool. LITERATURE CITED 1. Kern, D. Q., Process Heat Transfer, McGraw-Hill (1950). Example 20.1 2. Foust et al., Principles of Unit Operations, 2nd Edition, John Wiley & Sons (1980). Example 15.1 3. McCabe, W. L., J. C. Smith, P Harriott, Unit Operations of Chemical Engineering, 4th Edition, McGraw-Hill (1985). Example 16.1 [ejij I book reviews DISTILLATION TRAY FUNDAMENTALS by M. J. Lockett Cambridge University Press, Cambridge and New York, 1986. 224 + xxiii pp. $54.50 Reviewed by James R. Fair The University of Texas at Austin The distillation column continues to be the princi pal separation device for the chemical and petroleum processing industries. For many years it was charac terized as a vertical, cylindrical column containing plates or trays upon which rising vapor and descend ing liquid were brought into intimate contact, for pur poses of effective mass transfer. In recent years de signers of distillation columns have shifted some at tention to the use of structured (as opposed to ran dom) packings as vehicles for effecting intimate phase contacting. However, such packings are considerably more expensive than trays, and their cost is often jus tified only when their lower pressure drop carries an economic advantage, as in some vacuum distillations. Thus, the tray column remains as the standard and basic device for vapor-liquid contacting. Despite the title, this book deals not only with 90 trays for distillation services but also covers applica tions in absorption and stripping. It covers all of the important aspects of tray design, those of a more hy drodynamic nature as well as those relating directly to the mass transfer propensity of the two-phase mix ture on the tray. Considerable space is devoted to the characterization of this mixture: foam, emulsion, froth, spray, and so on. The overall coverage is quite complete, with no detail of design left unaddressed. Such important topics as phase flow distribution, ca pacity limits, pressure drop and interphase mass transport are dealt with on quantitative bases. A com mendable effort has been made by the author to con sider all historical approaches (mostly empirical) that deal with the various design parameters. The litera ture coverage is near exhaustive, and the reader will not find elsewhere as complete a bibliography on the distillation tray as is provided here. For each design consideration, a method with some fundamental and mechanistic support is provided-and for practition ers of distillation system design this is a welcome ad vance from the art and empiricism that have often prevailed. There are some limitations to the treatment that should be mentioned. First, the author has not always found it possible to make a forthright recommendation when several alternate models or procedures are available for a particular design step. The reader must then make his own choice. Also, despite the title, all tray-type contacting devices are not considered. There is very little on valve trays, essentially nothing on bubble-cap trays, and complete silence on dualflow trays (those without downcomers). Emphasis is clearly on crossflow sieve trays, but this is not all bad. While there are still many bubble-cap tray columns in operation, very few new ones have been designed dur ing the last few decades. The valve-tray is really a proprietary contactor, with design often left up to the proprietor. The dualflow tray is a rather specialized device (and tricky to design), used mostly for fouling services. On the other hand, the sieve tray is an effi cient and relatively inexpensive non-proprietary de vice that has been the object of many basic studies, and its simple geometry (in effect, one or more sheets of perforated metal, joined to a downcomer for hand ling liquid passage) makes it reasonably amenable to fundamental modeling. Still, the title might have read "Distillation Sieve Tray Fundamentals." The book might have been improved by the inclu sion of some worked-out design examples, some ad vice on laboratory or pilot plant scaleup procedures, and an author index. Still, the development of ra tional, fundamental-based approaches to the handling CHEMICAL ENGINEERING EDUCATION

PAGE 37

of complex two-phase mixtures, as are found on trays, is refreshing and encouraging. The author has an ex tensive background in the research, testing and mod eling of distillation devices, and his authoritative text reads very well There is no equal to the book pre sently on the market. Anyone concerned with the de sign or analysis of distillation, absorption or stripping columns of the tray type will want to take advantage of the modern approaches presented in this book. D GAS SEPARATION BY ADSORPTION PROCESSES by Ralph T. Yang Butterworth Publishers, 80 Montvale Avenue, Stoneham, MA 02180; 352 pages, $52.95 (1986) Reviewed by D. M. Ruthven University of New Brunswick The importance of adsorption as a separation pro cess in the chemical and petroleum industries has in creased dramatically in recent years, but the subject is still not covered in any significant way in most chemical engineering curricula. There have been three recent books on the subject: Principles of Adsorption and Adsorption Processes, by this reviewer, pub lished by Wiley in 1984; Large Scale Adsorption and Chromatography, by P. C. Wankat, published by CRC Press in 1986; and the present volume by Ralph Yang, published by Butterworth Publishers in 1987. None of these is really a textbook in the formal sense, but any of them could be used as the basis for a graduate level (or possibly a final year elective) course on the subject. The coverage of the present volume is broadly similar to that of Principles of Adsorption and Ad sorption Processes, and there is considerable overlap, which is probably inevitable since many of the source references are common. The emphasis is, however, different-reflecting the different areas of interest and expertise of the authors. The book provides a coherent and comprehensive account of the subject, including the basic physico-chemical principles as well as process technology. Although the title is Gas Sep aration by Adsorption Processes (and this is indeed the main focus), liquid phase separation processes such as the "Sorbex Process" and parametric pumping are also covered briefly. As with its predecessors, the emphasis is on fundamentals rather than on technolog ical details, and the level of background knowledge which is assumed is also similar. The book is di vi d ed into eight chapters: 1. IntroSPRING 1988 duction; 2. Adsorbents and Adsorption Isotherms; 3. Equilibrium Adsorption of Gas Mixtures; 4. Rate Pro cesses in Adsorbers; 5. Adsorber Dynamics--Bed Profiles and Breakthrough Curves; 6. Cyclic Gas Sep aration Processes; 7. Pressure Swing Adsorption Principles and Processes; 8. Pressure Swing Adsorp tion-Models and Experiments. It is not a book for the undergraduate, but it should be easily understood by graduate students and those with some experience in research and develop ment. Since most adsorption processes operate under transient conditions, some familiarity with partial dif ferential equations is needed to follow the sections dealing with column dynamics and process modeling (Chapters 5 and 8). I found Chapter 3, which contains a good review of the various approaches to the corre lation and prediction of multicomponent adsorption equilibria, and Chapters 7 and 8, which provide an authoritative summary of PSA technology and model ing, to be most useful. There is no discussion of membrane separation processes which compete directly with pressure swing adsorption in a number of applications. While mem brane separations may not be included within the nar rower definition of adsorption processes, some such discussion would have been useful to allow the reader to assess the relative merits of either approach, par ticularly in view of the publisher's claim that this is a "complete treatise covering all aspects of adsorption processes . Inevitably in a book which covers such a wide range of subjects, one can expect controversy over the treatment of certain topics. For example, in the discussion of surface diffusion and intracrystalline dif fusion in zeolites (pp. 113-121) it should probably have been pointed out that the kinetic treatment (which is emphasized) and the quasi-thermodynamic treatment (which is criticized) are not necessarily in conflict, but merely represent different ways of looking at the same phenomenon. The advantage of the quasi-ther modynamic treatment is that it allows meaningful transport co-efficients to be derived without know ledge of the detailed diffusion mechanism. This may not be obvious to the casual reader. Such criticisms are, however, minor, and any lack of balance is more than offset by the advantages in the presentation of coherent perspective. Taken as a whole, the book presents a concise and readable summary of the voluminous literature of the subject. It will no doubt become required reading for those wo r king in this area, both in universities and in industry. At U S $52. 95 it is (just abou t) within the affordable price range for individuals. D 91

PAGE 38

[j;jI lecture THE MYSTIQUE OF ENTROPY B.G.KYLE Kansas State University Manhattan, KS 66506 S TUDENTS OF thermodynamics soon learn to ap preciate the utility of entropy in making various calculations involving process heat and work effects. These are direct applications found in engineering and physical science. However, in addition to these quan titative applications, one finds qualitative and metaphoric uses of entropy in a wide diversity of fields. The extent to which the concept of entropy has suffused contemporary thought is well illustrated by Lord C. P. Snow's assertion that any definition of cul ture should include a technical component and that an understanding of the second law of thermodynamics is the cultural equivalent of a familiarity with the works of Shakespeare [1]. !goring incidental uses, a few examples will be pre sented here in which the concept of entropy is central to the development of a theme or is thought to provide insight. The intent is to illustrate the pervasiveness of the fascination evoked by the entropy concept-the entropy mystique. COSMOLOGY Of the many formulations of the laws of ther modynamics, the boldest and most provocative was advanced in 1865 by Rudolf Clausius: The energy of the universe remains constant. The entropy of the universe tends to a maximum. Here Clausius has taken concepts arising from limited, earth-bound experience and with a great leap of imagination has vested them with cosmic signifi cance. While both statements are daring, the second law statement has sparked the most interest and con... a few examples will be presented in which the concept of entropy is central to the development of a theme or is thought to provide insight. The intent is to illustrate the pervasiveness of the fascination evoked by the entropy conceptthe entropy mystique. 92 Benjamin G Kyle is profe sso r of chemical engineering at Kansas State University and is enjoying his thirtieth year of teaching He holds a BS from the Georgia Institute Technology and a PhD from the Univer sity of Florida He has not outgrown an early fascination with ther modynamics and i s interested in practically all aspe cts of the subject He is the author of a recently published (Prentice-Hall) thermodynamics textbook troversy because it leads to speculations regarding the birth and death of the universe. Birth and Death of the Universe By definition, the universe must be a closed sys tem, and the uncritical extrapolation of our terrestrial experience would suggest that its entropy is increas ing toward a maximum. Because an increase in en tropy is associated with a decreased ability to perform work the second law implies that the univer se will ultimately reach a dead state referred to as thermal death. This dead state is an equilibrium state in which all thermodynamic potentials have been leveled and processes yielding work are no longer possible. If one accepts Clausius' statement of the second law with its implication of thermal death, then by the following simple argument one can show that the uni verse had a beginning. A maximum is characterized by a first derivative of zero. However, the derivative, or rate of change, is finite until the maximum is reached. Thus, if the entropy of the universe is tend ing to a maximum, it is doing so at a finite rate and will reach its goal in a finite time. A universe of infi
PAGE 39

nite age would have already reached its equilibrium state of maximum entropy, and since this is not the case, the universe must have a finite age. Thus, the universe had a beginning and we have a position that is at least congenial to the inference of creation and a creator. Undoubtedly, the inclination to draw this in ference is strengthened by a scientifically more ac ceptable theory originated by Lemaitre in 1927 and more recently popularized [2] as the Big Bang" theory. The widely published English prelate, William Inge, eagerly accepted the proof of a creator as im plied by the "law of entropy" [3]. Inge used the pros pect of the thermal death of the universe to argue that God the Creator could not be merely a pantheistic god found only in nature because such a god would be under an inexorable death sentence. Such would not be the fate of the transcendent god of Christianity. It is not Christianity but modem pantheism and the myth of unending progress which are undermined by the de gradation of energy. Most notable among those espousing the creation view was Pope Pius XII, who stated that Clausius' law of entropy provides "eloquent evidence of the existence of a Necessary Being." [4] An argument against the inference of creation from the laws of thermodynamics was advanced by the Russian physicist, I. P. Bazarov [5], based on the dialectic materialism of Engels. Identifying flaws in the thermal death argument for the existence of a creator, Bazarov points to the unfounded assumption that the laws of thermodynamics apply to the entire universe and to Engels' argument that the creation implied by the second law would be in violation of the first law. In Engels' words, those advocating the pro creation view saw the universe as winding down and thus assumed that an initial winding up had been pro vided by "a stimulus from without." However the process of winding up imparted energy to the universe and hence the total energy of the universe has not always been constant as required by Clausius' first law statement. It now appears that Engels' argument of a first law violation may be vitiated by recent calculations that indicate the total of all energy in the universe is zero [6]. This is possible because energy of motion and the energy equivalent of mass are positive while energy of gravitational or electromagnetic attraction is negative. A zero energy sum would imply that cre ation could have occurred without an energy input and brings to mind the theological doctrine of creation e x nihilo SPRING 1988 The theory of relativity shows that time and space are part of the structure of the universe and not ... an empty stage on which physical events are enacted. However, within this purely physical picture there remains the basic question of the direction of time. Bazarov's criticism regarding the unwarranted ex tension of the laws of thermodynamics to the entire universe seems well taken. This is especially so con sidering that on a cosmic scale the predominant energy effects are associated with gravitation and radiation, types of energy which are usually neglected when applying thermodynamics in its usual terrestrial context. In spite of this obvious incongruity, the ques tion is still unresolved for today one finds astrophysi cists concerned with entropy changes of various cos mic processes in which quantum and relativistic ef fects predominate [6]. Time Of all the basic quantities in the physical world, time is the most elusive. This is because we not only experience it as an abstraction useful in physical sci ence, but also in a psychological sense. Because the latter lies beyond the realm of physics, any strictly physical description of time can never be entirely satisfactory. The theory of relativity shows that time and space are part of the structure of the universe and not, as pictured by classical physics, an empty stage on which physical events are enacted. However, within this purely physical picture there remains the basic ques tion of the direction of time. The equations of classical mechanics quantum mechanics, and relativity theory are symmetric in time and do not preclude the rever sibility of processes. On the other hand, we know that all naturally occurring processes are irreversible and hence travel only in what we have designated as the forward direction of time. For this reason the second law of thermodynamics is said to define the "arrow of time." Time increases in the direction of increasing entropy. The problem of reconciling the reversibility of the microscopic world of colliding molecules obeying the laws of classical mechanics with the irreversible be havior of the macroscopic world was first undertaken by Boltzmann. This search for the origins of irreversi bility provides an interesting chapter in the annals of science [7] and has led to the development of statistical mechanics, an area of science which has proved quite useful in the calculation of thermodynamic and trans port properties. However, despite this practical sue93

PAGE 40

cess, the origins of irreversibility have yet to be deter mined to the satisfaction of scientists of a philosophical bent [8]. LIFE AND EVOLUTION At present there is disagreement among scientists as to whether living systems can be completely de scribed by the laws that apply to inanimate matter or whether additional, but as yet undiscovered, laws ap plying only to living matter are also needed. However, the various identifiable physical and chemical proces ses occurring within living systems require no addi tional laws for their explication and therefore conform to the laws of thermodynamics. Additionally, it is known that the laws of thermodynamics are obeyed by each living system as it interacts with its environ ment. Our present societal obsession with diets and calorie counting serves as a monotonous reminder of our bondage to the first law, but the dictates of the second law are not so obvious and, in fact, may seem counter-intuitive. The conceptual difficulty appears when we regard life as a striving to maintain order and the second law as a principle of degradation. This apparent conflict is due to our failure to recognize liv ing systems as open systems. In an open system the combined entropy of the system and surroundings must increase. Thus, a living system can maintain it self or grow, and thereby decrease its entropy, if the surroundings undergo the appropriate increase in en tropy. This leads directly to the statement that life is sustained by the conversion of low-entropy resources into high-entropy wastes. The ultimate origin of low entropy resources is the sun which drives the photo synthetic mechanism in plants whereby solar energy is converted into chemical energy and stored as carbo hydrates. Despite reconciliation of the life force with the law of entropy, there is still a persistent undercurrent of nagging doubt when the origin and evolution of life are considered. These activities imply purpose, a goal that seems totally opposed by the entropy principle. This contradiction was recogized by Teilhard de Char din in formulating his grand scheme of evolution to wards an Omega Point of pure consciousness and ecstatic union with God [9]. Because of the law of entropy with its opposing tendencies and prospect of thermal death, Teilhard postulated two classes of energy: tangential and radial. Tangential energy is simply the energy identified by physical science and subject to the laws of thermodynamics. On the other hand, radial energy is of a psychic or spiritual nature and is not subject to the laws of thermodynamics. The evolution of life toward the Omega Point is said to be 94 driven by radial energy and thus is free of second-law restrictions. The relationship of the two classes of energy is tenuous and is best illustrated by Teilhard's words: To think, we must eat. But what a variety of thoughts we get out of one slice of bread! Like the letters of the alphabet, which can equally well be assembled into nonsense as into the most beautiful poem, the same calories seem as indif ferent as they are necessary to the spiritual values they nourish. Teilhard was a Jesuit priest as well as a scientist, but he wanted his work to be judged on the basis of its scientific merit His Phenomenon of Man is a monumental attempt to meld science and metaphysics to provide an answer to the question of meaning. De spite his heroic effort to fuse faith and reason, science, because it recognizes only reason, could do nothing but render a negative verdict. Nevertheless, his work has wide appeal today, some thirty years after his death. Perhaps this is because the soul of a poet and mystic shows through. After all, who but a poet or mystic would dare address such questions. SOCIAL SCIENCE Henry Adams, the noted American historian and man of letters, was the first to view the course of civilization (i.e., history) from a second-law perspec tive [10] In the opening years of the twentieth cen tury he expounded a theory of history in which the Social Energy of civilization is continually dissipated. According to Adams, the second law of ther modynamics required that, ... the higher powers of energy tended always to fall lower, and that this pro cess had no known limit. In the category of energy Adams included the Vital Energy of an individual and the Social Energy of society and stated that, "The law of entropy imposes a servitude on all energies, includ ing the mental." All this naturally resulted in a pes simistic prognosis for civilization where ... the ash heap was constantly increasing in size." According to one of its most famous practitioners Claude Levi-Strauss the field of anthropology could appropriately be called entropology." Levi-Strauss [11] sees man and his civilizations as "instruments in tended to create inertia, at a rate and in a proportion infinitely higher than the amount of organization they involve." He further states Thus it is that civilization taken as a whole, can be de scribed as an extraordinary complex mechanism, which we might be tempted to see as offering an opportunity of sur vival for the human world, ifits function were not to produce CHEMICAL ENGINEERING EDUCATION

PAGE 41

what physicists call entropy, that is inertia Every verbal ex change, every line printed, establishes communication be tween people, thus creating an evenness of level, where be fore there was an information gap and consequently a greater degree of organization. We find familar the concept that the net result of human life is an increase in entropy, but the idea of a leveling of social organization through communication is a bit unexpected and would appear to be a metaphoric extension of the thermal death concept. The second law of thermodynamics has been shown to provide a realistic perspective for economics [12]. According to Nicholas Georgescu-Roegen, the conven tional view of economic process as circular and time less ignores the increase in entropy accompanying every human endeavor. Georgescu-Roegen sees economic activity as turning low-entropy inputs into high-entropy outputs. All species depend on the sun as their ultimate source of low entropy except man, who has learned also to exploit the terrestrial stores of low entropy such as minerals and fossil fuels. Life feeds on low entropy and so does economic life. Objects of economic value such as fruit, cloth, china, lum ber, and copper, are highly ordered, low-entropy structures. For low entropy is the taproot of economic scarcity. He observes that economic development as presently practiced is based on rapacious consumption of our terrestrial dowry of low-entropy materials and con cludes that this can not be indefinitely sustained. Using the entropy-conscious economic views of Georgescu-Roegen, Herman Daly [13] argues persua sively against the fallacy of perpetual growth and even suggests that growth itself may be illusory when mea sured by the GNP. This closely watched economic in dicator includes, besides the value of goods produced and services rendered all costs associated with pro duction, including pollution control costs. As the richer and more accessible resources are preferen tially consumed, the remaining lower quality re sources require higher production costs and thus a rise in GNP may reflect increased cost of production rather than increased level of production. This is the entropic factor largely ignored by conventional economics. As an alternative to growth, Daly has out lined in some detail the workings of a steady-state economy. The work of Georgescu-Roegen has also inspired Jeremy Rifkin's Entropy: A New World View [14] in which the entropy law is credited with singlehandedly undermining the Newtonian-Cartesian mechanistic world view. Rifkin believes that a new world view based on the entropy law would expose the fallacy in SPRING 1988 our present obsession with growth and would allow us to meet the future with awareness and acceptance of nature's constraints. This change in outlook would af fect practically every area of human endeavor ranging from technology to religion and could move us toward the long-sought but ever-distant utopian dream. There is no doubt that Rifkin has overstated his case in a single-minded and over-simplified manner, but exaggeration is a forensic device which often becomes acceptable when used in passion for a worthy cause. Because the inculcation of an entropy-conscious world view is considered by many to be such a cause, perhaps Rifkin should not be judged too harshly for his excesses. LITERATURE AND ART Metaphoric entropic themes abound in modern fic tion [15] as well as in serious science fiction [16]. Most applications are implicit, although several writers explicitly use the term entropy in their work. These writers are predominately American and include John Barth, Donald Barthelme, Saul Bellow, Stanley Elkin, Norman Mailer, Walker Percy, Thomas Pynchon, and John Updike. As expected of fertile imaginations, one finds many variations of the entropic theme whether used explicitly or implicitly The theme has been applied to either individuals or entire societies, some times within the context of a closed system, with ulti mate states as extreme as chaos and stagnation. Norbert Wiener's famous book on cybernetics [17] first appeared in 1950 and was undoutedly instrumen tal in popularizing the application of the entropy con cept to the many aspects of the human condition as found in modern literature. Moreover, Wiener's iden tification of entropy as a measure of the information content of messages stimulated writers to examine the very act of writing [18]. In Wiener's words Messages are themselves a form of pattern and organiza tion. Indeed, it is possible to treat sets of messages as having an entropy like sets of states of the external world. Just as entropy is a measure of disorganization, the information carried by a set of messages is a measure of organization. In fact, it is possible to interpret the information carried by a message as essentially the negative of its entropy, and the negative logarithm of its probability. That is, the more prob able the message, the less information it gives. Cliches, for example, are less illuminating than great poems. The idea behind this passage is easily grasped when applied to the transmission of factual data (e.g., a seven-digit telephone number) or a coded message (e.g., one ifby land, two ifby sea) but quickly becomes fuzzy when considering the transmission of subtle or 95

PAGE 42

One is left wondering why the entropy concept, which offers so little insight, has been so widely employed outside its narrow scientific context and why any serious artist or writer would even consider the application of scientific principles to the very act of creating. There does indeed appear to be an entropy mystique ... abstract ideas or the more vital aspects of communica tion inherent in the sender and receiver. This becomes apparent when the idea is pushed to its limit. The writer, in an attempt to counter the entropic effect of banality, looks for unexpected or improbable ways of using words and thus runs the risk of verbal chaos. The dilemma is articulated by Lewicki [12]: In order to avoid entropy, a writer must therefore walk a narrow path between the danger of producing probable mes sages of low informational value (such as, for example, new versions of old themes, written in a conventional manner), and the risk of turning out incomprehensible bodies of words that would seem disorderly to the reader. Rudolph Arnheim [19] has recognized the con tradictions in information theory as applied to art: Here order is described as the carrier of information, be cause information is defined as the opposite of entropy, and entropy is a measure of disorder. To transmit information means to induce order. This sounds reasonable enough. Next, since entropy grows with the probability of a state of affairs, information does the opposite: it increases with its improbability. The Jess likely an event is to happen, the more information does its occurrence represent. This again seems reasonable. Now what sort of sequence of events will be least predictable and therefore carry a maximum of information. Obviously a totally disordered one, since when we are con fronted with chaos we can never predict what will happen next. The conclusion is that total disorder provides a maxi mum of information; and since information is measured by order, a maximum of order is conveyed by a maximum of dis order. Arnheim attributes the contradiction to a failure to distinguish order from structure: Any predictable regularity is termed redundant by the in formation theorist because he is committed to economy: every statement must be limited to what is needed. He shares this commitment with scientists and artists; its meaning, however, depends on whether one chops up patterns into elementary bits or whether one treats them as structures. A straight line reduced to a sequence of dots for the purpose of piecemeal analysis or transmission can be highly redundant; in the drawing of a geometrician, engineer, or artist it is not. In explaining the creative process underlying a work of art, Arnheim sees an interplay between the anabolic tendency which establishes the structural theme and the entropic principle which is manifested both as a catabolic destruction of patterns, or disor96 dering, and a simplification or ordering due to tension reduction. Here we see the entropy principle invoked metaphorically and identified simultaneously with two opposing tendencies: ordering and disordering. COMMENTARY Of all the properties of matter, entropy is most difficult to conceptualize and seems the most con trived. It cannot be evaluated directly from an experi mental measurement, but must be calculated from a somewhat arbitrary computational path and thus its existence and evaluation are closely tied to human ac tivity. It is tainted with a human scent. Perhaps this is a reason this most anthropomorphic of all the con cepts of science has fascinated the human mind. Despite its mystique, entropy actually provides very little insight into the mysteries of nature. This can be appreciated by recalling the familiar textbook derivations. A Carnot cycle is used to define entropy and demonstrate that it is a state property. Next, an isolated system undergoing a spontaneous process is considered and a simple argument shows that an in crease in entropy results. Ramifications of this result are extremely useful for the detailed calculations made by engineers and scientists, but in what way has it improved our general understanding of nature? Are not the following statements equivalent and equally enlightening? A. Every spontaneous process tends toward a condition of deterioration or stagnation. B. Entropy is the measure of deterioration or stagnation. Every spontaneous process results in an increase in en tropy. Statement A is a direct statement of experience. Statement B is expressed in the language of an empir ically based science and is therefore merely an indirect but more formalized statement of experience. The idea of things running down, which can now be proclaimed as a decree of science, has been found to be a dominant theme in ancient myths [20]. This condition also has been lamented by Hesiod, the Greek poet of the 8th century B. C., who told of the five descending ages of Man beginning with a pristine age of gold and ending in his own worldly age of iron. Also, somewhat later, we find St. Paul [21] referring to a coming time of glory when "creation itself will be set free from its bondage to decay." These may be CHEMICAL ENGINEERING EDUCATION

PAGE 43

regarded as generalizations of statement A and ap pear more naive than the B-based concept of thermal death. Yet, even though we favor statement B today, let us not be deceived into thinking it more insightful than statement A. With regard to fixing the direction of time, it is also instructive to consider two alternative state ments. A. In the usual sense of before and after, spontaneous pro cesses proceed in the forward direction of time. B. All spontaneous processes are accompanied by an in crease in entropy. The forward direction of time is the direction in which the entropy increases. While the practical person would give no advan tage to statement B, many philosophers prefer some thing similar to it because it appears to have elimi nated some of the subjective element that clings to statement A. Their object is to define time without reference to the human mind [22]. However, in this respect an obvious problem arises in the actual deter mination of entropy. This requires a human mind to devise a reversible path between the initial and final states which is essential to the calculation of the en tropy change There is no entropy meter and the in voking of entropy, which is itself a construct of the human mind, does little to remove the human scent from the concept of time. To place the question of time in perspective we should recognize that it is a construct of the conscious mind. The subconscious mind, the Freudian id, has no awareness of time-a fact known to science and easily verified personally by recalling the particulars of our dreams. What we have done is to construct from the regular rhythms of the universe (e.g., the motion of the earth about its axis or about the sun) a lifeless, linear time scale upon which we can place in monotonic order the events of our physical world This seems so natural that it comes as something of a shock to learn that it is possible to have a valid and effective world view that does not include the concept of time. An outstanding example of this is the language of the Hopi indians of the American southwest which con tains no reference to "time" either explicitly or im plicitly. This language is capable of accounting for and describing correctly, in a pragmatic or operational sense, all observable phenomena in the universe with out the mental construct of time [23]. The avowed aim of science is to establish the basis of physical reality. However, two different interpreta tions exist: external and internal. The external ver sion is the older, traditional view in which the human mind objectively probes and observes nature and thereby discovers natural laws. Recent developments SPRING 1988 in physics, however, have cast doubt on the concept of an objective observer independent of the system observed [24], and this has led to the view that science merely creates a set of interlocking laws which pro vides a consistent description of nature. This descrip tion usually involves the use of abstractions and men tal constructs far removed from our everday experi ence and may thus be termed a view of physical reality internal to the mind [25]. The previous discussion of time and entropy is obviously an argument for the internal view. One is now left wondering why the entropy con cept, which offers so little insight, has been so widely employed outside its narrow scientific context and why any serious artist or writer would even consider the application of scientific principles to the very act of creating. There does indeed appear to be an entropy mystique, although the reasons for it will not be explored. Perhaps the following verse is appropriate. ENTROPY For thermodynamics we feel a sense of awe And marvel at the power of those early minds Who showed that mundane matters lead to natural law And boldly stated that the universe unwinds. The second law is never denied its due In science, belles-lettres, and philosophy. It pales the onward-and-upward view, For no one is consoled by entropy. While nothing temporal eludes its iron rule And most would take decay's decree as true, The sage's bane can be the builder's tool As entropy shows the best that we can do. This useful concept prompts a primal groan: A dread we've only named but always known. REFERENCES 1. Snow C. P., The Two Cultures : And a Second Look, The New American Library New York, 1964. 2. See, for example, Weinberg, S. The Fir s t Thre e M i nutes, Fontana, 1978 3 Inge W. R. God and the Astronomers, Longmans Green & Co London, 1934. 4. Address to the Pontificial Academy of Sciences, Rome 1951, reprinted in English in Bull e tin of the Atomic Scientists, 8, 143 (1952). 5. Bazarov, I. P., Thermodynamics, pg 74 Macmillan New York, 1964. 6. Davies, Paul God and the N e w Physi c s, Ch. 3, Simon and Schuster Inc ., New York, 1983. 7. Brush, S., The K i nd of Motion W e Call Heat North Holland Pub. Co Amsterdam 1976. 8 For a review of this problem, see Hollinger H. B. and M. J. Continu e d on page 10 2 97

PAGE 44

[eJ ;j ?I classroom DISCRETE-EVENT SIMULATION IN CHEMICAL ENGINEERING DANIEL J. SCHULTHEISZ a n d JUDE T. SOMMERFELD G e orgia Institut e of T e chnol o gy Atlanta, GA 3 0 3 3 20100 R E C ENT YEARS H A VE witnessed a rapid and dr a matic change in the nature of the chemical pro cess industries in the developed countries of the world. Specifically, there ha s been an intense revival of commercial interest in batch chemical processes, such as those employed in the manufacture of fine and specialty chemicals, at the expense of traditional con tinuous steady state processes for the manufacture of commodity chemicals. One large British chemical com pany reports that sp e cialty chemicals manufactured by batch processing contributed over 30 % to their total profits in 1983 as opposed to 18 % in 1977 [1]. Certainly, one of the primary driving fo r ces for this change has been the recent commissioning of many world-scale commodit y chemicals plants in various de veloping countries. SIMULATION SYSTEMS Concomitant with these indus try c h a ng es signifi cant developments ha v e occurred in t he modeling and simulation of chemical processes. To be sure, usage (including academic) and development of c onventional steady-state process simulators continue at an active level. Thus, the FLOWTRAN s ystem [2] developed by the Monsanto Company w as made a v ailable to chemical engineering schools in 1973 and has been ex tensively employed for educational purposes e v er since [3]. Subsequently, newer stead ystate proce s s simulation systems such as PROCESS, ASPEN PLUS and DESIGN II became available to academic users. Discrete-event simulators were originally developed as numerical aids to solve complex queuing theory problems which were not amenable to analytical solution. Such problems occur routinely in the field of industrial engineering ... Co pyright C h E Di vision AS EE 19 88 98 Daniel J. Schultheisz receiv e d hi s bachelor s degree in chemical e ngineering from the Uni v ers i ty of P enns ylvonio ond is currently com pleting the requir e ments for his master s degree in chernicol engineer ing ot Ge o rgie Tech. F o r his thesis he is preparing on instructor s monu o l of e x er c ises i n c hemi ca l engi n e er ing using GPSS to be distri but e d and marketed b y the CACHE Corporation (L) Jude T Sommerfeld is p r ofessor and ass o ciate dire c tor of the School o f Chemi ca l Engineer i ng at Georgia Tech. He re c e i ved his BChE degree from the University o f Detroit and his MSE and PhD degrees also in chemical engineering from the University of M i chigan His 25 years of industrial and academic e x perience have been primar i ly in the area of computer-aided design and he has published o v er seventy articles in this and other areas (R) We ha v e al s o witnessed the development and ap pli c ation of va rious simulators for batch chemical pro cesses in recent years These developments have in cluded both discrete-e v ent and combined (discrete + continuous) systems, as employed in the industrial en gineering field. There is an unfortunate confusion in terminology here: the industrial engineering interpre tation of the term continuous' i s not the same as that associated with chemical engineering usage, namely, steady state operation. Rather, the industrial en gineering meaning of continuous should be construed by chemical engineers as dynamic or unsteady-state. The progenitor of discrete-event simulation sys tems is GPSS [ 4] which dates back to 1959 and is still used extensively in many manufacturing sectors. Be caus e of its easy use, availability reliability, and effi cient operation (integer arithmetic only in many ver sions), GPSS is a very effective tool if only discrete simulation capabilit y is required Other popular disCHEMICAL ENGINEERING EDUCATION

PAGE 45

TABLE 1 Recent Applications of GPSS to Chemical and Allied Processes APPLICATION [REF.] Two batch reactors in parallel followed by a batch still [7] DDT manufacture [8] Chocolate manufacturing [9] Sequence of batch di st illation columns [10) Large-scale poliomyelitis vaccine production [11) Choline chloride manufacture [12) Polyvinyl chloride (PVC) manufacture [13) Sequencing batch reactors (SBR) for wastewater treatment [14) crete-event simulation systems include SIMULA [5] (more prevalent in Europe) and SIMSCRIPT [6]. In general, however, there are not many published appli cations of discrete-event simulation systems to batch chemical processes. Morris [7], for example, has de scribed a very simple application of GPSS to a batch process comprised of two reactors in parallel followed by a still. Other recent applications of GPSS to chem ical and allied batch (or semi-continuous) processes are listed in Table 1. DISCRETE-EVENT SIMULATION Discrete-event simulators were originally de veloped as numerical aids to solve complex queuing theory problems which were not amenable to analytiTABLE 2 Terminology in Usage of the FLOWTRAN and GPSS Simulators FLOWTRAN (Stea dy-State GPSS Process (Discrete -Event ITEMS Simulator) Simulator) Precoded functional Blocks Blocks subroutines Arguments of functional Parameters Operands subroutines Items moving through Streams Transactions the model Numerical characteristics Properties ( e.g., Parameters of moving items temperature, composition) Output quantitie s from Retention vector Standard subroutines (ot her contents numerical than moving items) attributes SPRING 1988 Despite their considerable differences in origin and application, there are noteworthy similarities among the various types of simulators ... For example, FLOWTRAN and GPSS have a number of precoded functional subroutines. cal solution. Such problems occur routinely in the field of industrial engineering and typical example applica tions include machine shops, customer service sta tions, and transportation networks. Most discrete simulation systems have stochastic capabilities for the scheduling of time events. To sup port this function, most such systems also have one or more built-in random-number generators. Output from the latter is used to sample event times (or dura tions between time events) from various probability distributions. In GPSS, the only easy-to-use, built-in distribution is the uniform or rectangular distribution. Thus, for example, a service time can take the form, A B, where A represents the mean value and Bis the half-width, in appropriate time units, of the distri bution. SIMILARITIES IN SIMULATORS Despite their considerable differences in origin and application, there are noteworthy similarities among the various types of simulators described above. For example, FLOWTRAN and GPSS have a number of precoded functional subroutines (generally written in FORTRAN). In both FLOWTRAN and GPSS, these functional subroutines are known as blocks. There is a number of other similarities between these two systems, obscured only by the technical jar gon employed. In conventional steady-state chemical process simulators such as FLOWTRAN and PRO CESS, the items which move from one block to another in the model are known as streams. Each in dividual stream has a set of properties (composition, temperature, pressure) associated with it, which are typically modified as the stream passes through a block. In analogous fashion, the items which proceed from block to block in a GPSS model are known as transactions. Also in analogy with stream properties in a steady-state process simulator, GPSS transac tions have associated with them various parameters (such as priority level or lifetime in the model) which can be modified by the passage of the transaction through certain blocks. In a GPSS model of a batch chemical plant, for example, transactions could repre sent batches of material proceeding through the pro cess. Table 2 summarizes these similarities and ter minology for the FLOWTRAN and GPSS simulators. 99

PAGE 46

GPSS PROCESSOR There are about thirty-five different blocks in GPSS roughly the same number as in the FLOW TRAN system. A listing of these GPSS blocks is given in Table 3. It is common to construct block diagrams in the development of GPSS models. In contrast with FLOWTRAN where each block in such diagrams is represented more or less by a rectangle, each differ ent functional block in a GPSS block diagram has its own distinctive shape (see Schriber [4]) Some of the GPSS blocks listed in Table 3 are quite complicated and would typically be used only by more sophisticated analysts. There are others, however, which would be common to any GPSS model. Thus, GENERATE blocks are used to provide transactions to a model, much like a chemical engineer inputs feed streams to a FLOWTRAN model. Conversely, a transaction is removed from a GPSS model by a TER MINATE block. There is a block named SPLIT in both FLOW TRAN and GPSS, but there is one fundamental differ ence between the two. In the FLOWTRAN system, the sum of each extensive property over all of the output streams from the block equals that property for the incoming stream. In discrete-event simulation with GPSS, however, the SPLIT block really per forms a cloning operation. That is, one or more identi cal offspring transactions are created from the single parent transaction (which retains its existence) enter ing the block. GPSS OUTPUT As with the FLOWTRAN system which provides a summary table of the streams passing through the model and output results from each of the blocks in the model, GPSS automatically prints out a variety of output statistics at the conclusion of a simulation. These statistics pertain primarily to the various ADVANCE ASSEMBLE ASSIGN BUFFER DEPART ENTER FAVAIL FUNAVAIL GATE GATHER GENERATE 100 TABLE 3 Listing of GPSS Blocks LEAVE LINK LOGIC LOOP MARK MATCH MSAVEVALUE PREEMPT PRINT PRIORITY QUEUE RELEASE RETURN SAVEVALUE SEIZE SELECT SPLIT TABULATE TERMINATE TEST TRANSFER UNLINK facilities, queues, and storages in the mo~~l. Thus from an inspection of the facility output statistic~ from a GPSS simulation, an analyst might find that the average holding time per transaction for a given facility is considerably greater than_ ~he user supplied average service time for that facility. In _a chemical engineering application for example, this could indicate that a reactor, after finishing processing of a batch (transaction), often cannot discharge the batch because of an unavailable downstream facility. The latter might correspond to a storage tank ~hich is full or another processing unit (e.g still, centrifuge, dryer) which is engaged. The regular occurrence of such a situation would normally be accompanied by an average utilization (fraction of total time busy) ap proaching unity for the original upstream facility and would suggest the existence of some downstream bottleneck. The existence of similar bottleneck situa tions can also be deduced from the output statistics for GPSS storages. The productivity (number of batches produced) of the modelled process is, of course, related to the number of transactions passing through the GPSS model. EXAMPLE APPLICATION Let us consider a very simple application of GPSS to the modeling of a batch chemical process. This example is an adaptation of a problem (number 2.41.14) presented by Schriber [4]; the process flow diagram for this example is presented in Figure 1. CUSTOMER ORDERS -------IQ/-S TILL --0 0 0 -::::::::::: IQ/ ::::=; Q --IQ! PRODUCT ~E IQ/ S~,_~GE THREE IDENTICAL REACTORS FIGURE 1. Sketch of batch process for example problem. Thus, a small, single-product batch chemical plant has three identical reactors in parallel, followed by a single storage tank and a batch still. Customer orders (batches) to be filled (which begin with processing in the reactor) occur every 115 30 minutes, uniformly distributed. The reaction time in a given reactor is 335 60 minutes, and the distillation time in the still is 110 25 minutes, both times uniformly distributed. The holding capacity of the storage tank is exactly one batch. Hence, the storage tank must be empty for a given reactor to discharge its batch; if not, the reactor cannot begin processing a new batch until the storage CHEMICAL ENGINEERING EDUCATION

PAGE 47

GENERATE FIGURE 2. GPSS block diagram for example problem. tank becomes empty. The simulation is to be run for 100 batches. The model should have the capability to collect waiting line statistics for the queue im mediately upstream of the reactor. The GPSS block diagram for this example model is shown in Figure 2. Note the distinctive shapes for each of the blocks employed. The first executable block is the GENERATE block, which creates trans actions representing customer orders (batches). These transactions immediately queue up and attempt to capture an available reactor via the ENTER block. After capturing a reactor, a batch leaves the reactor queue through the DEPART REACQ block, and is processed in the ADVANCE 335,60 block. The batch must first be able to enter the storage tank (ENTER TANK block) before it releases its reactor in the LEA VE REACT block. The batch then attempts to capture the single still facility in the SEIZE STILL block. Having accomplished such, the batch leaves the storage tank, is processed in the still, releases the latter, and finally leaves the model through the TER MINATE block. Selected output statistics from this simulation are summarized in Table 4. From Table 4, one sees that the batch still was in use 91.1 % of the time, and the average holding (pro cessing) time per batch was 108 minutes. The average contents in the queue upstream of the reactors was 0.44 batch, and the average waiting time for all batches, including ones which experienced no waiting, in this queue was 50.5 minutes. The three reactors were in use 95.2% of the time, and the average holding SPRING 1988 time for a batch in a reactor was 329 minutes. Simi larly the storage tank (with a capacity of one batch) was full 41.4% of the time, and the average holding time therein was 48. 6 minutes. Although not pre sented in Table 4, the total simulation time, to com pletely process 100 batches, was 11,967 minutes. One can easily explore proposed modifications to this process. Thus, one more reactor could be added in an effort to increase productivity. One might find as a result, however, a significant increase in the aver age reactor holding time beyond the nominal average reaction time of 335 minutes. In this case, one could explore increasing the intermediate storage capacity (TANK) and/or improving the downstream distillation operation. SUMMARY This article has attempted to serve as a brief intro ductory tutorial on discrete-event simulation, with emphasis on chemical engineering applications. For some simple batch process applications, only discrete simulation capability is required. More complex appli cations would require usage of a combined (discrete plus dynamic) simulation system, but knowledge of the essential features of discrete-event simulation re mains useful background in such cases. REFERENCES 1. Preston M. L., and G. W. H. Frank, "A New Tool for Batch Process Engineers---ICl's BatchMASTER," Plant/Op e ra tions Progr., 4 217 (1985). 2. Seader, J. D., W. D. Seider, and A. C. Pauls FLOWTRAN TABLE 4 Selected GPSS Output from Example Simulation of a Batch Chemical Process Facility Statistics (batch still, STILL): Average utilization Average holding time per batch, min Queue Statistics (reactor queue, REACQ): Maximum contents Average contents Average waiting time (all batches), min Storage Statistics: Reactors (REACT): Average utilization Average holding time per batch, min Storage Tank (TANK): Average utilization Average holding time per batch, min 0.911 108 3 0.44 50.5 0.952 329 0.414 48.6 101

PAGE 48

Simulation-An Introduction, 2nd Ed., Ulrich's Bookstore, Ann Arbor, MI, 1977. 3. Clark, J. P., and J. T Sommerfeld, "Use of FLOWTRAN Simulation in Education," Chem. Eng. Ed., 10, No. 2, 90 (1976). 4. Schriber, T. J., Simulation Using GPSS, Wiley, New York, 1974. 5 Franta, W R., The Process View of Simulation, Elsevier North-Holland, New York, 1977. 6. Wyman, F. P., Simulation Modeling : A Guide to Using SIMSCRIPT, Wiley, New York, 1970. 7. Morris, R. C., "Simulating Batch Processes," Chem Eng 90, No. 10, 77 (1983). 8. Blaylock, C. R., C. 0. Morgan, and J. T Sommerfeld, "GPSS Simulation of DDT Manufacture," Proc. 14th IASTED Intl. Conj. on Appl. Sim. and Mod., Vancouver, June, 1986, p. 314. 9. Passariello, I., and J. T. Sommerfeld, "GPSS Simulation of Chocolate Manufacturing'' in Tools for the Simulation Prof es sion, Soc. for Computer Simulation, San Diego, 1987, p 1. 10. Barnette, D. T., andJ. T. Sommerfeld, "Discrete-Event Simu lation of a Sequence of Multi-Component Batch Distillation Columns," Comp. & Chem Eng., 11, 395 (1987). 11. Kenvin, J.C., and J. T. Sommerfeld, "Discrete -Event Simula tion of Large-Scale Poliomyelitis Vaccine Production," Process Biochem., 22, 74 (1987). 12. Bales, W. J., J. R. Johnson, and J. T. Sommerfeld, "Use of a Queuing Simulator in Design of a Batch Chemical Production System," Prodn. and Inventory Mgmt accepted for publica tion (1987). 13. Adebekun, A. K., Z. Q. Song and J. T. Sommerfeld, "GPSS Simulation of PVC Manufacture," Polymer Process Eng., 5, 145 (1987). 14. Glenn, S. L., R. T. Norris, Jr., and J. T. Sommerfeld, "Dis crete-Event Simulation of Sequencing Batch Reactors (SBR) for Wastewater Treatment," J. Water Poll. Cont Fedn., sub mitted for publication (1987). D ENTROPY Continued from page 97. Zenzen, The Nature of Irreversibility, D. Reidel Pub. Co., Dordrecht, Holland, 1985. 9. Teilhard de Chardin, P., The Phenomenon of Man, Harper and Row, New York, 1975. A good summary of Teilhard's arguments is presented in Barrow, J. D. and F J. Tipler, The Anthropic Cosmological Principle, Ch. 3, Clarendon Press, Oxford, 1986. 10. Adams, Henry, The Degradation of Democratic Dogma, Re print of the 1919 ed., Harper Torchbooks, New York, 1969 11. Levi-Strauss, C Tristes Tropiques, translated by J. and D. Weightman, Jonathan Cape, London, 1973. 12 Georgescu-Roegen, N., The Entropy Law and the E c onomic Process, Harvard University Press, Cambridge, 1971. 13. Daly, Herman E. Steady-State Economics, W. H. Freeman & Co., San Francisco, 1977. 14. Rifkin, J., Entropy: A New World View, Bantam Books, New York, 1980. 15. Tanner, Tony, City of Words, Chap. 6, Harper and Row, New York, 1971, and Lewicki, Zbigniew, The Bang and the Whimper, Greenwood Press, Westport, CT., 1984. 16. Greenland, Colin The Entropy Exhibition, Routledge and Kegan Paul, London, 1983. 102 17 Wiener, Norbert Human Use of Human B ei ngs Houghton Mifflin, Boston 1950. 18. Whether entropy can be linked to the information in a message is still an undecided question. The views of Brillouin and the Denbighs typify the pro and con, respectively. Brillouin, L., Scientific Uncertainty and Information, Academic Press New York, 1964. Denbigh, K. G and J. S. Denbigh, Entropy in Relation to Incomplete Knowl e dg e, Cambridge University Press, Cambridge, 1985. 19. Arnheim, Rudolf, Entropy and A rt, University of California Press, Berkeley 1971. 20. de Santillana, G., and H. von Dechend, Haml e t s Mill, Gambit Inc. Boston, 1969. 21. Romans 8:21. 22. For lucid and concise discussion of the philosophical problem of time, see Denbigh, K. G ., Thr ee C on ce pts of T ime, Springer-Verlag, Berlin, 1981. 23. Whorf, B. L., Languag e Thought and R e al i ty, M.I. T. Press, Cambridge, 1956. 24 See for example, Capra, F The T a o of Phy si c s, Bantam Books, New York, 1975. 25. Both the external and internal vi e w s are discussed by Morris Kline in Mathemat i cs and the S e ar c h fo r Knowl e dg e, Oxford University Press, Oxford 1985. Th e internal view is examined by Roger S. Jones in Physics a s M e taphor New American Library New York, 1982 D EDUCATOR: Bailey Continued from page 61. pitchers helps my students know each other better," Jay notes. "It also helps me maintain a friendly and open relationship with my group that's important in our work together." Research by Bailey and his students was recog nized by the Curtis W. McGraw Research Award of the American Society of Engineering Education in 1983, by Jay's election to the National Academy of Engineering in 1986, and by the AIChE Professional Progress Award in 1987. Bailey does have interests outside of the lab. Everyone who knows him remarks on his devotion to Sean, his 18-year-old son, who's now a freshman at the University of Colorado, Boulder. Jay's an avid amateur musician-the guitar is his instrument-and he loves active sports such as tennis, racquetball, and bicycling. He and Arnold also love to travel. Says Bailey, ''We went to Malaysia and Indonesia last sum mer and just wandered around for four weeks for ab solutely no professional reason whatsoever. It was wonderful." Frances Arnold sums up Jay Bailey's influence on his profession in the following way, "Jay stands out in the field as a pioneer in new techniques in the 8,000year-old discipline of biochemical engineering. You won't find many new products coming out of his lab, but you will find many new ideas." D CHEMICAL ENGINEERING EDUCATION

PAGE 49

REQUEST FOR FALL ISSUE PAPERS Each year Chemical Engineering Education publishes a special fall issue devoted to grad uate education. This issue consists of 1) articles on graduate courses and research, written by pro fessors at various universities, and 2) ads placed by chemical engineering departments describ ing their graduate programs. Anyone interested in contnouting to the editorial content of the 1988 fall issue should write to the editor, indicating the subject of the contribution and the tentative date it can be submitted. Deadline is June 1st. [eJ ;j #I book reviews CATALYST SUPPORTS AND SUPPORTED CATALYSTS by A. B Stiles Published by Butterworths, 80 Montvale Ave., Stoneham, MA 02180; 270 pages, $54.95 (1987) Reviewed by John B. Butt Northwestern University The title of this book is interesting enough since most of those who deal in catalysts, particularly of supported metals, often have an uneasy feeling that the "support" (dispersive phase, contact phase, car rier, holder-i.e., any number of names) has never been given enough attention. This book is a good start in trying to rectify this situation, and Dr. Stiles has collected a good group of reviews concerned with this. I particularly enjoyed the first five chapters, devoted to alumina, oxide supports other than alumina, acti vated carbon, and the associated information on their preparation and properties. The surface chemistry in volved in catalyst-support interaction is probably not as extensively dealt with here as in other sources, but the overall treatment considered together with prepa ration techniques is quite a satisfying and useful one. Chapter 7, on organic polymers, also falls into this category. The remainder of the book sort of strays from the announced title. Khoobiar has done a good job in Chapter 9 of "Spillover," and while significant oppos ing points are ignored, this is still a good review. Less satisfying are chapters on the "Commercial Applica tion of Molecular Sieve Catalysts" and "Multifunction ing Catalysts." This is all old stuff, it seems rather qualitative, and it strays far from tne announced title of the volume. The book is not very well proofread, as illustrated on page fourteen as well as many other places in the text. This makes me wonder how good the numbers SPRING 1988 in the many tables and illustrations are. The publisher should be more careful. In spite of these reservations, I would say that this is a book worth having. Get it, and learn about supports. D MASS TRANSFER WITH CHEMICAL REACTION IN MULTIPHASE SYSTEMS Vol I: Two-Phase Systems (679 pages) Vol II: Three-Phase Systems (399 pages) Edited by E. Alper; Martinus Nijhoff Publishers, The Hague, Netherlands, 1983. $140 Reviewed by Arvind Varma University of Notre Dame This two-volume book constitutes the proceedings of a NATO Advanced Study Institute held in Turkey in 1981. It includes thirty papers, primarily of a re view type, by twelve invited lecturers, and nine other contributions. Various topics in the area of mass transfer with chemical reaction in gas-liquid, liquid liquid, and gas-liquid-solid systems are covered. These topics arise in the context of either separation pro cesses or reaction engineering. Some of the papers treat the general problem of multiphase contacting and reactor design. Others deal with the modeling of specific types of contactors or reactors, and include methods for obtaining or estimating physicochemical and other data. Finally, some papers deal with a spe cific application, e.g., facilitated transport, bioreac tors, or reactors for coal conversion technology. The invited lecturers are experts in the area (mostly from Europe) who have written other reviews as well. The material is somewhat dated by now, and other more recent reviews and books have appeared in print. Nevertheless, these volumes constitute a rich source of information for this relatively narrow but important area, and they should prove quite useful to those involved with multiphase chemically reacting systems. The volumes were printed from camera-ready copy. For this type of production, the cost of the book is high. 103

PAGE 50

ti Na classroom LEVELS OF SIMPLIFICATION THE USE OF ASSUMPTIONS, RESTRICTIONS, AND CONSTRAINTS IN ENGINEERING ANALYSIS STEPHEN WHITAKER University of California Davis, CA 95616 The Na vier-Stokes equations P[* + vvv) = VP+ pg+ 2 v (l) are exceedingly difficult to solve in their general form. Thus there is great motivation to search for plausible simplifications. One of these simplifications takes the form: convective inertial effects are negligible. This allows us to extract the linear version of Eq. (1) which is given by p(~~) = Vp + pg + 2 y (2) One could express this simplification as an equation, and there is some advantage in identifying it as a Level I assumption and expressing the idea as Level I: pyVv = 0 (3) This type of statement indicates precisely what is being done in a mathematical sense, but it provides no basis for the action. For an engineer, it is more attractive to make a statement of the type: convective inertial effects are small compared to viscous ef fects. This leads to a Level II restriction of the form Level II: pvVv 2 v (4) In writing inequalities of this type it is understood that the comparison is being made between the abso~ lute values of the terms under consideration. Equation (4) has very definite advantages over Eq. (3) since a comment concerning the physics of the pro cess under consideration has been made. While Eq. (4) tells the reader what must occur in order that Eq. (2) be valid, it does not indicate, in terms of the pa rameters of the problem, when it will occur. In order to determine this, one must be able to estimate the magnitude of the terms in Eq. (4). The treatment here Copyright ChE Division ASEE 1988 104 Steve Whitaker received his undergraduate degree in chemical engineering from the University of California ot Berkeley and his PhD from the University of Delowar e. He is the outhor of three books: In troduction to Fluid Mechanics, Elementary Heat Transfer Analysis ond Fundamental Principles of Heat Transfer ond he is the co-editor (with Alberto Cassone) of Concepts and Design of Chemical Reactors His research deols with problems of multiphase transport phenomena, and he hos taught ot U.C. Davis, Northwestern University, and the Univer sity of Houston. will be brief since the details are given elsewhere [1]. We begin by expressing the velocity in terms of its magnitude and a unit tangent vector V = v). (5) so that the convective inertial terms take the form vVv = v>.Vv (6) Since X. is a unit tangent vector to a streamline, we have vVv = v dv ds (7) where s is the arclength measured along a streamline. The derivative in Eq. (7) is estimated as [2, Sec. 2.9] dV = t:.v ds LP (8) in which dv represents the change in v that occurs along a streamline over the inertial length LP. Use of Eq. (8) in Eq. (7) leads to an estimate of the convective inertial terms given by CHEMICAL ENGINEERING EDUCATION

PAGE 51

pvVv = o(P~:v) (9) It should be clear that a successful use of this estimate requires a reasonably good knowledge of the flow field. The viscous terms in Eq (1) can be expressed as v2v = a2v + a 2 v + a 2 v ax 2 al az 2 ( l 0) and the associated order of magnitude is given by v2v = o[LIVJX LIVJZ] L2 L2 L2 (11) X y Z Here Llvlx represents the change of v that occurs in the x-direction over the distance Lx, and the meaning of Llvl y and Llvlz is analogous for they and z-directions. We represent the largest of the three terms on the right hand side of Eq. (11) as Llv / L! and our estimate of the viscous terms takes the form v2v = ot;J ( 12) Here L is referred to as the viscous length. For many cases the value of Llv in Eq. (12) is comparable to the value in Eq. (9) and this allows us to substitute Eqs. (9) and (12) into the inequality given by Eq. (4) in order to obtain pvL 2 __ << l (13) LP Traditionally the Reynolds number is defined in terms of a length that is comparable to LThus we use pvL Re = :...______}!_ so that Eq. (13) takes the form Re (~) l p Level III: (14) (15) Obviously this Level III constraint has a great deal more utility than the Level I assumption given by Eq. (3) for it allows one to decide a priori whether the analysis is applicable to a particular problem. When simplifying the Na vier-Stokes equations on the basis of Eq. (15), one must remember Birkhoffs warning concerning the plausible intuitive hypothesis that "small causes produce small effects" [3]. While the route to Eq. (15) is straightforward, it is important to keep in mind that it is a scalar con straint associated with the magnitude of vectors and it must be used with care. In addition, it is crucial to understand that Eq. (15) has nothing to do with diSPRING 1988 My thoughts concerning the various levels of simplification began to develop several years ago, and while the origin remains diffuse, I might place it in the early stages of an undergraduate heat transfer course. mensional analysis, but is based entirely on the pro cess of estimating the derivatives of the velocity that appear in Eq. (4). When the flow is turbulent, Eq. (1) must be time-averaged and Eq. (15) then applies to the time-averaged inertial and viscous terms. Es timating the Reynolds stress, pv'Vv', is more difficult than estimating pvVv since a knowledge of the mag nitude of v' and the turbulent length scale is re quired. Often it is difficult to develop Level III constraints for complex problems (think about the "perfectly mixed" stirred tank reactor), and one must settle for the type of simplification indicated by Eq. (3) in order to meet deadlines and complete required course mate rial. From my point of view, the clearly stated Level I assumption is an acceptable simplification for it tells you what is being done and it reminds you that Level II restrictions and Level III constraints are waiting to be found. In addition, it should remind you that the analysis has an unspecified range of validity and that experiments and further analysis are in order. SCENE My thoughts concerning the various levels of simplification began to develop several years ago, and while the origin remains diffuse, I might place it in the early stages of an undergraduate heat transfer course. Because the subject under consideration was heat conduction, I began a lecture with v = 0 and quickly discarded radiant energy transport to arrive at pc aT = Vq {16) P at Since the assigned chapter and homework problems dealt with steady, one-dimensional heat conduction, we quickly moved to the following boundary value problem: o = L ( k dTJ dx dx (17) B.C. l: T = T 0 X = 0 (18) B.C.2: T = T 1 X = L (19) With the comment that "we can treat the thermal con ductivity as constant," I was on the verge of present105

PAGE 52

ing the classical result given by (20) However, there was a flaw in my development. The title of the chapter under consideration indicated that we were to study the subject of steady, one-dimen sional heat conduction, but it said nothing about the thermal conductivity being constant. One of the sages from the back row spotted the opening, and the trad itional train of events was disrupted by the observa tion that "nothing is truly constant." Delighted to find that a portion of the back row was awake, I pursued Eq. (17) a bit further to arrive at 0 = (d 2 T) + l (ak) (dT) 2 dx2J k aTJ dxJ (21) Since nonlinearities can be eliminated with impunity in an undergraduate class, I was willing to asume that ak/aT was zero and move on to the desired result given by Eq. (20). However, the back row was warm ing to the task, and one of its occupants persisted with, "But nothing is really zero is it?" A reviewer of this article suggested that I should have counter attacked with the Kirchhoff transformation [ 4, Sec. 2.16] so that Eqs. (17) through (19) could be expressed as B .C. l: B.C. 2: U = 0 X = 0 U = U 1 X = L Here the transformed temperature is given by U = i-T Mn2. dn ko n=T 0 (22) (23) (24) (25) in which k., is the thermal conductivity at the temper ature T 0 This approach would have avoided making the assumption that k was constant, but it would have delayed our arrival at Eq. (20) and the physical insight that can be gained from that result. While Eqs. (22) through (25) can provide an "exact" solution to the problem posed by Eqs. (17) through (19), we usually seek "approximate" solutions and often the approxi mations that we make are forced on our students by the title of the chapter and the name of the textbook. In engineering education there is a conspiracy among students and faculty to base their assumptions on the title of the chapter currently under considera tion. It is a game that is played with well established rules and most often both parties are loath to depart 106 from the tradition. Students who are tempted to ques tion the existence of the perfectly mixed stirred tank reactor are afraid that the instructor might plunge into a discourse on viscous dissipation, the Kol mogoroff length scale, and Damkohler numbers.* The terms of the treaty between students and faculty have been hammered out over the years, and by and large they work reasonably well. For example, can you im agine the difficulties of a study of fluid statics if it were preceded by the Level III constraints associated with av 2 P af pg pVVV pg 'v V pg (26) It is better to have a chapter entitled "Fluid Statics" so that the deck is cleared for an exploration of the pressure fields and forces associated with 0 = Vp + pg (27) Still the question was posed from the back row and it deserved an answer. Furthermore, it seemed to me that Eqs. (22) through (25) were most certainly not the answer, for the question was, in reality: How can you justify the simplification of Eq. (21) to arrive at d 2 T O=(~) dx 2 Clearly the second term in Eq. (21) cannot be neg lected on the basis of l (ak] (dT) 2 d 2 T k aT dxJ dx2 (29) but surely conditions must exi s t for which the vari ation of the thermal conductivity is small enough" so that Eq. (21) could be replaced by Eq. (28). This raises the question of "small relative to what?" and the fol lowing problem was devised to explore this question and to help students understand what is meant by quasi-st e ady. SAMPLE PROBLEM We consider the boundary value problem given by pc (ar] = k (a 2 TJ + (ak] (aTJ 2 ( 30 ) P at ax2 aT ax I.C T=T 1 t=O (31) B.C. l: T = T 1 + (T 0 \lg(t) x = o (32) B.C. 2: T = T 1 x = L (33) It is bad enough that the material would not be available in th e text but what is worse is that it w ould not be c overed on the final! CHEMICAL ENGINEERING EDUCATION

PAGE 53

Here g(t) is a function such that g(t) = 0, t = 0 g{t) = l t = t* {34a) {34b) The classic test piece in a study of separation of vari ables is associated with t 0 + and (ak/aT) 0, but in this case we should simply think of t* as some characteristic time associated with the boundary con dition at x = 0. Everyone knows that if t* is "large enough" and if the variation of k with T is "small enough," the solution to this boundary value problem will yield the linear temperature profiles associated with Eq. (28). The Level I assumptions related to these conditions are Level Ia: Level lb: (:~) = 0 (:~) = 0 {35a) (35b) and one should be careful to identify the first of these as the quasi-steady assumption. Our objective at this point is to develop the Level II and Level III restrictions and constraints that are associated with Eqs. (35a,b). Thus we seek to deter mine what is "large enough" and what is "small enough." If you have an idea that a satisfactory solu tion to Eqs (30) through (34) might be given by* T = T 1 + {T 0 \)g{t) (1 (r)) (36) the possibility can be explored by decomposing the temperature into the result represented by Eq. (36) and whatever else is left. One method of doing this is to arrange Eq. (30) as a 2 T = l (aT) l (elk) (aT)2 ax2 a at k aT ax) (37) and to form the indefinite integral in order to obtain :! :!I-,+ r [~ lm H:~Hm} ,,.i /;=O Use of the definition fl = l(aT) l(clk) (aT) 2 a at k clT els (39) *This solution is obtained by using Eqs. (35a,b) to reduce Eq. (30) to the form given by Eq. (28). When Eq. (28) is solved subject to Eqs. (32) and (33), the solution given by Eq. (36) results. SPRING 1988 along with a second integration leads to T(,,t) 'lx-o + ,[:!]._, + r rndtdn (40) T)=O s=O The boundary conditions given by Eqs. (32) and (33) can be used to evaluate the two constants in Eq. (40) and the general solution is given by T { x. t > = T 1 + T O T 1 > g { t > ( 1 ( r)] + r r:(dn [f] r r:(dn (41) n=o s=o n=o s=o One should keep in mind that this is an exact rep resentation for the temperature, but it is only useful when the integrals are negligible. The integrals in Eq. (41) can be estimated as r 1)=0 r 1)=0 l =T) x2 ndsdn = O(n} 2 s=O l =T) L2 ndsdn = O{n) 2 s=O and use of these estimates in Eq. (41) leads to T(x,t) =Tl+ {To \)g(t}(l (r) l + ~L (1 . (r) )O{fl) {42a) (42b) (43) We are now in a position to state that the solution for T(x, t) is qiven by T ( x, t) = T 1 + (T O T 1 ) g { t) [ l ( r)) { 44) provided that the following inequality is satisfied 2 {To Tl)g(t) >> t O{n) (45) This result allows us to replace the Level I assump tions given by Eqs. (35a,b) with the following Level II restrictions 107

PAGE 54

Level Ila: (T 0 T 1 )g(t) 0(~ 2 (~~]] (46a) Level Ilb: (T 0 T 1 )g(t) 0(~ 2 (:~] (~!r] (46b) Here we are beginning to see how "long'' one must wait before the solution becomes quasi-steady, and how "small" the variation of the thermal conductivity must be in order that the last term in Eq. (30) can be discarded. In order to proceed further, we must be willing to estimate the derivatives that appear in Eqs. (46a,b) and in this development we will be satisfied with the rather crude estimates given by [2, Sec. 2.9]. aT __ [Tlx=o Tlx=L] __ [(To Tl)g(t) at 0 t o t 't t (47a) ;: a o[r!.-,,T!xLl a o[(r, -,r,)g(tl (47b) This aspect of the problem could be considered more carefully by introducing the thermal boundary layer thickness; however, we are interested in knowing under what circumstances Eq. (44) is valid and the estimates given by Eqs. (47a,b) are consistent with that objective. When Eqs. (47a,b) are used in Eqs. (46a,b) we obtain the Level III constraints given by Level Illa: at l t > t* L 2 (48a) Level Illb: (:~] (T 0 T 1 )g{t) l (48b) The first of these clearly indicates that the process will be quasi-steady when t* is large compared to L2/cx and an exact solution of the boundary value prob lem will indicate that this is a conservative constraint. Since g(t) has an upperbound of one, Eq. (48b) can be replaced by Level Illb: (~~J(T 0 T 1 ) 1 (48c) While the results given by Eqs. (48a,b,c) are some thing that "everyone knows," not everyone knows how to arrive at these constraints without solving the full boundary value problem and exploring special cases. In addition, the identification of various levels of simplification is an important concept to bring to the attention of students, for it allows us to move quickly to certain simple engineering solutions while 108 reminding us of our obligation to be more thorough when time permits or necessity demands. Following up on our obligations is sometimes easy to do. For example, in the typical heat transfer course transient processes are always studied, an:d when exact solu tions are available it is easy to remind students of prior constraints that were developed on the basis of order of magnitude analysis. In the study of transient heat conduction in a flat plate one finds that Eq. (48a) can be replaced with cxt/L2 2:: 1, thus providing a clear indication that the original estimation was overly se vere. To support the result given by Eq. (48c), a homework problem associated with Eqs. (22) through (25) does rather nicely. The process of following order of magnitude estimates with exact solutions is an at tractive method of encouraging students to develop their own assumptions, restrictions and constraints. As they gain confidence in this process, chapter titles become guidelines for the voyage rather than con straints for the next exam. REFERENCES 1. Whitaker, S., 1982 Th e laws of continuum physics for single phase, single-component systems," Chapter 1.1 in Handbook of Multiphase Systems, edited by G. Hetsroni, Hemisphere Pub. Co., New York. 2 Whitaker, S., 1983, Fundat menta l Principles of H eat Trans fer, R. E. Krieger Pub. Co., Malabar, Florida. 3. Birkhoff, G. 1960, Hydrodynamic s A Study in Logic Fact, and Similitude, Princeton University Press Princeton, NJ. 4. Carslaw, H. S., and J. C. Jaeger, 1959, Conduction of Heat in Solids, Oxford Press, London D ljij=I books received Gaseous Detonations: Their Nature, Effects and Control, by M.A. Nettleton. Chapman & Hall, Methuen, Inc., 29 West 35th St., New York, NY 10001; (1987) 255 pages, $72 Radiation and Combined Heat Transfer in Channels, by M. Tamonis (edited by Zukauskas and Karni). Hemisphere Pub lishing Co ., 79 Madison Ave., New York, NY 10016; (1987) 239 pages, $69.95 Particulate and Multiphase Processes: Vol. 1, General Partic ulate Phenomena; Vol. 2, Contamination Analysis and Con trol; Vol. 3, Colloidal and Interfacial Phenomena, by Ariman and Veziroglu; Hemisphere Publishing Corp., 79 Madison Ave., New York, NY 10016 (1987); 932; 760; 544 pages, $133; $131;$131 Batch Process Automation: Theory and Practice, by Howard P. Rosenof and Asish Ghosh. Van Nostrand Reinhold Co., 115 Fifth Ave New York, NY 10003 (1987); 336 pages Vapor Cloud Dispersion Models, by Steven R. Hanna and Peter J. Drivas. AIChE, 345 East 4 7 St., New York, NY 10017 (1987); 177 pages, $40 members, $75 others Handbook of Thermodynamic High Temperature Process Data, by A.L. Surls. Hemisphere Publishing Co 79 Madison Ave., New York, NY 10016 (1987); 601 pages, $139.95 CHEMICAL ENGINEERING EDUCATION

PAGE 55

ACKNOWLEDGMENTS Departmental Sponsors The following 156 departments contributed to the support of CEE in 1988 with 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 California Institute of Technology California State Polytechnic University California State University, Long Beach University of California (Berkeley) University of California (Davis) University of California (Los Angeles) University of California (Santa Barbara) University of California at San Diego 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 Florida Florida State University Florida Institute of Technology 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 of Technology Louisiana State University Louisiana Technical University University of Louisville University of Lowell University of Maine Manhattan College University of Maryland University of Massachusetts Massachusetts Institute of Technology McGill University McMaster University McNeese State University University of Michigan Michigan State University Michigan Technical University University of Minnesota University of Missouri (Columbia) University of Missouri (Rolla) Monash University Montana State University University of Nebraska University of New Hampshire New Jersey Institute of Tech University of New Mexico New Mexico State University University of New South Wales City University of New York Polytechnic Institute of New York State University ofN Y. at Buffalo North Carolina A&T State University North Carolina State University University of North Dakota Northeastern University Northwestern University University of Notre Dame Nova Scotia Technical College 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 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 University University of Saskatchewan University of South Alabama University of South Carolina South Dakota School of Mines University of South Florida University of Southern California University of Southwestern Louisiana Stanford University Stevens Institute of Technology University of Sydney Syracuse University Tennessee Technological University University of Tennessee Texas A&M University University of Texas at Austin Texas Technological University University of Toledo Tri State University Tufts University Tulane University University of Tulsa Tuskegee Institute University of Utah Vanderbilt University Villanova University University of Virginia Virginia Polytechnic Institute Washington State University University of Washington Washington University University of Waterloo Wayne State University West Virginia College of Grad Studies West Virginia Institute Technology West Virginia University University of Western Ontario Widener University University of Windsor University of Wisconsin (Madison) Worcester Polytechnic Institute University of Wyoming Yale University Youngstown State University If your department is not a contributor, write to CHEMICAL ENGINEERING EDUCATION, c/o Chemical Engineering Dept., University of Florida, Gainesville FL 32611, for information on bulk subscriptions.

PAGE 56

Union Carbide Union Carbide Corporation is one of the nation's major industrial companies. We employ more than 50,000 people at 700 plants and facilities in 37 countries, and market our products in about 130 countries. Over the past few years, we have become a more streamlined simplified and focused organization structured to respond effectively to rapidly changing economic and market conditions Streamlined for Success Our principal businesses include chemicals and plastics, industrial gases, and carbon products Union Carbide is one of the world's leading producers of ethylene oxide/glycol and polyethylene. We have the widest range of solvents of any U.S chemical company and are the largest domestic supplier of industrial gases We're also the world s largest producer of carbon and graphite electrodes for steel-making. Union Carbide Corporation Department of University Relations 390/d Ridgebury Road Danbury, CT 06817-0001 An Equal Opportunity Employer