Chemical engineering education ( Journal Site )

Material Information

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


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


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

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 01151209
lccn - 70013732
issn - 0009-2479
lcc - TP165 .C18
ddc - 660/.2/071
System ID:

Full Text

Sma e g e i--o o

ackiczeede& amd laa*M.....



...004 apoli oi%
wilt a doatotiat 4 imndS.


Department of Chemical Engineering
University of Florida
Gainesville, Florida 32611

Editor: Ray Fahien (904) 392-0857
Associate Editor: Mack Tyner
Editorial & Business Assistant:
Carole C. Yocum (904) 392-0861
Publications Board and Regional
Advertising Representatives:

Lee C. Eagleton
Pennsylvania State University

Past Chairman:
Klaus D. Timmerhaus
University of Colorado

Homer F. Johnson
University of Tennessee
Jack R. Hopper
Lamar University
James Fair
University of Texas
Gary Poehlezn
Georgia Tech

Darsh T. Wasan
Illinois Institute of Technology
Lowell B. Koppel
Purdue University

William B. Krantz
University of Colorado
C. Judson King
University of California Berkeley

Angelo J. Perna
New Jersey Institute of Technology
Stuart W. Churchill
University of Pennsylvania
Raymond Baddour
A. W. Westerberg
Carnegie-Mellon University

Charles Sleicher
University of Washington
Leslie W. Shemilt
McMaster University
Thomas W. Weber
State University of New York

Chemical Engineering Education





Department of Chemical Engineering
2 ChE at Clemson, J. M. Haile, D. D. Edie

The Educator
6 J. M. Smith, N. McGuinness

10 R. H. Wilhelm's Influence on the Development
of Chemical Reaction Engineering: From the
Wilhelm Lectures at Princeton
Rutherford Aris

16 Integrating Chemistry and Engineering: A
Course in Industrial and Engineering Chemistry
G. L. Schrader, R. L. Pigford, B. C. Gates

20 Solar Hot Water Heating by Natural Convection
Richard D. Noble
28 Direct Digital Control Liquid Level
Experiment, M. F. Adb-El-Bary

24 A Process Control Undergraduate Option
J. C. Hassler, K. I. Mumme'

34 TUTCHE-A Program Package for Tutoring
Chemical Engineers
F. P. Stainthorp, D. Lomas, J. Alonso

32 Survey: Computer Usage in Design Courses
Ernest J. Henley

23 In Memoriam: Ralph E. Peck

27, 46 Book Reviews

CHEMICAL ENGINEERING EDUCATION is published quarterly by Chemical
Engineering Division, American Society for Engineering Education. The publication
is edited at the Chemical Engineering Department, University of Florida. Second-class
postage is paid at Gainesville, Florida, and at DeLeon Springs, Florida. Correspondence
regarding editorial matter, circulation and changes of address should be addressed
to the Editor at Gainesville, Florida 32611. Advertising rates and information are
available from the advertising representatives. Plates and other advertising material
may be sent directly to the printer: E. O. Painter Printing Co., P. O. Box 877,
DeLeon Springs, Florida 32028. Subscription rate U.S., Canada, and Mexico is $15 per
year, $10 per year mailed to members of AIChE and of the ChE Division of ASEE.
Bulk subscription rates to ChE faculty on request. Write for prices on individual
back copies. Copyright 1983 Chemical Engineering Division of 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 of the ASEE
which body assumes no responsibility for them. Defective copies replaced if notified
within 120 days.
The International Organization for Standardization has assigned the code US ISSN
0009-2479 for the identification of this periodical.


Aerial photo of Clemson campus.

I department


Clemson University
Clemson, SC 29631

tion for the State of South Carolina, is the real-
ization of a long-held dream of its founder-
Thomas Green Clemson. Clemson was born in
Philadelphia in 1807 and, although the events of
his childhood are obscure, it is certain that during
his mid-teens Clemson acquired a lifelong interest
in science in general and chemistry in particular.
By 1826 Clemson was in Paris where he audited
lectures of Thenard, Gay-Lussac and DuLong at
the Sorbonne, studied in the laboratory of Robi-
quet, and attended the Royal School of Mines. In
1831 Clemson received an assayer's certificate
from the Royal Mint and during the years 1831-
39 he developed a profitable business as a consult-
ing and mining engineer in Paris, Philadelphia,
and Washington.
While in Washington in 1838 Clemson met
Anna Maria Calhoun, the daughter of the South
Carolina statesman, John C. Calhoun. By this time
Copyright ChE Division, ASEE, 1983

Clemson was a successful engineer and business-
man, a world traveler, a linguist and conversa-
tionalist, a man interested in science and politics
though himself neither scientist nor politician,
and one whose hobbies included painting in oils
and music-making on the violin. The Calhoun-
Clemson marriage took place in November 1838
and subsequently Clemson assumed the manage-
ment of the Calhoun plantation, Fort Hill, located
in the northwest corner of South Carolina. Im-
pressed by the value of proper training for those
who farm, Clemson began to consider ways of
applying scientific knowledge to agriculture. From
1844 until 1851 Clemson served as charge
d'affaires to Belgium and he took the opportunity
of those years in Europe to broaden his study of
In 1852 the Clemsons bought a small farm in
Prince Georges County, Maryland, where Clemson
began testing his ideas on scientific methods of
agriculture. He wrote and published extensively on
agricultural chemistry; he promoted, the estab-
lishment of the Maryland Agricultural College
(later a part of the University of Maryland) ; he
actively supported passage of the Morrill Act for


Land-Grant Colleges; he gave, at the Smithsonian
Institution, a series of lectures entitled "Chemistry
Applied to Agriculture." Following his appoint-
ment in 1860 as U.S. Superintendent of Agricul-
tural Affairs, Clemson began developing plans to
create a U.S. Department of Agriculture. Those
plans and his position as Superintendent had to
be abandoned when war broke out in 1861.
In 1866, with both John C. Calhoun and his
wife dead, the Fort Hill plantation became the
object of legal entanglements that were resolved
by holding a public sale. Clemson bought the
plantation at that time. Over the last decade of
his life Clemson's principal interest was establish-
ing an agricultural college on the Fort Hill plan-
tation. Clemson died in 1888, leaving Fort Hill's
814 acres to the State under the provision that
the land become the site of the Clemson Agricul-
tural College of South Carolina. The South Caro-
lina legislature ratified an Act of Acceptance of
Clemson's will on December 24, 1888.

Construction of college buildings on the Fort
Hill plantation was begun in 1890 and in July
1893 the college opened its doors to its first class-
446 students. From its inception until 1954, the
student body was all male and organized as a
corps of cadets.
In the early 1930's the college benefited from
the land resettlement policy of Roosevelt's New
Deal: the government purchased some 27,469 acres
within a ten mile radius of Fort Hill and assigned
Clemson College the responsibility of overseeing
its revitalization. The entire acreage was deeded
to the college in 1954 and the land now serves as
an extensive laboratory for the colleges of Agri-
culture and Forest Resources.
In 1964 the name of the college was formally
changed to Clemson University. Today, the uni-
versity is composed of a graduate school and nine
colleges: Agriculture, Architecture, Education,
Engineering, Forest and Recreation Resources,
Liberal Arts, Nursing, Sciences, and Commerce
and Industry. From circa 1970 the university ad-
ministration has limited total enrollment to about
10,500 students. The university library is housed
in an attractive modern building and its holdings
number well over one million items, including
bound volumes, microfilm, and microfiche. The uni-
versity computer is an IBM 370/3033 supported
by a complete selection of peripheral devices. In
addition, the College of Engineering has purchased

two DEC VAX computers: one dedicated to com-
puter graphics, the other to "number crunching"
research problems.
No description of Clemson University would
be complete without mention of intercollegiate
sports. Because of fanatical alumni loyalty and a
well-organized athletic fund-raising machine,
Clemson has been able to sustain national promi-
nence in several minor sports, especially soccer,
baseball, and tennis. As our obligatory comment on
football, we note that the legendary John Heisman
coached at Clemson from 1900 to 1903 and during
that period led the team to its first undefeated

Chemical engineering was first introduced as
a course of study at Clemson in 1917. There was
no chemical engineering department or any fac-
ulty; the curriculum was drawn from courses
in mathematics, physics, chemistry, mechanical
engineering, etc. The university catalog for 1920
is the first to show any enrollment in this curric-
ulum-seven students. In the spring of 1923 four
of these were the first chemical engineering gradu-
ates from Clemson; the attrition rate has changed
little since that first class.
The chemical engineering curriculum does not
appear in the University Catalogs for the years
1923-1933 but was reintroduced in expanded form

Undergraduate chemical engineering
students achieve a high degree of camaraderie
and identify strongly with the department, both as
students and later as alumni. Much of this is
due to an extremely active student
chapter of the AIChE.

in 1933 again with no department, courses, or
faculty of its own. The catalog attempts to entice
students into the curriculum with the following
Competition is compelling the industries to abandon
rule-of-thumb methods. They are using more and more
men trained in the principles of Chemical Engineering-
to design their plants and to supervise the operations of
various processes.
In 1934 there were 27 students in chemical en-
gineering and the number grew steadily to 81 by
1939. In the 1939 catalog the curriculum no longer
appears under the supervision of Engineering
but pops up in the Chemistry Department under a


new name, "Chemistry-Engineering," which per-
severed until 1946.
The 1946 catalog lists the curriculum back
under the supervision of Engineering and, for the
first time, includes a separate Department of
Chemical Engineering with its own courses. The
catalog of 1947 indicates two faculty members
in the fledgling department: C. E. Stoops, Jr.,
Professor and Head, and C. E. (Charlie) Little-
john, Assistant Professor.
In the 1948 catalog Professor Stoops does not
appear; rather, the Department Head is listed as
Allan Berne-Allen, who remained as Head until
1955. The year 1955 marked something of a turn-
ing point in the development of the department:
Charlie Littlejohn was the sole faculty member
and, hence, Head of the department. The cata-
log lists sixteen chemical engineering courses, of
which eleven were required, and the enrollment
stood at 125. It is intriguing to try to imagine
Charlie Littlejohn meeting all those courses

Earle Hall, home of the Chemical Engineering De-

and grading all those papers. In fact, however,
Dr. Littlejohn had teaching help from a professor
in agriculture who had a Bachelor's degree in
chemical engineering.
In January 1956 George Meenaghan joined the
faculty and enrollment was 156. In 1958 Chris
Alley and Bill Barlage brought the department's
faculty complement to four and enrollment had
climbed to 184. In 1958 Clemson received a grant
of $1.175 million from the Olin Foundation for
construction of a building and purchase of equip-
ment for chemical engineering. Construction was
begun in September 1958 and Earle Hall was dedi-
cated at the end of 1959. (Samuel B. Earle was
Dean of Engineering from 1933 to 1950 and
acting President of Clemson in 1918 and again
in 1924-25.) Earle Hall is a 50,000 square-foot
facility containing five classrooms, eleven fac-
ulty offices, a library, an auditorium, a student

lounge, a seminar room, shop, eight general pur-
pose laboratories, thirteen two-man research lab-
oratories, and a three-level 9,000 square-foot unit
operations laboratory. The UO Lab was laid out
and equipped under Charlie Littlejohn's direction
and has served as a model for similar labs at sev-
eral other universities. The undergraduate chemi-
cal engineering curriculum was first accredited
by the Southern Association of the Engineer's
Council for Professional Development (ECPD)
in 1959.
In 1960 a Master of Science program in chemi-
cal engineering was started and in 1962 the PhD
program was added. The first PhD in engineer-
ing in South Carolina was awarded in 1965 to
Jerry A. Caskey, a student under Bill Barlage. To-
day Dr. Caskey is a member of the Faculty at
Rose- Hulman Institute of Technology. Other PhD
graduates who are now following academic careers
include: Dick Stewart, 1966 (Northeastern) ; Dan
Reneau, 1966 (Louisiana Tech); Dendy Sloan,
1974 (Colorado School of Mines); Gary Mock,
1976 (North Carolina State) ; Eric Snider, 1978
(Tulsa) ; David Cooper, 1980 (Central Florida).
In May 1975 Charlie Littlejohn died of cancer.
He had taught chemical engineering at Clem-
son since the founding of the department and
had served as Head since 1956. It was his
strong personality that molded the attitudes and
aspirations of the department-both faculty and
students. His philosophy was simple: the depart-
ment at Clemson exists to provide training in
chemical engineering for undergraduates who will
follow industrial careers. Charlie was an excellent
teacher, that rare individual who inspires dili-
gence, respect, professionalism, and affection in
his students. On his death, alumni spontaneously
created the C. E. Littlejohn Scholarship Fund to
provide support for the education of highly quali-
fied undergraduates in chemical engineering. In-
dustrial gifts were also forthcoming to refurbish
a room in Earle Hall as the C. E. Littlejohn Me-
morial Student Lounge. Since 1975 W. B. (Bill)
Barlage, Jr. has served as Head of the department.


In 1981 the department was composed of 168
freshmen, 89 sophomores, 78 juniors, 90 seniors,
22 graduate students, and a dozen faculty. Fresh-
men enrollment in the department has increased
steadily from 32 in 1972. (The department does
teach a course to second semester freshmen.)


Equally dramatic has been the increasing coed
enrollment: in the fall of 1981, 30% of the incom-
ing class was female.
The undergraduate curriculum is a strong tra-
ditional program requiring 144 semester hours for
graduation. Thirty-eight of these are taught in the
department. Required courses include stoichiome-
try, numerical methods, kinetics, process design,
process control, transport phenomena, two semes-
ters of thermodynamics, and three semesters of
unit operations theory. Communication skills are
emphasized in two semesters of unit operations
laboratory, a FORTRAN programming course,
senior seminar, and junior plant-inspection trips.
Furthermore, selected juniors and seniors are in-
vited to join in work on research problems in col-
laboration with graduate students and faculty.
About six years ago an undergraduate coop-
erative education program was instituted. Un-
der the nurture of Joe Mullins, the depart-
ment's co-op program has been the most successful
of any on campus and co-op students formed one-
third of the chemical engineering class of 1981.
Undergraduate chemical engineering students
achieve a high degree of camaraderie and identify
strongly with the department, both as students
and later as alumni. Much of this is due to an
extremely active student chapter of the AIChE.
Each fall the chapter sponsors a shrimp boil to
welcome new and returning students; each spring
a pig roast is held to send off the graduating class.
During the academic year the chapter holds
monthly meetings at which invited speakers pre-
sent talks on a variety of technical, business, or
popular topics. In addition, the chapter holds in-
formal drop-ins throughout the year, organizes
tutoring services, sponsors a float in the fall foot-
ball parade, and makes sporadic trips into the
mountains or down the Chatooga River. The chap-
ter has received the Regional Outstanding Student
Chapter Award in five of the previous seven years
and has been recognized by the national AIChE
as an Outstanding Student Chapter in each of the
last eight consecutive years.
At the graduate level, the department offers
MS and PhD programs to full-time students and a
Master of Engineering (ME) program to part-
time students. The MS degree requires 24 hours
of graduate coursework plus a thesis written on a
research project; the ME degree requires 30 hours
of coursework plus satisfactory completion of
an advanced engineering design problem. In re-
cent years the graduate enrollment has averaged

about twenty students, of which about 20% have
been tramontane.
In 1981 two new programs were started with
the goal of bolstering graduate student enroll-
ments. One of these is a type of graduate co-op
program called the Industrial Residency Program.
This program terminates with an MS and func-
tions as follows: On graduation with a BS in
chemical engineering in the spring, the new grad-
uate resident selects the company and project
upon which he intends to work. The selection is
made from projects that participating companies
have previously submitted to the department. The
student is then assigned a faculty advisor from
the department and a project advisor from the
company. The student spends the first summer
at the company becoming familiar with the
project. In the fall the student returns to Clem-
son and takes two consecutive semesters of
graduate courses, completing the 24 semester
hour course requirement. The second summer
the student returns to the company and works
full-time for about seven months completing
the research project and writing the thesis.
The student is paid by the company at a BS-level
rate for the months he is actually working at

Professor Dan Edie with student and INSTRON Rhe-
the plant. Typically, the salary is prorated over the
nominal nineteen months needed to complete the
The second new graduate program is an In-
dustrial Fellowship Program in which exception-
ally promising students are awarded generous,
Continued on page 48.


nl educator


University of California, Davis
Davis, CA 95616

OE MAUK SMITH is known at the University of
California, Davis, as "Mr. Chemical Engineer-
ing". As founding Chair of the Chemical Engi-
neering Department and its tireless protector in
its fledgling years, he became so closely identified
with his department in the minds of his Academic
Senate colleagues across campus that they see
him largely as its guardian. His colleagues within
the profession give Joe the same appellation, but
for a different reason: for his prominence in his
As a member of the National Academy of En-
neering, J. M. Smith has been accepted by his
peers as one of a very select group of engineers.
As winner of the Richard H. Wilhelm Award of
the American Institute of Chemical Engineers in
1977, he was recognized for "advancing the fron-
tiers of chemical reaction engineering with em-
phasis on originality, creativity, and novelty of
concept and application." He was named to the
AIChE Wm. H. Walker Award in 1960. In addi-
tion to awards from the American Chemical So-
ciety, American Society for Engineering Educa-
tion, and the Guggenheim Foundation, he has been
recognized by the entire faculty of his campus, be-
ing named to the highest Academic Senate honor,
Faculty Research Lecturer, in 1970.
The present Department Chair, Ben McCoy,
describes J. M. Smith's contributions in this way:

"Dr. Smith makes it a point
to get to know students by name.'He
attends student functions whenever possible and
was a regular fan at the Senior class
intramural softball games."

Copyright ChE Division, ASEE, 1983

"Joe has made substantial contributions to the
areas of kinetics, reactor analysis, mass transfer,
and thermodynamics with both experimental and
analytical methods. His career work has resulted
in over 260 publications. One most influential
paper-N. Wakao and J. M. Smith, "Diffusion
in Catalyst Pellets", Chem. Eng. Sci 17, 825
(1962)-was recently identified and honored as
one of the most cited in its field. Joe Smith has
made major contributions in chromatographic
systems. He has pioneered in applying moment
methods to the analysis of dynamic response ex-
periments. Adsorption, intraparticle diffusion, dis-
persion, channeling, chemical reaction, and re-
generation are some of the phenomena examined,
clarified, and quantified. Whereas in pulse re-
sponse experiments his students once picked data
points off recorder chart paper for subsequent
numerical integration, they now punch buttons
on computers to get the temporal moments."
Recently Joe and his coworkers have been
focusing their attention on trickle-bed reactors.
These three-phase systems with gas and liquid
flowing over catalytic particles pose interesting
challenges with combined fluid mechanics, mass
transfer, and chemical reaction phenomena. Cur-
rent interests also include thermodynamics and


M. MSmadh



kinetics of supercritical extraction of oil shale
and tar sands. As is much of his research, Joe's
studies in these areas were motivated by practical
questions arising in industrial processes.
Joe's work has gained recognition through his
major textbooks, his journal publications, and
the prominence of his former students, now dis-
persed throughout the international chemical en-
gineering community. Professor McCoy reports:
"Joe Smith publishes at least eight papers each
year, in the most widely read research chemical
engineering journals. Nearly all of these have as
co-authors graduate students or post-doctoral
scholars who go on to responsible positions in
corporations and universities around the world.
Most of these research associates come to U.C.
Davis specifically to study with Joe." Not only
have graduate students and postdoctoral students
come to Davis to work with him, he has also
gone on Fulbright appointments to the Nether-
lands (1953), Argentina (1963), Spain (1965),
Ecuador (1970), and Chile (1970). His two text-
books have influenced a generation of chemical
engineers around the world. Introduction to Chem-
ical Engineering Thermodynamics, coauthored
with Hank Van Ness of Rensselaer, was published
in 1949. Chemical Engineering Kinetics first ap-
peared in 1956. Both texts are in the McGraw-
Hill Chemical Engineering Series, and both are
now in their third editions. The texts are filled
with problems of practical interest, reflecting Joe's
wide-ranging experience in industrial consulting.
His former and current students bear eloquent
witness to his teaching effectiveness. Bob Reid
of M.I.T. says this of his former professor (CEE
9, 106 (1975)): "Joe really turned me on; he
was the first teacher who gave me problems I
couldn't solve. As a matter of fact, I'm still using
some of them. I had never met a professor who
wanted to know you as an individual. If I were
working in the lab late at night, he'd stop in,
put his feet up on the desk, and talk with me,
not necessarily about my thesis, but about almost
Former student Ray Fahien of the University
of Florida says, "It is apparent from his numer-
ous papers and several widely adopted books that
Joe Smith is a scholar of international stature. But
what is not apparent is that he is also a warm,
sensitive person who thoroughly enjoys the com-
pany of others-as well as a charismatic leader
who knows how to motivate people. He does this
by imparting a sense of purpose or mission to

every activity in which he is engaged. For exam-
ple, in his lectures Joe would begin by asking
the question: 'Why are we, as chemical engineers,
interested in this topic?' In his discussions with
beginning graduate students, he would always
emphasize the project's importance in science or
engineering. His books also first make clear the
relevance of each chapter or section before going
into the theory.
"Joe also motivates his students by being pro-
fuse in his praise for work well done, but sparing
in his negative criticism. Even before Skinner
became a popular name in engineering education,
Joe recognized the motivating effects of positive

Joe's work has gained recognition through
his major textbooks, his journal publications, and
the prominence of his former students, now
dispersed throughout the international
chemical engineering community.

Each year Joe Smith has taught an average
of five courses, both graduate and undergraduate,
at U.C. Davis. The custom at Davis is to divide
the sections of senior chemical engineering labora-
tory among the faculty. Exercising his usual ini-
tiative, curiosity, and innovation, Joe recently
devised a new chemical reaction kinetics experi-
ment for the course. In addition to supervising a
number of research graduate students and post-
doctoral scholars, Joe advises his share of chemi-
cal engineering undergraduates.
Students applaud Professor Smith's teaching
style, which they characterize as organized and
thorough, with lectures clearly and enthusiastic-
ally presented. Joe frequently applies the Socratic
method in the classroom; all agree the method
encourages preparation.
One of Smith's recent undergraduate students,
Anne McGuinness of the Class of '82, gave this
view of Joe as a lecturer, lab instructor, and
mentor: "Any overview of his accomplishments
cannot neglect his ability as a teacher. Dr. Smith
received the AIChE Outstanding Teacher Award
from the Davis chapter in 1982, his 21st year of
teaching here. Dr. Smith begins his lectures with
a short review of homework, emphasizing the im-
portant aspects of the problems. When introduc-
ing new material he calls on students in class to
answer questions, insuring that the class is fol-
lowing the development of his presentation. He


is always willing to stop his lecture to answer
questions and uses them to clarify his points.
"In laboratory courses, Dr. Smith has students
review their proposed experiments with him prior
to lab sessions. He uses these sessions to ask ques-
tions and stimulate the students to look at other
possible aspects of the experiment.
"Dr. Smith makes it a point to get to know
students by name. He attends student functions
whenever possible and was a regular fan at the
Senior class intramural softball games."
Speaking of his methods of guiding students
in research Joe recently said: "As the years go
I i I in 0 l -H .

MINp. -. -
Joe examines output data from reactor for super
critical extraction of oil shale.
by it becomes ever more clear to me that there
are many effective ways to collaborate with stu-
dents in chemical engineering research. The most
effective approach for one professor may not be
suitable for another. Hence, the following com-
ments represent simply a single procedure which
has been useful for me. For research which is at
least in part experimental it is helpful to visit
students in the laboratory each day. Perhaps there
seems little to discuss, but, from a general con-
versation about the problem and apparatus, new
ideas and new approaches usually emerge. During
these discussions it may be helpful to present
unanswered technical questions to the student.
Also it is stimulating to offer challenges in the
direction and extent of the research that are per-

haps a bit beyond reasonable expectations in the
time available. On the other hand, students seem
to be discouraged if a thesis problem is not clearly
defined beforehand, but is left open-ended. Of
course research cannot be described in detail in
advance, but the impression can be given to the
student that diligent application to a particular
phase of a research project should result in a
thesis in a reasonable amount of time.
"Formal, weekly meetings, for which the stu-
dent must summarize prior work and plan for
future studies, have merit. When there are data
to analyze, such weekly meetings might last sev-
eral hours. During routine aspects of the project,
the meetings may last but a few minutes. Never-
theless, the idea of having a weekly meeting where
the students will have time to talk to the faculty
member at length, and must summarize their
thoughts, seems to be helpful. Also, in working
with graduate students it is often valuable to
have the students write down in a clear fashion
their progress during a previous period and their
plans for future work. The act of putting thoughts
and results on paper requires clear thinking about
the project."
The care for and attention to students, the
constant "offering of challenges", and the in-
sistence on clarity in expression of these remarks
characterizes Joe Smith's teaching methods. He
claims to have learned some of these from his
own teachers and colleagues.
"I was first introduced to chemical engineering
in a most fortunate way with William N. Lacey
teaching a course to two students using the classi-
cal book authored by Walker, Lewis, and Mc-
Adams. During that period at the California In-
stitute of Technology with the small number of
chemical engineering students, all chemical engi-
neering courses were taught by William Lacey
and Bruce Sage. These men were specialists in
thermodynamics and I was fortunate enough to
be able to take, while employed by Chevron Re-
search, a graduate course in thermodynamics with
Bruce Sage. These professors taught me some of
the elegance and precision involved in thermo-
dynamic thinking and also introduced me to the
fascinating area of the interaction of chemical and
physical processes which constitutes the essence
of chemical engineering."
Of Dr. Lacey's continuing influence Smith
goes on to say: "Dr. Lacey, along with Walter
Dayhuff of Chevron Research, was instrumental
in teaching me how to write clearly. Dr. Lacey


was kind enough to review a chapter in my first
attempts to write a text on chemical engineering
thermodynamics. He found 37 split infinitives
which he brought to my attention in a gracious
way. Ever since a split infinitive has, figuratively,
jumped off the page at me."
He was equally "fortunate" in his professors
at MIT where, he says, "my interest in thermo-
dynamics was further enlarged by the experience
of taking graduate-level courses with Harold
Weber. The contrast in approach to the laws and
applications of thermodynamics from these two
institutions was, at first, perplexing and confus-
ing. With the passage of time, the strong points
of the methodical, careful approach at Cal Tech
and the thought-provoking, more applied meth-
ods at MIT provided a more complete and satis-
fying understanding of the subject.
"Thesis work with W.K. Lewis was an inspira-
tion. One of the most lasting benefits of this asso-
ciation was learning the importance of being able
to generate new ideas about chemical engineering
problems. Dr. Lewis was an inspiring innovator
and his students could not help but see the beauty
and advantages of innovative approaches to diffi-
cult problems. Dr. Lewis had no patience with
inferior work and transferred this characteristic
to his co-workers." The same might be said by
the students of Dr. Smith.
Joe Smith's influence on students is not lim-
ited, however, to those he individually reached in
lecture rooms or labs. Not only is he "the man
who wrote the book", he is also one of "the men
who built the College at UC Davis".
Roy Bainer, Dean Emeritus of the College of
Engineering at UC Davis, describes Joe's hercu-
lean efforts in these early years. "Hired as a
chemical engineer in the Department of Food
Science and Technology, Smith was a member of
the UC Davis faculty when the possibility of the
College was being considered. Local committees
were appointed to develop curricula in the main
disciplines, utilizing the engineering talent at
Davis. For example, Dr. Smith of Food Science
and Technology was asked to chair a Committee
for Chemical Engineering." When the local com-
mittee's report was accepted, chemical engineering
was approved as part of the College of Engineer-
ing, not, as at Berkeley, as part of the School of
Chemistry. To continue with Dr. Bainer's ac-
count: "This meant that Dr. Smith could serve
as first chairman of chemical engineering at Davis
and could become involved in hiring new faculty

Essie and Joe Smith enjoying a departmental
social function.

and the planning of the new building... The cur-
rent enrollment in engineering at Davis reached
1900 students twenty years after the Regents'
establishment of the College. Much of the credit
for success must go to the many contributions of
Dr. Smith".
Not only the College benefited from Joe
Smith's talents as an administrator. The chemical
engineering department owes its existence to his
persuasive advocacy. After discussing the struggle
to establish a separate department ("It became
clear that if chemical engineering were to develop
as a recognized discipline at Davis a separate
department was necessary"), Joe describes the
continuing struggle, familiar to chemical engi-
neering educators everywhere, the need for which
he attributes "to a lack of understanding of the
essence of chemical engineering": "After the
establishment of the department our worries about
developing a strong program were not completely
eliminated. It seemed a continual struggle dur-
ing the late sixties to justify inclusion of sufficient
credits in chemistry and separate courses in ther-
modynamics and fluid mechanics, distinct from
the existing courses in these subjects taken by all
other students in the College of Engineering."
Professor Smith worked equally hard on cam-
pus recruiting for his department. Contributing
to the struggles on behalf of the department in
those early days were Steve Whitaker and Dick
Bell, who joined it in 1964 and 1965, respectively;
Ben McCoy and Alan Jackman, both alumni of
Minnesota, arrived in 1967 and 1970; Ruben
Carbonell and Dave Ollis, both with backgrounds
at Princeton University, joined the department
Continued on page 47.


mn lecture 1


University of Minnesota
Minneapolis, MN 55455
WE ARE TOLD ON THE highest authority that "no
prophet is accepted in his own country". That
may well be, for the charge of the prophet is usu-
ally an uncomfortable message aimed at the con-
science rather than the intellect. Scholars fare
rather better. Their message is to the mind with
its rational and aesthetic sensibilities and very
happy it is when a man's work receives that quiet,
but coveted, comment that it is "true scholarship".
It is therefore a pleasure to recall that Princeton
did not fail to recognize one of its most distin-
guished professors,* and that the profession at
large will forever keep his name in verdant mem-
ory through the Wilhelm Award, designed to rec-
ognize significant and new contributions in the
field to which Wilhelm contributed so much.
Twelve years and more have passed since his
death and it seems not inappropriate to take this
opportunity to consider the nature of his contri-

*He was appointed the Henry Putnam University Professor
in 1968.

After a few years in chemical industry and a brief spell teaching
mathematics at the University of Edinburgh, Aris came to the Univer-
sity of Minnesota in 1958 and has been in its Department of ChE ever
since. He particularly appreciates the lively intellectual atmosphere of
the department and the excellent quality of students that it serves
and has endeavored to make some contribution to its good repute.
His research has centered on the mathematical models of chemical
engineering, particularly those of chemical reactors. He is currently
Regents' Professor of ChE at the University of Minnesota.

bution to the subject of chemical reaction engi-
neering. It would be impertinent for me to add
many personal reflections, for I knew him far
less well than many. But if I may be permitted
to be personal for a moment, I can illustrate how
his papers influenced at least one of the lesser
mortals who have climbed on his shoulders. I was
in the employ of Imperial Chemical Industries in
a Physical Chemistry Research Department when
I was taken off work on the theory of chroma-
tography, in which I was just beginning to find
some interest, and was ordered to "get up" re-
actor design. In those days basic design of their
converters was in the hands of mathematicians
(mathematicians were called mathematical physi-
cists and lumped with physical chemists) and
stalwarts like F. F. Snowdon in a so called
Technical Department. My predecessor, Doro-
thy Annable, had had notable success in using
Temkin's expression for ammonia synthesis.
In trying to learn what reactors were, it was
Wilhelm's papers to which I turned and that I
found the most enlightening. These included his
1943 paper on "Conduction, Convection and Heat
Release in Catalytic Converters" and his review
of "Rate Processes and their Application to Re-
actor Design" (1949c). His paper on the measure-
ment of diffusion in beds of solids by a frequency
response technique (1953a) linked with one of
the approaches that I had used in connection with
the theory of chromatography. It is therefore to
Wilhelm in large measure and after him to
Hougen and Watson, that I own an introduction
to chemical reactor design. To analytic under-
standing of reactors it was the work of Amund-
son that later provided the great illumination. (I
cannot leave this personal intrusion without ac-
knowledging my further indebtedness to Wilhelm.
In 1955 I was fortunate enough to catch his atten-
tion with some early work on optimal design and
it was his good offices that obtained for me my
first research grant-from the Research Corpora-
tion, to whom he was an advisor. I am sure that
I am far from being alone among those whose

Copyright ChE Division, ASEE, 1983



from the Wilhelm Lectures at Princeton

early efforts have been encouraged in this typically
self-effacing way by a truly great scholar.)

In trying to discern the current of Wilhelm's
thoughts on the great questions of chemical engi-
neering, it will be useful first to discover the main
lines of his work. I will consider only those publi-
cations that have to do with reaction, leaving aside
some very interesting but marginally relevant
matters that he took up in other papers. Thus, I
will not refer to his work on fluid flow and vis-
cometry in 1939 or on the rate of solution in
1941, .some questions of textile drying in the late
40's or of parametric pumping in his last years.
Of these miscellaneous topics his discussion of
flames is really the only one close to our subject.
In a certain sense all Wilhelm's work served
the general problem of the design of reactors on
an a priori or scientific basis. Specifically there
is a series of general, or review, papers* (1949c,
1950, 1951a, 1953b,c, 1954) that lead up to the
general survey of 1962. On more specialist topics,
we note two papers on the tubular reactor, the
first (1950e) a kinetic determination, and the
second (1956d) a discussion of the effect of a
viscous profile on the conversion. In dealing with
packed beds he quickly recognized the importance
of the interaction of physical and chemical factors
and there is a whole series of papers concerned
with questions of longitudinal dispersion (1953a,
1956a, 1957c), of heat and mass transfer (1943,
1950d, 1959, 1963b) and of the measurements of
fluctuations of concentration (1957a, 1960, 1963a).
He took an early and continued interest in flu-
idized beds (1948c, 1949a, 1950f, 1951b, 1952,
1956b, 1958b), and a somewhat lesser interest in
stirred vessels, though he discussed the fluctua-
tions in the stirred vessel (1963a) and was con-
cerned with the mass transfer of oxygen in a

*All references to Wilhelm's papers are by year and
letter in keeping with arrangement of his bibliography
in the Appendix. Other references are to be found in the
list just before the appendices.

In his discussion of
packed-bed reactors, the question
of lateral and longitudinal dispersion by the
circuitous paths that fluid elements take
in passing around the particles was
a constantly recurring theme.

fermentation reaction, (1950a, b). Though the
diffusion within the catalyst particle received only
one paper in its own right (1961), questions of
mass transfer are. the continual burden of his
interest in the reactor design problem as a whole.
Without attempting to be comprehensive I will
sketch the lines he followed in some of these
Let us take a particular component of Wil-
helm's work and examine it in more detail. In
his discussion of packed-bed reactors, the question
of lateral and longitudinal dispersion by the cir-
cuitous paths that fluid elements take in passing
around the particles was a constantly recurring
theme. His first paper on this was one with Ber-
nard (1950a). In it they solved the steady-state
equations for diffusion with plug-flow for the
boundary condition corresponding to a finite tube.
By injecting methylene blue from a point source
on the axis and comparing concentration profiles
they estimated a radial Peclet number ranging
from 8 to 15. The equations contain no hint of the
anisotropy that Wilhelm commented on in his
1962 review (see quote below), but this he was
to measure later with McHenry (1957c). Using
a sinusoidal wave input produced by an ingenious
mechanism which varied the individual flows of a
binary mixture without varying the total flow,
they found an axial Peclet number of 1.88 0.15
over the Reynolds number range of 100-400. It
was an extension to higher frequencies of earlier
work with Deisler (1953a), who had measured
both the axial and the internal diffusivity when
the pellets of the packing were porous. It comple-
mented and was complemented by some of Praus-
nitz' work (1956c) on fluctuations of concentra-
tion in a packed bed.


It is instructive to see how Wilhelm brought
this all together in his plenary lecture in 1962.
Because it shows the mode of his thought and his
manner of expression so well, I will take the lib-
erty of quoting five paragraphs in extenso.

We now come to an interesting question: Why do
the magnitudes of the dispersive Peclet numbers,
although constant, have different values in the fully
developed mixing regimes Why is the system aniso-
tropic? The radial Peclet number is noted to have a
numerical value of about twelve and that of the axial
group, of about two. These values would appear to
arise naturally from the properties of the cell mixing
model and the ideas involved merit brief discussion.
The lateral dispersion effect is consistent with a
random walk, a Galton quincunx description. As tracer
material flows from one mixing cell level to the next
in the tube, a succession of lateral displacements must
occur because of the physical presence of the particles.
The thorough mixing at each stage assures that the
direction of travel of a tracer element shall be sta-

in contrast... the modern scientific
design method tries to make as much progress
with a model of the system in projecting forward the
behavior at each stage.

tistically independent of the direction of lateral motion
in the previous cell. Through a straightforward ele-
mentary statistical development of the mean square
displacement of tracer substance after passage through
n layers of mixing cells, it may be shown that the
lateral Peclet group should be constant (this was veri-
fied experimentally). It was also shown to depend in
its absolute value on (a) the average scale or side-
step distance in walking around a particle and on (b)
the number of layers of mixing cells between source
and measuring point. If we take the number of mixing
cells in the axial direction to be of the order of the
number of particles in that direction and the average
scale of individual lateral displacements to be a frac-
tion of the particle diameter, an estimate of the value
of the lateral Peclet number is made which closely
approximates the experimental value. Prausniz* in a
detailed study of concentration fluctuations in a tracer
plume, rather than of the time-average values of
which we have been speaking, was able to determine
directly that the scale of lateral displacement mixing
was one-quarter of a particle diameter. In short, an
elementary random walk analysis serves to describe
the observed lateral dispersion effects. Such descrip-
tion has been made by a number of investigators in-
cluding Baron1, Boreskov2, Latinen3 and Singer4.
As we turn to the axial dispersion process, a key
point, as suggested by Beek5 is whether, in a sequence
of interconnecting and thoroughly mixed stages, there
is a significant amount of fluid backflow against the
main stream. The state of the reactor in a local region

*See 1957a.

will depend greatly on the presence or absence of
such flows. For these to be effective it is necessary for
the distance that should be traversed by backflow ele-
ments of fluid to be larger than the average size of
the particle-associated mixing stages or permanent
eddies. Such backflow streamers would of course mix
with the fluid in stages directly upstream and thus
would be the means for conveying a signal against
the main stream. Such random backflow motions would
provide the physical basis for a purely diffusive trans-
port. If backflow is a predominant effect the form of
the system may be closely approximated by a normal
differential diffusion equation and characterized by
a meaningful diffusion constant.
However, in the absence of backflow, and when the
predominant dispersion effect is due to mixing in stages,
reactor design should be accomplished by the solution
of the difference equations which represent the discrete
nature of the physical situation. Under the latter
circumstances a diffusion constant in the local region
of a well-stirred eddy must approach a value of zero
and the designation of the diffusion equation to de-
scribe the state becomes meaningless. Although there
have been no direct experiments devised, to date, to
measure the extent of backmixing, the value of the
axial Peclet number of two as determined by measure-
ments in the limits of deep beds gives strong evidence
that stage-wise mixing rather than backmixing is
the predominant dispersion action. It will be recalled
that the Peclet number is an inverse proportionality
constant in the Einstein equation; it represents the
spreading rate of a tracer pulse after a large number
of mixing cells have been traversed if each cell is one
particle diameter in depth. Aris and Amundson6 an-
alysed the axial dispersion property by tracing the
changes in the probability density function with dis-
tance into the bed. This function, it will be recalled,
is the probability that a molecule, introduced into the
first mixer at the bed entrance at time, T = 0, will be
located in the nth layer at time, T. As expected, a
Poisson distribution represents conditions in the ini-
tial discrete stages at the bed entrance; with increas-
ing bed depth and number of mixing stages a normal
distribution is approached for which the diffusion equa-
tion becomes an appropriate description. A substan-
tial approach toward this limit occurs in a depth of
about twenty particles. The first authors to propose
the cell mixing model, Kramers and Alberda7, an-
alysed the same problem by making theoretical com-
parisons of the frequency response properties of the
solutions of difference and of differential equations
at various bed depths for a forcing sinusoidal tracer
concentration input.
In order to summarize this point, we can state that
within the specified high-velocity fluid mechanical
regime a single model now exists which describes
satisfactorily both lateral and axial dispersion effects
in packed bed reactors. However, if the only advan-
tage in the prescription of a difference equation formu-
lation were to describe properly reaction behaviour
in the regions of the bed entrance the accomplishments
would be small at best. The significant gain is derived
from a mathematical simplification which has a de-


Another paper that illustrates his paene-Taylorian combination
of theory and experiment is his discussion coauthoredd with Cleland) of
a viscous flow profile on the yield of a tubular reactor.

cisive advantage as we look forward to the detailed
design of reactors for realistic chemical kinetics. As
stated previously, if backmixing were important in
contributing to axial dispersion, coupled non-linear
differential equation heat and mass balances would be
the appropriate system descriptions and the state
within the reactor would depend on both the upstream
and downstream boundary conditions. This state would
lead to a mathematical situation that for a complete
model is analytically intractable and numerically be-
yond present digital solution capacities in a practical
sense. On the other hand, in the difference equation
stirred-tank model, only conditions at the bed inlet
need be satisfied and calculation may proceed down
the bed in a so-called step by step marching technique.
Of course he was standing at the brink of
some important numerical developments that
would have changed some of his last sentences.
Of this he was not unaware, though he does not
emphasize it, preferring to describe the cell model
of the packed bed that Lapidus and Deans [37]
had recently devised.
Another paper that illustrates his paene-
Taylorian combination of theory and experiment
is his discussion coauthoredd with Cleland) of
the effect of a viscous flow profile on the yield of
a tubular reactor (1956d). In it, an analysis of
the equations and computational solution were
combined with some simple and direct experiments
which were not only confirmed by the theory but
also served to show its limitations.

Let us try and locate Wilhelm's contribution
within the general course of the development of
chemical reaction engineering as a self-reflective
subdiscipline of chemical engineering in general
(see Table 1). Chemical reactors, of course, have
been built, and have often performed excellently,
from the earliest times. Their analysis and their
design, by other than ad hoc or trial and error
methods, however, is a fairly late phenomenon.
In the growth of chemical industry the standard
technique was to rely on scale-up by stages in
which a potentially useful chemical reaction found
in the laboratory would first of all be tried in one
or more intermediate sizes of apparatus (often
known as semi-technical stages) before being in-
corporated in a main-line plant. It was hoped that

some of the problems which might arise in the
larger scale would be discovered in the semi-
technical stage or stages and any difficulties over-
come in this way. In contrast to this the modern
scientific design method tries to make as much
progress with a model of the system in projecting
forward the behavior at each stage.
The interaction of physical and chemical effects
and their just disposition was recognized as one of
the keys to the understanding of reactors. Though
hinted at in a half-formed notion of Lomonosov,
the first definite applications of this was the work
of Nernst and Brunner [8] which recognized that
the rate of reaction at a heterogeneous surface
might be limited by the transfer through a stag-
nant film of liquid. By 1908 Bodenstein and Wol-
gast [10] had used a flow reactor for the homo-
geneous hydrogen-iodine reaction, but they were
only able to analyze certain limiting cases. Lang-
muir [9], in the same year, treated the problem of
simultaneous diffusion and reaction and obtained
the correct boundary conditions for this system.
This last was a point which Wilhelm was later
to investigate with Wehner in 1956.
In 1913 a physicist by the name of Ferencz
Jiittner [11] solved the equations for diffusion and
reaction in a porous catalyst but, so far as is
known, no notice was taken of his paper. Mean-
while, in connection with combustion limits, Taf-
fanel and LeFloch [12] had recognized that multi-
ple steady-state could arise though they were
more interested in the close approach of the heat
generation and rejection curves than in their inter-
section. However, it was Liljenroth [13] who, in
1918, first made this explicit in his discussion of
ammonia burners. Little seems to have been pub-
lished of an analytical nature about chemical re-
actors until the late 30's when the whole question
of design and performance was opened up by
Damk6hler [15] in his papers on "the influence of
flow, diffusion and heat conduction on the yield
of reactors." Almost simultaneously Thiele [16]
and Zeldovich [17] obtained the same essential re-
sult on the question of diffusion and reaction in
catalysts. Namely, that there was a dimensionless
parameter, of the nature of a (length)2 x (the
first order rate constant) / (the diffusivity), which
governed the seriousness of any diffusion limita-




.* Books and
Scc .' g S- Wilhelm's coauthor Monographs
Year Notable Papers (0) & interest (W) (B)

1904 Nernst/Brunner8

1908 Langmuir9
1909 Juttner"
1913 Taffanel/LeFlochl2
1918 Liljenroth'3
1934 Hottel/Tu/Davis14

1938 Damkohlers1
1939 Thiele'6/Zeldowitch1'
1941 Zeldowitch/Zysin'9
1945 Wagner20
1944-8 Denbigh21

1949 Wicke & Brotz22

1951 Hougen23
1953 van Heerden24
1955 Amundson/Bilous25

1957 Amundson/Aris27
1961 Carberry28Weiss/
Amundson/Schilson :30
Tinkler/Metzner :31

1963 Amundson/Schmitz32

Bischoff33 :Petersen34:






W* O W


(*denotes paper with RHW
as sole author).

RHW joined Princeton


Johnson/Wynkoop Collier


W* *:McCune
W Toner :Singer:Wynkoop
Bernardi: Bartholomew et al
W Valentine

W W*



W WB Wehner: Prausnitz:
Hanratty: Latinen: Cleland
W McHenry: Prausnitz
B Rice
B,W Hill

B Villet

IUPAC Lecture

WO B B W B Lamb: Manning

BW Blum

RHW's death

J. M. Smith

W. Britz, Birdfoot

Aris (Opt. Design)


Aris Denbigh:

Gavalas :Boudart

tion. It was left to Thiele however to see that this
modulus, which is now called after him, could be
incorporated in a scientific fudge-factor called the
effectiveness of the catalyst particle. This is the
ratio of the actual rate of reaction to that which

would obtain if there were no diffusion limitation
and it is, I think, a mark of true engineering
insight that Thiele should have seen the oppor-
tunity to wrap-up a lot of difficult calculations in
so simple a factor. This betrays an essential dif-




ference in outlook between the engineer and the
physicist, the latter portrayed in Jiittner's work.
By this time Wicke [16] was also beginning his
studies in adsorption, desorption, diffusion, and
their influence on chemical reaction, studies which
have continued until the present day. The question
of multiplicity of steady-states and, indeed, of
isolas was also investigated in Russia by Zeldo-
witch and Zysin [19] in a paper which has largely
been overlooked. (D. Luss drew it to my attention
and reference is made to it in Gray's valuable
chronology [39].
It was in the mid-forties that Denbigh [21]
built on his experience in industry and produced
a general analysis of the well-stirred reactor.
His was the first mathematical analysis of tran-
sient behavior and later he was a leader in some
aspects of optimal design. To the late forties be-
longs the confirmation of the diffusion limitation
of a reaction by Wicke and Br6tz [22] through the
observation that the apparent activation energy
was reduced by a factor of 2 at high temperatures.
An earlier result belongs to the 1930's when the
external film diffusion limitation was shown to
give an even greater reduction in the apparent
activation energy for the combustion of carbon
spheres [14].
Of course the need for various synthetic chem-
icals during the war had greatly spurred interest
in the general design of reaction and the book of
Hougen and Watson published in 1945 represents
the outcome of much practical experience. Watson
had gained a great reputation for the start-up of
the butadiene plants so necessary to the synthetic
rubber program. Hougen [23] was later to sum-
marize his experience in the 1951 Institute lecture
published as the first of the Chemical Engineer-
ing Progress Monographs. The 50's saw the pub-
lication of van Heerden's [24] paper on auto-
thermic reactors which, done in ignorance of
Liljenroth [13] and Zeldowitch [19], was very in-
fluential in bringing questions of uniqueness and
stability to the attention of a much wider public
than ever before. In particular, the nature of the
multiplicity of steady states and the associated
transient behavior was taken up by Amundson
who recognized it immediately as cognate with
the work on nonlinear differential equations that
had prospered so notably in the previous decade.
Amundson in the mid 50's was also concerned
with the sensitivity of such features as the hot-
spot in the exothermic tubular reactor and his
papers with Bilous [25] mark the founding of the

mathematical study of stability and sensitivity in
chemical reactors. Indeed, Amundson [32, 36] con-
tinued this work in the late 50's showing that
under the influence of control there could arise
limit cycles. He went further in the early 60's
to discuss two-phase reactors, polymerization and
was approaching the more difficult question in the

Wilhem stands out for his marriage of the
concept of a priori design, as he called it... and the
laboratory-scale experiments that would be needed
to understand the physical effects.

distributed system, the tubular reactor, at about
the time of Wilhelm's death.
The internal economy of the catalyst particle,
a question which Wilhelm was always aware of
and frequently refers to, though he did not con-
tribute much to the direct literature on this topic,
was also the subject of much research during this
period. In 1951 Wheeler had given a valuable re-
view of work on diffusion and reaction which
helped greatly in drawing attention to Thiele's
work (Advances in Catalysis 3, 249, Academic
Press, New York). He further discussed some
questions of selectivity in 1955 [26]. Just as there
had been a flurry of independent work in the
late 30's which resulted in the basic notion of the
effectiveness factor, so in the early 60's there were
almost simultaneous solutions to the nonisother-
mal problem. Of these Amundson and Schilson's
[30] went back to Schilson's PhD work in the mid
fifties and of which Wilhelm had probably learned
directly from Amundson. But almost simultane-
ously Carberry [28], Weisz and Hicks [29] and
Tinkler and Metzner [31] obtained essentially the
same result. It was the curves of Weisz and Hicks
which became so well known and served to remind
people that multiplicity of steady state could occur
within the catalyst particle itself. Another curi-
ous independent simutaneity occurred in 1965
when Bischoff [33], Petersen [34] and Aris [35]
found that the effectiveness factors for all shapes
and all kinetics could be brought together in an
asymptotic sense.

The plenary address to the International Union
of Pure and Applied Chemistry which Wilhelm
Continued on page 38.


M@12 n classroom


A Course in Industrial and Engineering Chemistry

Iowa State University
Ames, IA 50011

University of Delaware
Newark, DE 19711

C HEMISTRY IS AT THE heart of chemical engi-
neering practice. And chemistry-as the life-
blood of innovation and design-is intimately in-
volved in the creation of new technology. The
ability to integrate chemistry and engineering is
a hallmark of the chemical engineer. A course
taught to seniors and graduate students at the
University of Delaware and at Iowa State Uni-
versity is based on such a confluence of chemistry
and engineering. The theme of the course is pro-
cess conceptualization: the creation of chemical
processing routes. There are no new subjects in
the course; rather, information and ideas are
synthesized from organic and inorganic chemis-
try, stoichiometry, thermodynamics, kinetics, re-
actor design, physical chemistry, heat and mass
transfer, and fluid mechanics. This emphasis on
synthesis is the key both to creating processing
the routes and to demonstrating the interplay be-
tween chemistry and engineering.

H, ----- -



A \ /



This emphasis on new routes rather than old,
and synthesis rather than analysis, provides a course
with an orientation toward innovation and creativity.
Many aspects of the discussion of the processing
routes are . qualitative, but an exclusively
descriptive orientation is avoided.

The course is organized around a sequence
of chemical processes. The processing chemistry
and engineering are covered in detail but not to
the extent which occurs in a typical senior-level
design course. Since all processes depend on both
chemical and engineering insights, it is the com-
bination of these insights which is prime. Some
processes rely on well-understood routes; how-
ever, the discovery of new routes is emphasized.
This emphasis on new routes rather than old-
and synthesis rather than analysis-provides a
course with an orientation toward innovation and
creativity. Many aspects of the discussion of the
processing routes are necessarily qualitative, but
an exclusively descriptive orientation is avoided.
Quantitative evaluation and reasoning are es-
sential to understanding industrial chemistry.
The format for teaching the course is uncon-
ventional. But we have discovered that students
can be led to view chemistry and engineering as
being interdependent-and often inseparable.
Here we present the content of the course and the
teaching method which has been developed during
the past four years.

The course in industrial and engineering
chemistry is organized on a case-study approach,
and the processes are shown in Fig. 1. This master
flowchart presents a large portion of the struc-
ture of the chemical and petroleum industries as
an interrelated whole. This approach increases the
students' recognition of specific chemicals as raw
materials, intermediates, end-use products, by-
products, or wastes. Students particularly need
Copyright ChE Division, ASEE, 1982


information about existing industrial paths and
respond eagerly to the real-world orientation. A
sequential presentation reinforces the strong
interdependence of industrial processes.
The case studies were also chosen on the basis
that they illustrate many important aspects of in-
dustrial chemistry. Ammonia and ammonium
phosphate production and chlorine production are
representative of inorganic syntheses. Classic
organic reactions are involved in the production
of nitrobenzene and aniline. Some reactions rely
on relatively unspecific homogeneous free radical
chemistry, such as ethane cracking to ethylene;
reactor design is relatively simple. In contrast,
the conversion of ethylene to ethylene oxide
relies on the complexities of catalytic selec-
tive oxidation. Most industrial processes in-
volve catalysts, and the integration of chemistry
and engineering can be vividly demonstrated for
these processes. In addition, a discussion of ma-
terials of construction and catalyst properties
(surface area, porosity, dispersion) permits
aspects of solid state chemistry to be introduced.
The production of poly (vinyl acetate) is an excel-
lent example of the combination of polymer
chemistry and engineering. Environmental
chemistry is encountered in the synthesis of
chlorinated hydrocarbons and in the production
of synthesis gas from coal. Thus, from the
beginning, the overall flowchart provides a basis

Glenn D. Schrader received his
B. S. degree in Chemical Engineer-
ing from Iowa State University in
1972. He then attended the Univer-
sity of Wisconsin-Madison, where
he received his doctoral degree. In
1976, he joined the faculty of the
University of Delaware, and be-
came a member of the Center for
Catalytic Science and Technology.
In 1980, Dr. Schrader returned to
Iowa State University as an Asso-
ciate Professor of Chemical Engi-
neering and as a Chemical Engineer
of the Ames Laboratory-U.S. De-
partment of Energy. Dr. Schrader's
research interests are in catalysis
by metal sulfides, metal oxides, and
transition metal complexes for applications in fuels and chemicals
production. He also serves as an industrial consult in these areas. (L)
R. L. Pigford was born in Meridian, Mississippi, and got his B.S.
at Mississippi State College in 1938. M.S. and Ph.D. followed from
the University of Illinois in 1940 and 1942. He worked for E. I. duPont
de Nemours and Company at the Experimental Station in Wilmington,
Delaware, until 1947 under T. H. Chilton. Then he replaced A. P.
Colburn as Chairman of the Chemical Engineering Department at the
University of Delaware, later becoming the first Colburn Professor.

for identifying the broad range of concepts which
are integrated in industrial practice.
However, the goal of the instructor is to stress
the impermanence of this overall scheme. Evolu-
tion and modifications in the flowchart are empha-
sized throughout the course. The student is en-
couraged to consider drastic changes-typically
leading to the disruption of an entire train of pro-
cesses. The motivation is frequently the need for
alternate sources of raw materials. Several spe-
cific examples can be chosen:
vinyl acetate production without the availability of
ethylene from refinery gas or natural gas
aniline production without the availability of benzene
derived from petroleum refining
methanol production without the availability of coal
or natural gas

The development of alternative industry-wide
routes requires basic information concerning
sources of elementary chemical constituents. It
is useful to preface the coverage of the overall
flowchart by a discussion of sources of carbon,
nitrogen, hydrogen, phosphorus, and chlorine.
The discussion is much more than a listing of
raw materials and their relative abundance.
Students are directed to critically evaluate the
sources-primarily from a chemical viewpoint.
This exercise provides a good opportunity to re-
view elementary chemical concepts such as oxida-
tion state and stability. As an example, although

Between 1966 and 1975 he was Professor of Chemical Engineering
at the University of California in Berkeley. (C)
Bruce C. Gates is Director of the Center for Catalytic Science and
Technology and Professor of Chemical Engineering at the University
of Delaware. His research group includes graduate students in chemi-
cal engineering and postdoctoral fellows with backgrounds in physical,
organometallic, and surface chemistry. They are studying catalysis by
supported metal complexes and clusters, zeolite-entrapped metal
clusters, supported superacids, and metal sulfides. The research is
directed toward fundamental characterization of industrial catalysts
and reactions, and conception and design of new catalysts. (R)


Most industrial processes involve catalysts, and the integration of chemistry and engineering
can be vividly demonstrated for these processes. In addition, a discussion of materials of construction and
catalyst properties .. permits aspects of solid state chemistry to be introduced.

students may be surprised to discover the over-
whelming abundance of carbon in precipitate or
solid form, they quickly distinguish the utility of
carbonates versus carbohydrates; a similar com-
parison can be made between gaseous nitrogen
and nitrides. Frequently, a change in raw ma-
terial may mean the introduction of new impuri-
ties into a processing route. Thus, lignite might
be considered as a replacement for coal or natural
gas, as a source of carbon monoxide. Lignite, how-
ever, contains considerable amounts of water,
with many consequences in the downstream pro-
cessing. In proposing oil shale as an alternative
source of benzene, the relative amounts of sulfur,
nitrogen, and heavy metals change, all with
enormous consequences in the catalytic chemistry
of reforming.


Flowcharting is the primary mode used to
demonstrate process conceptualization. The
ability to think in terms of flowcharts (including
their creation, modification, and evaluation) is
emphasized throughout the course.
Flowcharts are introduced at various levels of
detail. The detail is determined by the utility of
the flowchart, i.e., to quickly express a new chemi-
cal route and possible alternatives or to describe
a key scheme of separation processes. In other
cases, crucial aspects of processing may not appear
until recycle streams are drawn or heat transfer
operations are introduced.
The most elementary form of a flowchart ex-
presses the stoichiometry of the process: raw ma-
terials transformed into products. The essence of
the chemical innovation is frequently contained
within this simplest stoichiometry. We have found
that students are adept at suggesting alternatives.
But, sometimes, these "inventions" must be pre-
sented to the student with development of the re-
mainder of the flowchart being an assignment.
In writing a simplified stoichiometry, we stress
that industrial processes rarely involve only the
nominal reaction. During the initial portion of
the course, the production of side products, wastes,
and pollutants is emphasized. Students are re-
minded of the limitations in the purity of their

raw materials, since the fate of impurities fre-
quently is not expressed in the simplest stoichio-
metry. The dire consequences of ignoring impuri-
ties can be illustrated by "well-developed" pro-
cesses which are fatally flawed by catalyst poisons
or sources of water or oxygen.
Consideration of specific unit operations adds
the next level of detail to the flowcharts. The most
frequently expressed initial combination of unit
operations involves reactors and separation units.
Even the most elementary flowchart is influenced
by the chemistry of the process. Generally, the
selection of specific unit operations proceeds from
the restrictions dictated by the chemistry. To il-
lustrate these ideas, the instructor can lead the
student to think of fluidized bed reactors for
highly exothermic oxidations and fixed bed re-
actors for reforming petroleum distillates. Re-
actor systems for synthesis gas reactions using
homogeneous catalysis provide a vivid contrast to
those using heterogeneous catalysis. Separation
processes can be dominated by the chemical route
chosen initially. Methanol carbonylation requires
separation of a heavy catalyst, a light promoter,
and an intermediate product.
Conversely, engineering requirements often
strongly restrict the chemistry which can be em-
ployed. This interplay between chemistry and
engineering is presented by the modification of
flowcharts. Typical alterations include the ad-
dition of recycle streams and by-product purges.
Heat and mass transfer requirements may spe-
cifically limit the chemistry which can be at-
tempted. The students can also be confronted with
reactions such as the production of methanol, for
which equilibrium limitations require low temper-
atures, but for which the kinetics demands higher
temperatures. The selection of specific unit
operations may also be affected by safety or en-
vironmental restrictions. For example, the con-
ceptualization of a process for nitrating benzene
is dominated by the recognition of the high exo-
thermicity of the reaction. Safety precautions,
which are necessary to prevent reaction runaway
and explosion, strongly influence reactor design.
Ethylene chloride production is dominated in some
respects by environmental concerns.
This approach emphasizes the creative develop-


ment of chemical processing routes rather than a
descriptive presentation-and invention rather
than memorization of the details of the process.
The challenge to the instructor is to create a class-
room atmosphere conducive to such an orientation.
In the early portion of the course, the develop-
ment of process flowcharts is fostered by having
the students pose questions concerning the basic
chemistry and engineering. Students are en-
couraged to ask a wide range of questions, in-
cluding those that they or the instructor may be
unable to answer fully. At this point the students'
curiosity is undisciplined; this approach appears
to be essential for providing an open "free-wheel-
ing" orientation towards creativity. A typical set
of questions posed by a class during the early
portion of the course is shown in Table 1.
Flowcharts are constructed as these questions
are considered. Initially, the students have diffi-
culty in posing the "right" questions. However,
skill in asking a sequence of questions can be de-
veloped by the instructor. Initially, a large set of
questions is accepted, but an ordering and ranking
is essential. Students are confronted with the need
for a sequence in their consideration of the pro-
cessing routes. Thus, consideration of the thermo-
dynamic effects of temperature and pressure must
be preceded by stoichiometry. Questions concern-
ing kinetics of specific reactions should not arise
until thermodynamic feasibility is examined.
Some questions can be shown to lead naturally to
others: knowledge of the heat of reaction raises
the question of heat transfer in reactors. The in-
structor can guide this thought process through
examples and illustrations offered by the case
studies. The complexity of the questions generally
parallels the detail of the flow chart, and there-
fore more explicit questions can be asked as the
conception of the process develops. From our ex-

Student Questions on Developing a Flowchart for
Production of Methanol from CO and H2
1. Is a catalyst needed? What is the catalyst?
2. Does the reaction require heat?
3. How can heat transfer be performed?
4. What temperature and pressure should be used?
5. Are there important side reactions?
6. What yields can be obtained?
7. Can a fluidized or a fixed bed reactor be used?
8. Are reaction rates available?
9. What is the purity of the feed?
10. What product separations must be performed?
11. What materials of construction can be used?

perience, a specific sequence of questions often
evolves, dealing with the following subjects:
stoichiometry of proposed reactions and potential
side reactions
reactor design
mass transfer and heat transfer
fluid flow
environmental restrictions
The students recognize many of these as subjects
of chemical engineering. A more mature set of
questions posed by a class later in the course is
shown in Table 2.

Student Questions on Developing a Flowchart for
Petroleum Reforming
1. What reactions are involved in reforming?
2. What is the chemical composition of the feed stream?
3. What concentration of impurities can be accepted?
4. What is the heat of reaction?
5. How do temperature & pressure affect the equilibrium
6. What are the rates of the reactions?
7. Can a catalyst be used to improve the rates of re-
action or selectivity?
8. Can the catalyst be poisoned or deactivated?
9. What is the role of hydrogen in the process?
10. Should a fixed bed or fluidized bed reactor be used?

However, the students are not encouraged to
view any one sequence as a pattern for innovation
or even for the duplication of existing processes.
Examples are encountered which apparently in-
volve complete reversals of the sequence, such as
benzene nitration or hydrocarbon chlorination.


The discussion of the basic chemistry of in-
dustrial processes requires a qualitative knowledge
and understanding of chemical composition, rela-
tive stability, solubility, catalytic activity, and
other intuitive concepts related to structure and
reactivity. Although students have previously
studied many of these ideas, their general useful-
ness is usually not appreciated and their integra-
tion into chemical engineering calculations is
normally missing. We have deliberately attempted
to incorporate these concepts into the course in a
quantitative manner.
The complexity of industrial processes fre-
Continued on page 42.


laboratory i



University of Colorado
Boulder, CO 80309

T HE PROBLEMS ASSOCIATED with fossil fuel sup-
plies and the associated escalation in price has
sparked renewed interest in solar energy as an
alternate fuel source. One application of solar en-
ergy which has received considerable interest is
domestic hot water.
This paper presents an undergraduate labora-
tory experiment which uses a solar collector to
heat water for domestic usage. The water is cir-
culated by natural convection so no pumping is
The experimental apparatus is relatively inex-
pensive to acquire. The solar collector can be built
on site to reduce costs. This can be accomplished
by students with supervision for which they can
receive credit for construction and start-up of the
The quality of the data produced is very good.
This enhances the educational experience since
the students are more enthusiastic about an ex-
periment which "works." The students can also
sample the hot water produced to "experience"
the results.

Rich Noble received his B.E. degree in 1968 and M.E. degree in
1969 from Stevens Institute of Technology. In 1976, he received his
Ph.D. degree from the University of California, Davis. His current re-
search interests include facilitated transport in liquid membranes,
transient heat transfer, and problem solving skills.

Hot Water

1o Ambient


Inside Building


Solar Cc
Outside Building

FIGURE 1. Schematic diagram of solar water heater
with thermocouple locations.

The educational objectives of this experiment
are the following. 1) to expose the students to
the principles of solar energy collection, 2) to
demonstrate the use of solar energy to heat water,
3) to demonstrate the use of natural convection
to move fluid, and 4) to use experimental data ob-
tained to do a preliminary design for a feasible
household system.

The experimental apparatus is shown sche-
matically in Fig. 1. The students take temperature
readings at least hourly for the 10 locations shown
and also the short-wave solar radiation flux normal
to the collector surface. A short-wave radiometer
is attached to the collector with a millivolt read-
out. This is done for the period 8 am to 5 pm and
can usually be accomplished quite easily by the
students in the group performing the experiments.
They can take turns for each hour and obtain the
data between classes if necessary. One set of data
can be obtained in 2 or 3 minutes with the aid of
a digital thermometer with a multiple dial so the
student simply dials each thermocouple reading
and writes down the temperature.

Copyright ChE Division, ASEE, 1983


The students are asked to determine:

Overall system efficiency.
Overall heat-transfer coefficient for the collector.
Overall heat-transfer coefficient for the heating coil.
Collector and heating coil surface area needed to pro-
vide 0.189m3 (50 gal) of water daily at 500C which is
initially at 150C.


A diagram showing the construction of the
solar collector is shown in Fig. 2. The solar panel
is an aluminum plate (2.11m x 1.14m) which is
painted black. Attached to the collector are 9 cop-
per tubes (1.59cm O.D.) spaced 12.7cm apart
("a" in Fig. 2) and also painted black. The copper
tubes are connected to a copper manifold (3.81cm
O.D.) at the top and bottom. To avoid relative
heat expansion effects, it is recommended that the
plate and all tubing be made of the same material.
Below the collector is 0.15m of insulation. The
collector and insulation are housed in a wooden
frame (3.81cm thick) which is 1.22m x 2.43m x
0.24m. A piece of plate glass sits above the collec-
tor to prevent long-wave radiation from escaping.
The angle of the collector relative to the ground
surface can be adjusted using a support frame
with adjustable support arms on the top and a
hinge on the bottom of the collector.
The return pipe from the heating coil should

1.22m n
\ -- 1.14 m



Plate Glass-

Solar Collector -T
Insulation 0.15 m
FIGURE 2. Solar collector

This paper presents an ... experiment
which used a solar collector to heat water for
domestic usage. The water is circulated by natural
convection so no pumping is required.

be downward sloping to the collector to aid the
natural convective fluid motion. It is initially rec-
ommended that the angle of the collector be set
at latitude + 10 [1].
Depending on climate, it is desirable to insulate
all piping and the hot water storage tank to mini-
mize losses. Also, in cold climates, an anti-freeze
solution can be used as the working fluid since the
heat transfer is through a closed loop.


The data are analyzed using the following pro-
cedure: First, the heat transferred from the heat-
ing coil to the water in the tank is calculated.
qt = mCp( At (1)

where: qt


= heat transferred to the water in
tank (W)
= mass of stored water (kg)
= isobaric heat capacity of water

AT ) = change in water temperature in
tank over time interval At

Measurement of the short-wave solar radiation
incident on the solar collector q1 determines the
heat input to the system.
Therefore, the thermal efficiency of the system
(1) can be determined.

qt (2)
This assumes that the system has no other heat
gains or losses. If the storage tank and piping are
well-insulated, this is a good assumption.
An overall heat-transfer coefficient for the
solar collection (U,) can be calculated as follows

A (A) (3)
A, (AT) LA

where: A,

= surface area of the collector


TABLE 1. Experimental Data

Temperature (oC) Reading
TIME T1 T2 T3 T, T5 Te T7 T8 T9 T10 (mV)

8:00 3 5 7 8 23 17 21 19 24 8 3
9:00 13 35 41 29 16 22 21 18 24 12 8.5
10:00 23 63 71 57 54 24 22 19 24 15 13.6
11:00 25 79 88 68 66 25 24 19 24 16 16.5
12:00 26 74 83 67 66 26 24 20 25 16 19.5
1:00 25 36 44 46 46 26 24 20 26 15 4.5
2:00 27 59 68 60 60 27 26 20 26 19 19
3:00 29 83 92 70 69 29 29 21 27 23 17.5
4:00 29 72 78 61 59 29 27 22 28 23 12.5
5:00 30 59 68 61 61 30 27 23 28 23 9.0

Tank Capacity: 0.454m8 (120 gal): Area of Tank Coil 1.86 m2: Radiometer Calibration constant: l(mW/cm2) = 0.272mV
Area of Solar Panel: 2.38m2

(AT)LM = log-mean temperature differ-
ence (K)
(T3- T4) (T2- T1)
(T3- T4)
(T2- T)
The subscripts on the temperatures refer to the
locations shown in Fig. 1.
Similarly the overall heat-treansfer coefficient
for the heating coil (Ut) can be determined.

Ut = t (4)

where: At = surface area of coil (m2)
(AT) (Tj-Tg) (T5-T9)
In (T( T,)
(T, T,)

By scanning the data in Table 1, it becomes
apparent that the collector does not start supply-
ing any appreciable heat to the storage tank until

approximately 10 am. This is the time that the
storage tank outlet temperature (Te) equals the
storage tank temperature (T,). Before this time
T, is greater than T6 and the storage tank is heat-
ing the fluid in the coil. Therefore, using 10 am
as a starting point for heating, the hourly data
was sub-divided into periods where there was a
rise in T,. The average values of the data during
these periods is shown in Table 2.
Sample calculations for the 10 am to 12 noon
time period are shown below.

qi = (radiation flux) (collector surface area)
16.5mV x 1(mW/cm2) 1W
0.272mV x10mW
x 2.38 x 104cm2 = 1.45 x 103 W
qt = mC, ( ) = 0.454m3 x 996.4 x 4.183
At m
J 1K 1 hr
x 103 x x 2.63 x 102W
kg K 2 hrs 3600s
qt 2.63 x 102W 0
S q 1.45 x 10W .1

TABLE 2. Averaged Data

Avg ation
Avg Temperatures (OC) mV Rate
TIME PERIOD 1 2 3 4 5 6 7 8 9 10 Reading (kW)

10:00 12:00 24.7 72 80.7 64 62 25 23.3 19.3 24.3 15.7 16.5 1.45
12:00 1:00 25.5 55 63.5 56.5 56 26 24 20 25.5 15.5 12.0 1.05
1:00 3:00 27 59.3 68 58.7 58.3 27.3 26.3 20.3 26.3 19 13.7 1.20
3:00 5:00 29.3 71.3 79.3 64 63 29.3 27.7 22 27.7 23 13.0 1.14


Calculated Results

TIME qi qi Uc Ui
PERIOD (kW) (kW) (%) (W/m2K) (W/m2*K)

10:00 12:00 1.45 0.263 18 20.6 15.2
12:00 1:00 1.05 0.524 50 28.1 38.6
1:00 3:00 1.20 0.262 22 33.8 15.7
3:00 5:00 1.14 0.262 23 18.0 12.9

q= 1.45 x 103W
c Ae (AT) LM 2.38m2
47.3 W
x = 20.6
(16.7 47.3) K m2.K
Sqt 2.63 x 102W
Ut At(AT)L,, 1.86m2
37.7 W
(0.7 37.7) K m2 K
Results for all time periods are shown in Table 3.
Sizing requirements for obtaining 0.189m3 (50
gal) of water at 500 which is initially at 15C
are shown below.
It is assumed that the actual time period for
heat transfer is 8 hours. This assumption is due
to the fact that the sun did not set until 7 pm and
the collector surface was still quite hot at 5 pm.
AT kg
qt = mC,( A ) t = 0.189m3 x 99.64 x 4.183
At m3
J 35K 1 hr
x 103 x 35 x hr 9.57x 102W
kg'K 8 hrs 3600s
qt 9.57 x 102W
qi --= = 3.81 x 103W
Assuming that the heat transfer rate is di-
rectly proportional to surface area
At 957W (1.86m2) = 5.93m2
3.81 x 103W
A = x 10 (2.38m2) = 7.37m2
e 1.23 x 103W

The average value of qt is 300W and the aver-
age value of qi is 1230W for the calculated results.

A laboratory experiment to demonstrate the
feasibility of using solar energy to heat has been

discussed. The working fluid is moved by natural
convection so no pumps are required. The experi-
ment exposed students to the principles of solar
energy and natural convection. The experiment
also demonstrated the use of solar energy to heat
water. Students used data obtained in the experi-
ment to do preliminary design calculations to size
equipment for a domestic hot water installation.
The experimental apparatus is simple in de-
sign and operation. The students can take data
quickly and easily. The experiment also performed
as desired (it works) so students can see the per-
formance. O

1. "Energy Primer: Solar, Water, Wind, and Biofuels,"
Portola Institute, Menlo Park, California, 1974, pp.
2. Daniels, Farrington, "Direct Use of the Sun's En-
ergy," Ballantine Books, 1974.
3. "Solar Hot Water and Your Home," National Solar
Heating and Cooling Information Center, Rockville,

9a Memo'uam

Ralph E. Peck
Ralph E. Peck, professor emeritus of chemical
engineering at the Illinois Institute of Technology,
died November 6, 1982, at the age of 71.
Dr. Peck began at the institute in 1939 and
was chairman of the chemical engineering depart-
ment from 1953 to 1967. He was internationally
recognized as an expert in drying, and was active
as a consultant for industry and government. He
taught in Israel, Brazil, Algeria and Korea, and
established a department of chemical engineering
at India's University of Punjab.
Dr. Peck was involved in peace and nuclear
arms freeze organizations and was sometimes
called on by the U.S. government to disarm chemi-
cal gas weapons.
He was a member of the Ethical Humanist
Society of Chicago. He was honored for teaching
excellence from the institute and from ASEE and
was a fellow of the AIChE. An annual Ralph Peck
lectureship was established in 1973 at Illinois
Institute of Technology.
His work after his retirement in 1977 led to
his invention of a process for removing sulfur
oxide gases from burning coal and converting
them to high-grade fertilization. O


I Curriculum


University of Maine at Orono
Orono, ME 04469

T HE PRACTICE OF PROCESS control in the chemical
processing industries is undergoing rapid tech-
nological change. The availability of powerful and
inexpensive computers and micro-processor based
controllers has made the use of sophisticated con-
trol algorithms routinely possible and economi-
cally necessary. Unfortunately, the supply of com-
petent process control engineers lags far behind
the demand produced by the available technology
and potential applications. The typical B. S. chem-
ical engineering graduate has been exposed to no
process control beyond a first course in process
dynamics. These graduates are completely unfa-
miliar with the concepts of "modern" or multi-
variable control. Graduate work does not meet
this need either, since there are few graduate
students in chemical engineering and fewer still
who are interested in studying advanced process
control. The relationship between processes and
their control systems is becoming increasingly
sophisticated. Hence, there is an obvious need for
control expertise in the chemical engineering pro-
fession which is not being met by the conventional
chemical engineering curriculum [1,2].
Two factors have made it possible to initiate
a course sequence in process control in the chem-
ical engineering department at the University of
Maine. The first was a bold change in the curric-
ulum in the department. A program was instituted
in which a student must choose an area of special
concentration. Current options are process control,
pulp and paper, polymers, honors, or, in excep-
tional cases, an individualized program. Each op-
tion consists of at least four courses at the junior
and senior levels. The second is that there are
two of us who are interested and competent in
computers and process control. This allows us to
overcome the very real barrier imposed by the
"critical mass effect" as described by Waller [2].
The Process Control Option is a dramatic

Copyright ChE Division, ASEE, 1983

break with the traditional teaching of control
theory in this department, and we believe that it
is a unique program at the undergraduate level
in chemical engineering. Teaching in process con-
trol formerly consisted of an introductory process
dynamics course for all students, a graduate
course when there was sufficient demand, and an
occasional special undergraduate course in re-
sponse to demand. We now provide the coordi-
nated sequence which is described in this paper.
One of us has had industrial experience in
process control, and the other is knowledgable in
real-time computer hardware, software, and inter-
facing. Neither of us has much interest in pure
theory divorced from application. Hence, the en-


John C. Hassler is currently an Associate Professor of Chemical
Engineering at the University of Maine at Orono. His degrees are in
physical chemistry from Kansas State University. He spent several
years in the "post-doc holding pattern," including four years in the
Electrical Engineering Department at the University of Illinois, work-
ing on lasers. He joined the Chemical Engineering faculty at Virginia
Polytechnic Institute and State University in Blacksburg, Virginia, in
1972, and moved to Maine in 1977. His research interests are process
instrumentation, modeling, and control, with an emphasis on the
hardware and software involved in the application of computers to
real-time problems. (L)
Kenneth I. Mumme' is an Associate Professor of Chemical Engineer-
ing at the University of Maine-Orono. He received a B.S. in physics
from Lawrence College (1954) and the M.S. and Ph.D. degrees in
Chemical Engineering from the University of Maine (1970). He worked
for eight years in the pulp and paper industry and has spent two
years doing research at the Institute for Engineering Cybernetics in
Trondheim, Norway. His research, publications, and patents are in the
general area of instrumentation, process control, and system mod-
elling. (R)


tire sequence tends to be strongly applied in na-


This is the typical required course for all students in
the department. It covers LaPlace transform theory, trans-
fer functions, PID control, etc. The text in current use is
by Luyben [3]. Because this course is required of all stu-
dents, it is not actually part of the Process Control option,
although it is a prerequisite for the option.

This course considers single-input single-output (SISO)
systems using the Z-transform. The text used is by Kuo
[4]. The course covers all of the sections of the text which
do not involve state variables. (The reason for this rather
uncomfortable division of material is the scheduling prob-
lem created by our coop program. This will be discussed
later.) The use of fully interactive computer methods for
design is emphasized. Many of the "standard" algorithms,
such as the Smith Predictor, the Dahlin algorithm, etc.
are presented and compared with direct Z-transform de-
sign methods.
Unfortunately, there is no formal laboratory with this
course. However, we both feel very strongly that practical
experience is an essential part of learning, so we "steal"
some laboratory time by omitting a certain number of
lecture periods. Although we continue to look for good
experiments, the system used this past year worked very
well. It consists of a series of interacting tank levels, with
the series ranging from one to four tanks. The levels are
controlled directly by a PDP 11/60 computer programmed
in FORTRAN. The students were asked to identify the
process by any means they wished and then to control the
process on the basis of one of the design methods learned
in the lectures. For example, this year they were asked
to design a digital PID-exponential filter controller for a
third order system, a dead beat controller for a first order
system, and to apply any technique they wished, other
than PID, to a third order system.
The experiments can take a long time, far more than
the class time allotted, for several reasons. One group
was very slow because of poor planning. Another group
got very interested, tried all sorts of control schemes, and
produced a truly excellent report. All of the students
agreed later that the laboratory, in spite of the great
amount of time spent on the experiment, was an absolutely
essential part of the course.

This course concentrates on multivariable systems. The
text in current use is the portion of Kuo [4] not covered in
ChE 154. Virtually none of the students who enter this
course have taken mathematics beyond the normal re-
quirement of 16 semester hours. Thus, they know little
about linear algebra and are inexperienced in the use of
the computer for matrix manipulations. Therefore, some

The Process Control Option is
a dramatic break with the traditional
teaching of control theory in this department,
and we believe that it is a unique program
at the undergraduate level in
chemical engineering.

time must be used to remedy these deficiencies. Once this
has been accomplished, we can proceed to the representa-
tion of systems by state variables. This leads naturally
through such concepts as controllability, observability,
multivariable systems, and control of linear (or linearized)
multivariable systems.

The major emphasis of this course is on system identi-
fication through filtering and estimation. A portion of
this course must be devoted to developing the practical
concepts of probability and stochastic systems. Following
this, we consider system identification through techniques
such as the Kalman filter, least squares, maximum likeli-
hood, etc. Applications of these techniques are examined
through case studies, the modelling and control of esti-
mated parameter systems, feedforward systems, model
reference systems, and adaptive control systems. Labora-
tory work is used to reinforce the lecture material but,
unfortunately, here again we must "steal" lab time from
the lecture schedule. We have not found a truly suitable
text for this course, but have based much of the material
on the books by Gelb [5] and Meditch [6]. Current litera-
ture is also an essential source for this course.

To complete the process control option, the student
must take one more course related to process control We
have been very liberal in this matter. For many students,
it is advantageous to take a processing course, in polymers
or pulp and paper for example. This gives better insight
into the problems which a control engineer might be ex-
pected to solve within a processing industry. We do not
encourage co-op students to make this choice, as these
students already have two semesters of professional work
experience behind them. Many students will elect a course
in mathematics, computer science, or electrical engineering.
The remaining students will usually choose to do a thesis
within the honors program (directed to a control problem),
or an individual senior project having something to do
with control. The actual choice which any particular
student makes depends upon his interests.


We feel that there is a need for suitable text-
books for this program. We look forward to the
publication of a chemical engineering text suit-
able for the first course, ChE 152, which covers
the introductory material from the "modern con-
trol theory" point of view. For ChE 154 and 156


The program is ... unique and valuable. There is a high demand for our graduates, and they
seem to fill an obvious industrial need which is not being met by graduate schools. The sequence blends with
the other options within the department so that all students can find something of interest to them.

the text by Kuo is quite suitable. It is only a
minor handicap that the text is not oriented to-
wards chemical engineering or the process indus-
tries. The most serious text problem is the lack
of a suitable text for the final course, ChE 158.
This lack can only be partially offset by lecture
notes and the use of the periodical literature.
Within the department, we have excellent com-
puter facilities, and we are developing laboratory
experiments suitable for the program. We have
interested and capable students, and a good mesh
of faculty interests and expertise. On the other
hand, teaching loads in the department are very
high, making it difficult to teach, do research,
and to find the time to continue to develop and
improve the sequence of courses in the program.
It has required an almost fanatical belief in the
importance of what we are doing to push the
courses to their current state of development. Ac-
cording to Waller, this would probably be true
anywhere in the country, as control theory has not
been much appreciated [2].
About half of our students are in the co-op pro-
gram. It is impossible for these students to take
the control sequence in four consecutive semesters.
Because we cannot offer extra sections of any of
the courses, we have had to design the sequence
so that students could take either ChE 154 or ChE
156 as the first course beyond the introductory
course. This is the reason for the artificial division
of the material in Kuo [4]. Though this is less than
ideal, it seems to work. It should be mentioned that
the work experience of the coop students, about
half of the class, tends to raise the level of effort
and understanding of the entire class.
The constraints of university, college, and de-
partmental requirements so far have made it im-
possible to develop and include a formal laboratory
course in the program. Therefore, we must sacri-
fice lecture time in order to provide essential
laboratory process experience. Laboratory is very
time consuming for students and faculty. Because
of this pressure, we feel that we must provide
more setup and laboratory instruction than we
would like. This problem will be alleviated some-
what by development of more experiments which
can be run simultaneously by different groups.

The problem here is instrumentation cost. Avail-
able computing power is more than sufficient.

The program is, we believe, unique and valu-
able. There is a high demand for our graduates,
and they seem to fill an obvious industrial need
which is not being met by graduate schools. The
sequence blends with the other options within the
department so that all students can find something
of interest to them. It is important to recognize
that the option programs work within what most
curricula consider to be technical electives. There-
fore, every student completes the full ABET-
AIChE accredited program in Chemical Engi-
The "critical mass effect" has been very im-
portant. It is not only necessary that more than
one faculty member be closely involved in the
program, but it is equally important that the fac-
ulty have interest and expertise in certain speci-
fic areas. We found it necessary that the following
areas be fully covered: mathematics, instrumen-
tation hardware, computer hardware and software
(on-line, real time system), interfacing of com-
puter and process, computer operating system
maintenance, industrial experience, and the areas
of process control covered by the courses. With-
out skill in the above "auxilliary" areas, our pro-
gram simply would not exist.
Although we feel we have developed an excel-
lent program, we have certain constraints which
we have been unable to overcome. Our students
generally lack background in statistics, proba-
bility, and linear algebra. Although we do con-
sider these topics within the sequence, we cannot
spend enough time on them. We do not have suffi-
cient time to spend on hardware, and we have no
time to spend on microprocessors or assembly
language. We agree with our industrial advisors
that these topics are important, but we feel that
the material we do cover is more important, and
that any time we can "steal" must be used for
laboratory work. Those students keenly interested
in any of the topics mentioned above can learn
more about them through courses given by the
mathematics and electrical engineering depart-


ments. These courses can be taken as the fourth
elective or as overload courses.
Industrial response to our process control pro-
gram has been positive and encouraging. Indus-
trial representatives have helped us to choose the
material to be presented and have helped with
financial support for our computer facility. Posi-
tive feedback from instrument and control system
vendors and from manufacturing companies, es-
pecially pulp and paper companies, has consisted
not only of verbal approval but also the ultimate
test: They need, want, and hire the graduates of
the program. O

1. Seborg, D. E., "A Survey of Process Control Educa-
tion in the United States and Canada," Chem. Eng.
Education 14, No. 1, 42 (1980).
2. Waller, K. V., "Impressions of Chemical Process Con-
trol Education and Research in the USA," ibid. 15,
No. 1, 30 (1981).
3. Luyben, W. L., "Process Modeling, Simulation, and
Control for Chemical Engineers," McGraw-Hill
4. Kuo, B.C., "Digital Control Systems," Holt, Rinehart,
and Winston (1980).
5. Gelb, A. (ed.), "Applied Optimal Estimation," MIT
Press (1974).
6. Meditch, J. S., "Stochastic Optimal Linear Estima-
tion and Control," McGraw-Hill (1969).

W book reviews

By W. H. Ray
McGraw Hill, NY, 1981
Reviewed by John C. Friedly
University of Rochester

There are few available books suitable for a
graduate-level course in chemical process control,
although there are a number of more general as
well as a number of more specialized advanced
control texts. That fact alone would make this
text a useful addition to the literature. However,
this book has more to appeal to students and
practitioners alike.
The book covers nearly all of the most used
ideas of modern process control, from multivari-
able control, through optimal control, to state
estimation. Ray tries to present a balanced cover-
age of control of both lumped and distributed
parameter systems, linear and nonlinear systems,
theory and practice. From the large literature

available, choice of the basic principles has been
skillfully made. As in most of Ray's work the
concepts are illustrated with a wealth of rather
detailed examples, all pertinent to chemical engi-
neering. In addition, there is a summary of several
more elaborate case studies taken from the litera-
ture. The only noticeable omissions are parameter
estimation and adaptive control.
The style is mainly expository with approaches
to solving the problems stressed rather than an
axiomatically rigorous mathematical treatment.
As should be expected, vectors and matrices are
used throughout and the manipulative skills of
linear algebra are essential. For the most part,
mathematical methods are introduced where
needed as unobtrusive digressions in the text. The
prose is eminently readable and the typescript is
free of obvious errors.
The bulk of the material is jammed like meat
on a delicatessen sandwich into just three rather
long chapters. The advanced control concepts are
first introduced in Chapter 3 on control of lumped
parameter systems. Multivariable control, non-
interactive control, modal control and optimal
control are all covered in some detail. Both the
theoretical basis and the more practical imple-
mentation of the control is discussed, including
computation and approximation problems. The
following chapter attempts to give the same
type of coverage for control of distributed pa-
rameter systems. This is done to the extent
possible and the distinctive problems associated
with the more complex distributed systems are
covered well. The coverage of optimal filters and
observers in Chapter 5 is complete, but the con-
cepts of stochastic control are only paid lipservice.
In the same chapter both lumped and distributed,
linear and nonlinear systems are included. These
meatier chapters are contained between two
which, while appropriate, suggest that the sand-
wich could well have been served open-faced.
Chapter 2 gives an overview of the hardware
and practical details of the use of minicomputers
for process control and is virtually not referred
to again in the text. The final chapter reviews
several case studies from the literature that use
many of the concepts covered earlier, but is also
somewhat disconnected from the rest of the ma-
terial in the text.
The text is not without its faults however. In
using many parts of the text for a course in pro-
cess control, students found that it contained a
Continued on page 46.


F H laboratory


New Jersey Institute of Technology
Newark, NJ 07102

DIRECT DIGITAL CONTROL OF liquid level was de-
veloped at the New Jersey Institute of Tech-
nology as an experiment in the undergraduate
process control laboratory. The objective of the
experiment is to give the students exposure to the
area of digital control which is widely used by
industry, although not much attention is given to
it in most undergraduate process control curric-
ula. Students can be exposed to writing their con-
trol program, and can experimentally determine
the transfer functions for all system components.
They can compare the actual response of the con-
trolled system with the theoretically predicted
response following a step disturbance in the feed
rate and also study the effect of sampling time
on system response.

The experimental set-up consists of a plexi-
glass tank 0.14 meter in diameter, fed with water
from the bottom through a rotameter with a max-
imum capacity of 0.0015 m3/s or 24.5 gpm. A
Fisher Governor control valve type 657A, 3-15 psi
is located on the tank outlet pipe. The original
set-up, designed for pneumatic analog control in-
cludes a Foxboro Model 58P5 controller, and a
3-15 psi type 13A-1 Foxboro transducer. To pro-
vide direct digital control, a Devar Inc. air to
electric converter (A/E) type 18-118L, 0-5 volts
was connected to the transducer, and a Devar Inc.
electric to air converter (E/A) type 18-150, 0-5
volts was connected to the control valve. Plumbing
was done in such a way that the system can either
be operated by the controller or by the digital
computer. The digital computer installed was a

Students can be exposed
to writing their control program and
can experimentally determine the transfer functions
for all system components.

C Copyright ChE Division, ASEE, 1983

*oome te

FIGURE 1. Schematic drawing of System.

Digital MINC-11 minicomputer equipped with
four laboratory modules: a preamplifier, an analog
to digital (A/D) converter, a clock, and a digital
to analog converter (D/A). A schematic drawing
of the system is shown in Fig. 1.

A material balance in terms of deviation values
can be written as

A-= X- (1)

where A = tank cross sectional area; X = feed
flow rate; U = outlet flow rate.
The outlet flow rate "U" is a function of the pres-
sure to the valve "P" and the liquid level "Y".
This can be written as
U = (au/ay)ssY + (au/ap),sP (2)
where the subscript "ss" represents steady state
value. Combining the above equations, transform-
ing to the Laplace domain and rearranging gives

Y(s) = [X (s) kP (s)] (3)
(7s + 1)
where 7 = A/(au/ay) ss; kp = 1/(au/ay),,; kv
= (?u/ap),s.

1. T, k, can be determined by making a step
change of magnitude "a" in the feed rate while
the analog controller is placed on manual status.


eter scale and utilizing the following experimen-
tally determined parameters.
kp = 17.85 min/ft2, (1.153 s/m2)
kv = -0.151 ft3/min.psi, (-1.034
x 10-3m3/s.Pa)
ke = 3.53 psi/psi, (3.53 Pa/Pa)
T = 2.98 psi/ft, (6.74 x 104 Pa/m)
T = 176.40 s
The solution of the above equation for liquid level
deviation in the units of meters can be written as
Y = 0.0729 [1 exp (-0.1664t)] (7)
where the time units are seconds.

M. F. Abd-El-Bary is an Assistant Professor at the New Jersey In-
stitute of Technology. He holds degrees from Alexandria University,
MIT, and Lehigh University. He has a strong interest in the area of
digital control. His research work is primarily in the field of water
and air pollution.

The two parameters can be calculated from the
Y = ak,[1 exp(-t/r)] (4)
2. With liquid level being kept constant, and the
controller on manual status, Eq. (3) can be
written as
P (S)
P () k (5)
The valve transfer function k, can be obtained
by changing the flow rate to the tank and vary-
ing the pressure to the valve to keep the liquid
level at a specified value.
3. In a closed loop experiment with proportional
control only, the proportional band was set at
25, and the set point on the recorder controller
was set at 9 psi. From the chart recorder,
values of the error and pressure to the valve
were obtained and the proportional control
gain ke can be calculated.
4. The transducer transfer function was obtained
by changing the water level in the tank and re-
cording the resulting transducer output (in psi
as obtained from the chart recorder).

The closed loop transfer function for a pro-
portionally controlled system can be written as
Y(s) = k (6)
X (s) (TS + 1) k3kpkT
For a step change from 18 % to 30 % on the rotom-


Liquid level was converted to an air pressure
signal through the transducer, then to an electric
voltage signal through the A/E converter, and
finally to a digital signal through the A/D con-
verter. The developed computer program com-
pares the measured liquid level to the set point,
thereby generating the error. Control actions pro-
vided in the software were; proportional (P),
proportional-integral (P-I), proportional-derivi-
tive (P-D), and proportional-integral-derivative
For continuous sampling an ideal P-I-D con-
troller is described by the following equation.

r t
V o k eedt+ de
V = Vo + k e + edt + t (8)

V = controller output signal at time t
Vo = controller output signal at t = 0
e = error signal
,r = integral time
7D = derivative time
The discrete equivalent of the above equation can
be obtained by replacing the derivative with finite
difference, and using rectangular integration for
the integral. The computer output at the nth
sampling instant can, therefore, be written as

V, = Vo + ke eo + -T2 en + (e. en)
71 0 T
T = sampling periods, seconds
V, = computer output at the nth sampling


en = error at the nth sampling instant
en-i = error at the (n-1) sampling instant
Similarly at the (n-1)th sampling instant, the
equation for the output is

V-= = Vo + ke


T n-i
+ -- e,
TI o

+ -q (e10 n-- e-_,)

Substracting Eq. (10) from Eq. (9) gives

Vn = V,, + k,

(e -en-) + Te,
-- en

+ (e, 2en-1 + en-) (11)

In the sampling method adapted in this experi-
ment one point (y2) was collected at the start of
each sampling loop and the error en was calcu-
lated. This error was used along with the errors
en-, and en-2 according to Eq. (11) to formulate
the computer output signal. The next loop began
with the collection of a new data point (y,) and
calculating a new error en the old value of en was
transferred to e,_-, the old value of e.n- was trans-
ferred to en-2, and the old value of V. was trans-
ferred to Vn-. The fastest sampling rate was
approximately one sample every 0.12 second for
proportional control. A pause statement was also
incorporated in the program to introduce a time
delay, thus varying the sampling rate. The soft-
ware also contained a program for calibrating
the "transducer + A/E" converter, and a pro-
gram for calibrating the "E/A converter +
valve." Fig. 2 represents the block diagram for the
digital control loop. The output of the A/E con-
verter which is a continuous electric signal of the
measured variable is fed to the A/D converter.
The A/D converter changes the continuous signal
to a discrete form (series of impulses of varying

S t Pon, Computer


FIGURE 2. Block diagram representation of the digital
control loop.

strength). The computer compares the discrete
forms of set point and measured value to produce
the error, and acts on the error by the appropriate
control algorithm. The output of the computer
generated every "T" seconds is sent to the D/A
converter where a continuous signal is produced.
In effect, the D/A converter clamps on the signal
until the next one comes along; i.e. the output
voltage of the D/A converter remains at a con-
stant value over the sampling period. The D/A
converter, therefore, acts as a holding device. For
a zero-order holding device, the transfer function

1 exp(-sT)
The analysis of a discrete system is conveniently
done in terms of z-transforms, defined as

F(z) = Z[f(t)] = f(nT) z-"


The use of the z-transform in discrete systems
is analogous to the use of Laplace transform for
continuous sampling, and the Laplace transforms
are usually tabulated alongside the time function
and the corresponding z-transforms [1]. This al-
lows direct conversion from the Laplace transform
of a continuous function to the z-transform of that
same function.
The closed loop transfer function for the

Y(z) _G,(z)
X(z) 1 (TkAkckEkVHG,) (z)


Since T, kA, ke, kE, kv are constants, they can be
combined in one term as K, and the output varia-
ble can be written as

(XG,) ,
Y (z) (XGp)


What is needed are the values of (XGp), and
(HGp), to plug in the above equation

Z(XGp) = Z [a k 1 z [ak, ak
Ls + s s + 1/TJ

= ak, z z
St-1 z exp(-t/T)J

Let e(-t/7) = b. Then

(X ) ak (1- b) z
Z(z 1)(z b)





I _--o-- Dig0 lal Control Exprimn Data
0,04 --- Equotion (20)
u- / - Analog Control Experimental Dato
0.02 _-- Equation (7)


0 10 20 30 40 50 60 70 80 90
TIME (Seconds)
FIGURE 3. Comparison between experimental and
theoretical responses.

Z(HGp) = Z -s 1 e]

Z kp[l exp (-Ts) ]
S(Ts + 1) s

=k,(l z-1) Z s (s+ 1)

(z b)
Substituting and rearranging gives


Y(z) = akp (1-b)z / z2 + z[Kkp(b-1)

(b + 1)] + [b K k(b- 1)]
with values of


kA = 0.358 volt/psi, (5.1924 v 10-5 volt/Pa)
kE = 2.649 psi/volt, (1.8264 x 104 Pa/volt)
K = -1.51 ft3/min.ft, (-2.3381 x 10-3 m3/s.m)

To examine the effect of sampling time on the
stability of the system one has to substitute dif-
ferent values of sampling time in the above equa-
tion and find the values of denominator roots. If
the roots are located within the unit circle of the
z-plane, the system is stable [2]. The following
table shows the values of the roots at different
sampling time.

Sampling Time
0.12 sec.
1.0 sec.
5.0 sec.
10.0 sec.
12.0 sec.
13.0 sec.
13.1 sec.
14.0 sec.

+1.000, +0.981
+1.000, +0.842
+1.000, +0.220
+1.000, -0.563
+1.000, -0.831
+1.000, -1.128

Notice that second root decreases in value as the
sampling time increases and the system becomes
theoretically unstable when the sampling time is
above 13.1 seconds. It can therefore be stated
that increasing the sampling time would desta-
bilize the system.
For a step change from 18% to 30% on the
feed rotameter scale, and for a sampling time of
0.12 seconds, Eq. (7) reduces to

S 0.001452 z
2 1.98103 z + 0.98103
z2 1.98103 z + 0.98103

= 0.0759 z .
=z -1 z 0.981
The pole 0.981 can be expressed as

0.981 = exp(-aT) = exp (-0.0192)

Y = 0.0759 z 1 z exp (-0.0192)

Inverting to the time domain gives

Y = 0.0759 [1 exp (-0.0192n)]




0 10 20 30 40 50 60 70 90 so
TIME (Seconds)
FIGURE 4. System response for different control modes
under digital control.


Eq. (7) and Eq. (20) are plotted in Fig. 3
along with experimental data taken with analog
proportional control, and digital proportional con-
trol at 0.12 seconds sampling interval. Agreement
between the equations and the actual response was
quite good.
Fig. 4 shows the response of the system at the
fastest sampling time to a step change in the feed
rate using proportional control (ke = 3.53 volt/
volt), proportional-integral control (ke = 3.53
volt/volt, Ti = 20 seconds), proportional-deriva-
Continued on page 47.



University of Houston
Houston, TX 77004

SE REPORT HERE THE results of a survey of
chemical engineering departments regarding
computer usage in senior design courses. All
chemical engineering departments in the United
States and Canada were polled, and two-thirds
responded. That the sample was representative
was verified by checking the number of schools
reportedly using FLOWTRAN (12) against the
actual number (18) : (Personal Communication,
Prof. J. D. Seader, University of Utah, CACHE

1. Use of Process Simulators
Forty seven of the ninety eight
schools responding used process simula-
tors. The most popular were: CHESS
(5), GEMCS (4), ASPEN (2).
2. Student-Written Programs
Fourteen schools responded that the
students write programs in the design
3. Faculty-Written or "Canned Programs"
Four of the five schools using canned

Ernest J. Henley has been a professor of chemical engineering at
the University of Houston since 1964. He received his Ph.D. from
Columbia University in 1953 and has been on the faculty of Columbia
University and Stevens Institute of Technology.

programs included CSMP on their list.
Many schools use mixtures of the
above, as follows
1. + 2. Number of Schools = 7
1. + 3. Number of Schools = 4
2. + 3. Number of Schools = 25
1. + 2. + 3. Number of Schools = 30
Only seven schools do not use the
computer in the senior design courses.
B. Cost of Computer Usage
The schools were asked how much they
spend on computing per student, and whether
the funds came from "soft" or "hard"
sources. The replies were
Soft Money
$ 10- 30 : 13 schools
31- 70 : 12 schools
71- 150 : 18 schools
151- 300 : 9 schools
301- 500 : 6 schools
501- 1500 : 3 schools
Average: $178/school
Average for Schools Using Simulators:
Hard Money
$10- 30 : 4 schools
31 100 : 7 schools (4, partial soft)
331 : 1 school (partial soft)
500 : 1 school
C. Available Software
One of the questions was whether the
schools had software they would share with
other schools. This produced the list shown
in Table I. Entries 14-17 reflect the recent
decision of the four major purveyors of
industrial-level simulators to make their prod-
ucts available to Universities. Except for the
case of ASPEN, only load models are being
provided. These, of course, are computer de-
The use of large simulators in senior design
courses is clearly established. Over sixty five per-
cent of the schools will be using them in 1983.
Computing costs are high. Unmonitored student
use can easily result in expenditures of over
$1,000/student. E


Copyright ChE Division, ASEE, 1983


1. Dr. R. D. Weir

.2. John A. Meyers

Jim Douglas

John H. Erbar
Jan Wagner

Paul Babcock

R. L. Motard

M. V. Svrcek

W. D. Seider

B. A. Finlayson

E. J. Henley

Alberto I. LaCava

John L. Potter

R. F. Benenati

Vickie Jones

Larry J. Lesser

Jim Byrne

17. Margaret Butler

TABLE 1. Programs Offered


Royal Military College of Canada

Villanova University







University of Massachusetts

Oklahoma State University

University of Connecticut

Washington University, St. Louis

University of Calgary

University of Pennsylvania

University of Washington

University of Houston

City College of City University, NYC

New Mexico State University

Polytechnic Institute of New York

University of Utah

P. O. Box 1885, Houston, TX 77001

Simulation Sciences, Inc.
1400 North Harror Blvd.
Suite 250, Fullerton, CA 92635

National Energy Software Center
9700 South Cass Ave., Argonne, IL



CALYPSO-Neutron Diffusion Code

Activity coeff. from data. Tray efficiency
Antoine constants from data HTU in packed
column. Multipass ht. exch.
Non-ideal binary x-y.
Linear and non-linear regression.
Tray by tray by relaxation.
Multicomponent distillation.

Process synthesis programs (in preparation)

PAS, MINISIM (General purpose simulators
similar to CHESS, FLOWTRAN, etc.)

GEMS, FLOWSIM under restrictive agree-

CHESS ($1000/school)

HYSIM, HYDIS, (property of Hydrotech
Ltd. of Calgery)

GIBBS, Chemical and phase equilibrium.
HETDIS, three-phase distillation

DISTIL-Shortcut distillation

BCOST-Equipment costing and economic

Parametric Estimation. Dynamic Simulator.
Nonlinear Equation Solver.

Economic Evaluation. PCS (written by J.

Comprehensive collection covering unit ops
suitable for microprocessors (Fall, 1982)

FLOWTRAN (Load Module Only)

DESIGN/2000 (Load Module only)

SIMSCI (Load Module only)


1 international



Manchester M60 1QD, England

A LL ENGINEERS, AND particularly chemical engi-
neers, have needed to bring to their profes-
sional activities a blend of inspiration, information
and perspiration. A common complaint about the
academic training of engineers is cogently ex-
pressed in the Finneston Report where it is con-
cluded that not enough attention has been paid,
in the recent past, to developing the inspirational
aspects of design. This report recognizes the need
for more time and effort to be devoted to training
in design methodologies but, apart from one brief
reference and recommendation, it fails to antici-
pate the impact that developments in computing
techniques would have in the areas of training and
design methodology.
As computer technology has evolved in the
directions of large efficient time-shared machines
on the one hand and more powerful desk-top mini-
computers or micro-processors on the other we
are witnessing significant changes in the ways in
which we tackle the design both of individual
units and of integrated systems. It is particularly
in the opportunities that we now have for exam-
ining alternatives in large scale "integrated" de-
signs that we can see the greatest impact.
All the major chemical manufacturers, oil re-
finers, plant contractors and design offices have
built up their computer-aided engineering activi-
ties over the past fifteen years so that they now
possess significant libraries of well proven pro-
grams for use by their staff. The smaller firms
may not have been able to spare the effort to de-
velop the soft-ware themselves but in many in-
stances make use of bureaus which have estab-
lished themselves as reliable sources of design
capacity on large, shared, computers.

*A revised version of a paper originally presented to
the 1979 Montreux Conference-"Computer Applications
in Chemical Engineering."

The smaller firms may not have been able
to spare the effort to develop the soft-ware
themselves but in many instances make use of bureaus
which have established themselves as reliable
sources of design capacity on
large, shared, computers.

Clearly, the universities have a responsibility
to ensure that when their graduates emerge they
are prepared to work in this new manner and
many academic institutes have not been slow to
recognize the potential for using computers to
improve their teaching and training activities.
The original CACHE committee which published
a well-known set of some 112 programs in 1972
is now a Corporation which is fostering these
activities in the U.S.A. Of these the most ambi-
tious is the CHEMI project which aims to produce
some 500 tried and tested instructional modules.
CACHE also supports the use of FLOWTRAN as
a networked flow-sheeting package and has now
initiated a project to develop and exchange Micro-
computer Aids for Chemical Engineering.
CHICHEE and EURECHA have encouraged
parallel but smaller scale activities in the United
Kingdom and Europe.
Most of the activities referred to above were
directed to exploiting batch computing facilities
but it became apparent in the late '70's that devel-
opments in low-cost interactive computing were
going to have the most significant impact on
Computer-Aided Design (in all its shades of mean-
ing) and that, potentially, this mode of computa-
tion could be of equal importance for the develop-
ment of Computer-Aided learning. It was these
considerations which led the authors to examine
the problems of constructing a package which
would be designed to take advantage of the vibrant
characteristics of this mode of working.
After some initial trials using an existing
Cyber 72, the hard-ware described in the next
section was acquired to evaluate the package under
conditions which would provide some 250 chemical
Copyright ChE Division. ASEE, 1983


engineering students (3 years of B.Sc. and 1 of
M.Sc.) with a novel experience and also permit
colleagues in other departments to take advantage
of the facility in order to develop teaching pack-
ages appropriate to their own disciplines.


The hard-ware to be described is as it will be
at the completion of the first phase of the project
and is the best that we could afford and most
appropriate for the computers available to us. It
should be stressed that the TUTCHE package
does not depend on this hardware and could be
run on any interactive system ranging from a
single-user mini with a 10 char/sec teletype to a
multi-user main frame provided with a large
number of (colour) graphics terminals. We had
outgrown the former but could not afford the
The compromise was to set up three clusters of
five terminals arranged as shown in Fig. 1. In
any one cluster all five terminals have an upper
and lower case alphanumeric capability, three
have a graphics capability which is required to
give a full Tektronix 4010 emulation and one of
these has a micro-processor capacity (twin 51/2"
floppies) so that it can work in "stand alone"
mode or in transparent mode when it behaves as a
a/graphics terminal. All five may be connected
by a simple multi-pole switch to an a/graphics
printer with extended buffer storage that enables
any one user to dump a "page" as displayed at
the terminal onto the printer. (We should remark,

F. P. Stainthorp graduated in
chemical engineering from Man-
chester in 1941. After eight years in
industry he was transferred to Uni-
versity College, London to work for
his doctorate and then returned to
chemical industry for a further six
years. He was recruited to Man-
chester, UMIST, at a time of ex-
pansion with a special remit to
cultivate the growing fields of proc-
ess control and digital computa-
tion. He is chairman of CICHEE and
directs the chemical engineering
computing service at UMIST. (L)
David Lomas is a member of the British Computer Society and has
spent most of his working life in University engineering departments
first as a programmer but lately as the supervisor of the chemical
engineering computing service at UMIST. He has made a special
study of the problems of transferring nominally 'standard' programs
between different computing systems. (C)

\ _1



here, that we have had no problems with a-nu-
meric displays and printing, but we are still not
satisfied with the copying of graphics display and
are currently examining alternatives.) The am-
bition is to progressively upgrade the alpha-
numeric terminals as funds and graphics require-
ments increase.
Any of the fifteen terminals in the suite will
be connected via a "Gandalf Switch" to various
machines and currently we may access either a
Prime 750 or a Cyber 172 both of which provide
time-shared interactive operation at linespeeds
of 2.4kB which is just about the lowest acceptable
rate for graphics work and is a very comfortable
rate for conventional dialogue.
The Prime has proved to be a very effective
machine since, up to the limits of the locally set
connection limit, it gives a very fast response to
"trivial" dialogue-i.e. during data input sessions.
An important feature that we have been able to
exploit is a "shared code" facility which means
that when all fifteen terminals are used by stu-
dents on the same tutorial exercise we can arrange

J. Alonso received his Chemical Engineering and MSc. degrees from
the Universidad Central Venezuela in 1972 and IVIC 1976, and his
Ph.D. degree from the University of Manchester England in 1979. He
currently is lecturer of process simulation at the Universidad Central
Venezuela and his interests have generally been in software de-
velopments. (R)


for only one copy of the program to be resident
in the machine. This has the big advantage of en-
abling the normal load of other users to be satis-
fied without them seeing a significant impact from
the student users.
The Cyber 170, having nearly 200 terminals
connected across two campuses, tends to get over-
loaded at peak times, but a scheduling facility in
the switch should ease this problem in the future.
The big advantage of this machine is that it is
linked to a CDC 7600 which is used to process
the largest design packages. In this mode of work-
ing the intention is to use the interactive machine
for validation and preprocessing activities that
generate a suitable data file for subsequent batch
processing in the larger machines. The interactive
system is also used for post-processing the output
files resulting from the batch machine so that
only selected material required for permanent rec-
ord need be printed.

Fig. 2 is a diagrammatic representation of the
overall structure of the package which is shown
in this manner for ease of discussion of the vari-
ous components. There is some correspondence
with the actual system since many of the programs
need to use common data files and in general only
one program will be in use by a student at any

COSTS DYNPACK co1mp etc. pipeline
Haz. CONSYST Hexer



one time. We think of four banks which contain,
respectively, common data, overall systems related
programs, unit design specification programs, and
detail design programs. Another bank contains a
library of service programs or sub-routines which
can be used to deal with input/output problems,
graphical interactions, function fitting etc. Each
component of the total package communicates
with the user via a dialogue that is appropriate
to the application. On the first occasion that a

On subsequent occasions
abbreviated prompts are used but the
student may always ask for "Help," whereupon the
abbreviations are explained and amplified if
further requests for help are made.

program is used the requests are full but not
verbose. On subsequent occasions abbreviated
prompts are used but the student may always ask
for "Help" whereupon the abbreviations are ex-
plained and amplified if further requests for help
are made. We also find it convenient to include
a switch that enables the frequent user of a par-
ticular program to suppress the dialogue.

Students gain access to the suite via a program
named TUTCHE. This program is a purpose writ-
ten data base management system which serves
to control and record the use of the various pro-
grams by the students. It also tidies up the stu-
dent's own data and results in files on exit so
that they may be retrieved reliably on re-entry.
TUTOR is a separate interactive program
which enables a tutor to examine and modify the
record held in TUTCHE. The use made by all
students of any one program may be displayed
or the record of programs used by any one stu-
dent may be examined. In some cases a permit
has to be set and continued use of a program may
be contingent upon monitored successes at selected
points. The tutor may insert, examine and modify
these controls through TUTOR.
This control is not a case of "Big Brother
watching" since, in the main, the programs are
directed to use by responsible and mature students
who each work at their own pace but within
constraints imposed by the capacity of the re-
sources. Examples of the regulation of use will
be given later but in general they either ensure
that the student has understood the fundamentals
of the tutorial exercise and is not using the pro-
gram as a "black-box" or they serve to curb ex-
travagance on open-ended exercises. This disci-
pline is intended to correspond to that which they
will find necessary in industrial environments
when the use and time has to be paid for.
The titles shown in Fig. 2 are those that form
the current basis for the package. Those marked
with an asterisk have reached the stage of regular
student use, those in capitals but not starred are
undergoing trials and those in lower case are still


in preparation. More details of some of the pro-
grams and how they are used follow.

FLOWPACK and PROCESS are two major
flow-sheeting packages that have been made avail-
able to U.K. students by the generosity of ICI
Ltd and Simulation Sciences Inc. respectively.
Both packages are maintained on the CDC 7600's
at the University of Manchester Regional Com-
puter Centre by UMIST department of Chemical
Engineering for use by some fifteen departments
at other, remote, universities. A growing number
of students make extensive use of the packages
for their design exercises and we find the nomi-
nal royalty figure calculated and displayed for
each run of the PROCESS program a very con-
venient means of control. Each student may be
given an allocation of "cost- units" within which
the exercise should be completed and this gives
a strong incentive to rational use of the package.
At the moment the flowsheeting packages are
run as batch jobs from card input but we are de-
veloping an interactive preprocessor that will
validate the data as they are introduced. Ulti-
mately, users will start from a flow-diagram of
the type shown in Fig. 3 which has been produced
by using "FLOSSY" on a graphics terminal .

This is a three part package which has pro-
vision to generate costing information in any cur-
rancy using a reference year and a selected cost
index. After the initial dialogue the student may
enter any of three sections:-

"COSTIT" for costing items (vessels, heat-exchangers,
etc) using stored correlations relating F.O.B. costs to
a capacity factor with adjustment for selected mate-
rials of construction. In a few cases the cost of the
item can be estimated for detailed fabrication factors.
"COSTFCI" for estimating Fixed Capital Investment.
Two methods are provided: (a)Zevnick-Buchanon giv-
ing a very approximate figure based on the principle
units, and (b) a factorial method (Millar) applying
factors for installation, civil, structural work, etc to
all the items.
"COSTASS" for estimating product costs from FCI,
depreciation policy, taxes, materials, utilities, labour
rates, etc, etc.
Profitability analysis and comparisons based on eight
different criteria can be made and this may be followed
by a sensitivity analysis in which the effects of vary-
ing any of the principle costing components may be

All of the above are closely tied to tutorial ex-
ercises aimed at reinforcing an appreciation of
the economic factors that determine good design.

The determination of economic designs is in-
tended to be brought out in all, the programs
stored in the Specifications Bank and the approach
adopted is illustrated by these two programs
which deal, respectively, with the fundamentals


of multiple-effect evaporator design and the ex-
tension to more elaborate evaporator systems.
When using MEEVAP the first exercise is one
in which the' student initially determines by a
hand method the area required for a three effect
evaporator to carry out a particular duty when
using steam at a specified pressure. (Each student
has a different set of data). The student intro-
duces the same data to the program using a dia-
logue which validates every item and at the end
of the data input session is requested to provide
the calculated answer. If this is within 5% of
the (unrevealed) computed solution then the next
phase of the exercise can be commenced. If the
student's solution is outside the given tolerance
the data is filed and he or she is invited to check
their working and, if necessary, see a demon-
strator for further advice. After three failures
a lock is set to prevent access to the program until
the tutor has discussed the problem again and
released the lock.
The continuing exercise may take any direction
desired by the tutor since the student can start
from any one run and alter the evaporator ar-
rangement, number of effects, callandria area or
any other design or operating variables. From the
results obtained the student examines the econom-


ics of various alternatives and ends up with a
design which satisfies some agreed criterion.
With MEESYS, access to which may be
contingent upon demonstrated successes with
MEEVAP, the student can explore the economic
benefits of introducing heat-exchangers to preheat
the feed and/or the addition of condensate flash
Automatic optimisation is deliberately ex-
cluded from these packages but the student needs
to be able to compare several designs so an essen-
tial component of the package is a facility to gen-
erate tables which collate nominated variables
for a sequence of runs. Fig. 4 is an example of

6 BAC 0.581 500 65.0 257, 2.50
5 BAC 0.660 5.00 65. 1 273. 7.50
7 BAC 0.665 4.00 65.0 254. 2,50
4 BAC 0.702 5.00 72.9 240. 7.50
3 BAC 0.772 4.00 71.2 240. 7.50
2 BAC 0.910 3 00 69 6 240. 7.50
I FOR 0.984 3.00 69.7 240, 7.50

such a table in which the runs have been ranked
in order of increasing flow of steam to the first

After two years experience at using the pack-
age at UMIST and some attempts to transport
parts of it to other departments with different
systems we can comment on some aspects which
seem to us important.
There is no doubt that the impact made by a
well-prepared interactive exercise can be sub-
stantially greater than a conventional tutorial. The
opportunity for the student to examine aspects
of design which calculation time would hitherto
have prevented is particularly welcome. We would
certainly endorse the experiences of those such
as our colleagues at Illinois, who find the PLATO
system so effective. However, we also recognize
that a considerable effort is required to generate
the program, exercises and support documen-
tation needed for each module in the packet. We
believe that the only solution to this problem is
to ensure that the packages receive wide distri-
bution and use by generating modules to agreed
standards and by enlisting the co-operation of
several authors in many departments so that
modules may be exchanged.
The original version of the package was writ-
ten, as far as possible, in strict FORTRAN4

(ANSI 66) but we have now rewritten it to con-
form to strict FORTRAN77 (having looked at
PASCAL and rejected it). There are still several
machine dependant features, however, connected
with file creation, opening and closing, I/O pro-
cedures, graphics or quasi-graphics. Our solution
to the problems in these areas is to confine them
to a few sub-routines which have been coded so
that the local peculiarities may be incorporated
during the initial setting up. Particular attention
has been paid in writing the I/O procedures to
making the dialogue "user friendly' and to trap-
ping spurious or invalid responses.
The students need to be prepared for this type
of tutorial work by providing them with concise
documentation which sets out clearly the initial
objective and encourages them to extend the exer-
cises in an imaginative way. We arrange to sched-
ule the terminals so that their use coincides with
other practice and laboratory classes and find ses-
sions of about 45 minutes to be about optimal. E

Continued from page 15.
gave in Montreal in 1961 so much sums up his
work that it is worth examining in a little more
detail. The flavor of the argument and its presen-
tation will have been, gained from the lengthy
extraction which was quoted above; here I would
like to lay out its main lines. After an historical
introduction Wilhelm divided the field of reactor
studies into three parts. These were classified
according to the difficulty of applying design ob-
jectives and the first was that part which seemed
beyond immediate a priori design objectives be-
cause of incomplete theory. The second was the
domain of moderately well-understood processes
and the third was that part which was already
developing rapidly. In 1962 he saw flame theory
with its rapidity of reaction and complicated ther-
modynamics and other systems in which turbulent
fluid dynamics plays a part as in the first category.
Fortunately, the second category was larger and
comprised most of the conventional reactors for
which sensible models had been constructed. The
incorporation of some of these reactor types into
the third area of well-understood systems was
the thrust of his own concern and, as he said, "I
have chosen the packed bed reactor, a device
which clearly falls in the third category, as the
specific subject of my paper."
He commented that historically the packed


bed reactor is one of the oldest arrangements
for conducting gas-solid or liquid-solid chemical
reactions on an industrial scale, but that the
component parts which he proceeded to analyze
had only more recently come under investigation.
He divided this into the external field model, the
interparticle field, the interphase field and the
intraparticle field. In the first case he summarized,
with masterly touch, the state of knowledge on
axial and radial dispersion in the packed bed, an
area to which he had himself contributed so much.
He then turned aside to comment on the Lapidus-
Dean cell model, recognizing its potentials and
limitations. He also mentioned in passing the
currently exciting work of correlating experimen-
tal data on kinetics with scientific, rather than
empirical, expressions for the reaction rate. Non-
linear estimation of constants was at that time a
comparatively new topic and he draws attention
to some of the work that was going on. In con-
nection with the interparticle field he remarked
on the calculation of temperature and other pro-
files in the bed, the subject of one of his earlier
papers (1943). However, he quickly turned to the
question of the interphase field by which he meant
the diffusion-reaction problem at the surface of
a particle, then to the internal diffusion problem,
showing how the Thiele modulus could be recog-
nized by considering the relative rates of reaction
and diffusion. In particular he rejoiced in the prac-
tical test for diffusional effects which had been
proposed by Weisz [38].
To sum up I think it is fair to say that of
the pioneers of chemical reaction engineering
Wilhelm stands out for his marriage of the con-
cept of a priori design, as he called it in his 1962
review, and the laboratory-scale experiments that
would be needed to understand the physical effects.
In contrast to Hougen and Watson and their
school he did not pursue reaction kinetics or at-
tempt to fit these data in any systematic way,
though he had a keen appreciation of the impor-
tance of catalysis as shown, for instance, by the
early appointment of Boudart to his department.
He saw the interaction of chemical and physical
rate processes as central to the design of chemical
reactors and established a research program that
would elucidate the different factors which might
bear upon it. He was not so concerned, as was
Amundson, with the dynamics or behavior of the
reactor though he was interested in the fluctuating
processes as they gave insight into the transport
properties. In the dominant part of his research

he chose to investigate the properties of the tubu-
lar and packed-bed catalytic reactors, but this
did not prevent him from looking at other types
as in his work on fermentors and, more especially,
in his continued interest in fluidized beds. His
1962 plenary lecture represents his mature re-
flection on the whole question of design. It is not
his last publication, for he was about to turn to a
fruitful preoccupation with parametric pumping,
but it may serve as a terminus ad quem. "I wish
to suggest," he there said in conclusion, "that
substantial progress has been made, in the case
of packed-bed reactors toward the goal of being
able to design such reactors from basic principles,
rapidly and for realistic systems. The supporting
physical and engineering sciences and the develop-
ment of computational capabilities are sufficiently
well advanced to encourage computational design
explorations, at least in selected cases. A major
problem, always, is to secure reliable chemical
rate data and any wide-spread trend toward com-
putational simulation may well serve to encourage
the undertaking of basic laboratory chemical rate
studies even more than at present. A need for
continuing research in certain engineering as-
pects of chemical reactors is indicated." And in-
deed, the work has gone on well,* it is now cele-
brated by a biennial International Symposium and
perhaps, more than any other branch of chemical
engineering, has shown its intrinistic vitality-a
spur to engineering research, an incentive to phys-
ical and chemical experimentation, and both the
stimulator and benefactor of applied mathematics.

The last time I recall meeting Wilhelm was
in the spring of 1965. His thoughts had been turn-
ing to parametric pumping for some time and in
connection with this he had more to do with bio-
logical workers than before. In May of that year
a conference was held under the auspices of the
International Society for Cell Biology at the Villa
Falconieri. A very pleasant villa it was, situated
above the hill town of Frascati and overlooking
the plain that was once the northern end of the
Pomptine Marshes and, in the distance, the south-
ern aspect of the city of Rome. We assembled there
from all sorts of places one sunday evening in May,
seeing old friends and meeting new acquaintances.
I remember Wilhelm's eager interest in what

*This growth is illustrated by the multiplication of
texts ,and monographs on chemical reaction engineering.


each was doing and the unpretentious enthusiasm
with which he spoke of his own work. I recalled
that occasion when, a few years later, I read Dick
Toner's beautifully written Memorial Statement
in the issue of I. and E.C. Funds. in May of
1969. He drew attention to the very meetness of
an issue filled with papers written by former
students, colleagues and friends who had worked
with Wilhelm during his lifetime. "For it is
through what these men are now doing", Toner
wrote, "and will continue to do that his influence
on chemical engineering will continue to be ex-
erted. It is likely that he would have regarded
this Memorial Issue as the greatest honor ever
accorded to him." U


1. T. Baron. Chem. Eng. Prog., Symposium Series 48
118 (1952).
2. B. G. Bakhurov and G. K. Boreskov. J. Appl. Chem.
(USSR) 20, 21 721 (1947).
3. G. A. Latinen Ph.D. Dissertation, Dept. of Chem.
Eng., Princeton University, Princeton, New Jersey.
4. E. Singer, Ph.D. Dissertation, Dept. of Chem. Eng.,
Princeton University, Princeton, New Jersey.
5. J. Beek and R. S. Miller. Chem. Eng. Prog., Sympo-
sium Series 55, 23 (1959).
6. R. Aris and N. R. Amundson. AIChE. Journal, 3 280
7. H. Kramers and G. Alberda. Chem. Eng. Sci., 2, 173
8. W. Nernst and E. Brunner. Z. physik. Chem. 47, 52
9. I. Langmuir. J.A.C.S. 30, (1908).
10. N. Bodenstein and K. Wolgast. A. physik. Chem.
(Leipzig) 61 422 (1908).
11. F. Jiittner. Z. physik. Chem. 65, 595 (1909).
12. J. Taffanel and S. LeFloch. Comptes Rend. Acad. Sci.,
Paris 155, 1544 (1913); 157, 469, 595, 714 (1913).
13. F. G. Liljenroth. Chem. Met Eng. 19 287 (1918).
14. C. M. Tu, H. Davis and H. C. Hottel. Ind., Eng. Chem.,
26, 749 (1934).
15. G. DamkShler. Z. Electrochem. 42, 846 (1936); 43, 1
(1937); 44, 193, 228 (1938).
16. E. W. Thiele. Ind. Eng. Chem. 31, 916 (1939).
17. J. Zeldowitch. Act. Phisicochim. U.R.S.S. 10, 583
18. E. Wicke. Z. Elektrochem. 44, 587 (1938).
19. J. Zeldowitch and Y. A. Zysin. Z. Techn. Fiz. 11,
502 (1941).
20. C. Wagner. Die Chemische Technik 18, 28 (1945).
21. K. G. Denbigh. Trans. Faraday Soc., 40, 352 (1934);
44, 479 (1948).
22. E. Wicke and W. Britz. Chemie-Ing.-Techn. 21, 219
23. 0. A. Hougen. Chem. Eng. Prog. Monograph Series
No. 1, 47, 1-74 (1951).
24. C. vanHeerden. Ind. Eng. Chem., 45, 1242, (1953).
25. N. R. Amundson and 0. Bilous. AIChE Journal 1,
513 (1955).

26. A. Wheeler. Catalysis Chapter II (Ed. P. H. Emmett)
Rhinehold Publishing Co. (1954).
27. N. R. Amundson and R. Aris. Chem. Eng. Prog. 53,
227 (1957); and Chem. Eng. Sci., 7, 212 (1958).
28. J. J. Carberry, AIChE. Journal, 7, 350 (1961).
29. P. B. Weisz and J. S. Hicks. Chem. Eng. Sci., 17,
265 (1962).
30. N. R. Amundson and R. E. Schilson. Chem. Eng. Sci.,
13, 226, 237 (1961).
31. J. D. Tinkler and A. B. Metzner. Ind. Eng. Chem.,
53, 663 (1961).
32. N. R. Amundson and R. A. Schmitz. Chem. Eng. Sci.,
18, 265, 391,415,447, (1963).
33. K. B. Bischoff. AIChE. Journal, 11, 351 (1965).
34. E. E. Petersen. Chem. Eng. Sci., 20, 587 (1965).
35. R. Aris. Ind. Eng. Chem. Fundamentals 4, 227 (1965).
36. N. R. Amundson and D. Luss. Ind Eng. Chem. Funda-
mentals Vol. 6, 436 (1967).
37. L. Lapidus and H. A. Deans. AIChE. Journal, 6, 656
38. P. B. Weisz. Z. physik Chem. 11, 1 (1957).
39. P. Gray, Ber. Bunsen-Gesell, physik. Chem. 84, 309


Bibliography of Richard H. Wilhelm
a. (with D. M. Wroughton and W. F. Loeffel) "Flow of
Suspensions through Pipes", Ind. Eng. Chem. 31, 622
b. (with D. M. Wroughton) "Concentric-Cylinder Motor-
Driven Viscometer", Ind. Eng. Chem. 31, 482
a. (with L. H. Conklin and T. C. Sauer) "Rate of Solu-
tion of Crystals", Ind. Eng. Chem. 33, 453
b. "Simply Constructed Color Comparator", Ind. Eng.
Chem. 13, 123
(with W. C. Johnson and F. S. Acton) "Conduction,
Convection and Heat Release in Catalytic Converters",
Ind. Eng. Chem. 35, 562
a. (with D .W. Collier) "Vapor-Liquid Equilibria of the
System Budadiene Styrene", Ind. Eng. Chem. 40, 2359-
b. (with G. F. Paul) "Radiation Drying of Textiles", Tex-
tile Research Journal, Vol. XVIII, 573-597
c. (with M. Kwauk) "Fluidization of Solid Particles",
Chem. Eng. Prog. 44, 201-218
d. (with W. C. Johnson, R. Wynkoop and D. W. Collier)
"Reaction Rate, Heat Transfer, and Temperature Dis-
tribution in Fixed-Bed Catalytic Converters", Chem.
Eng. Prog. 44, 105-116
a. "Fluidization Nomenclature and Symbols", Ind. Eng.
Chem. 41, 1249-1250
b. (with L. McCune) "Mass and Momentum Transfer in
Solid-Liquid System", Ind. Eng. Chem. 41, 1124
c. "Rate Process-Application to Reactor Design", Chem.
Eng. Prog. 45, 208-218
d. (with J. B. Smith) "Transmittance, Reflectance, and
Absorptance of Near Infrared Radiation in Textile
Materials", Textile Research Journal, Vol. XIX, 73-88


a. (with W. H. Bartholomew, E. O. Karow and M. R.
Sfat) "Oxygen Transfer and Agitation in Submerged
Fermentations-Mass Transfer of Oxygen in Sub-
merged Fermentation of Streptomyces griseus", Ind.
Eng. Chem. 42, 1801
b. (with W. H. Bartholomew, E. O. Karow and M. R.
Sfat) "Oxygen Transfer and Agitation in Submerged
Fermentations-Effect of Air Flow and Agitation
Rates upon Fermentation of Penicillium chrysogenum
and Strentomyces griseus", Ind. Eng. Chem. 42, 1810
c. (with R. K. Toner) "Kinetics and Equilibria", Ind.
Eng. Chem. 42, 1644
d. (with E. Singer) "Heat Transfer in Packed Beds.
Analytical Solution and Design Method-Fluid Flow,
Solids Flow, and Chemical Reaction", Chem. Eng.
Prog. 46, 343-357
e. (with R. Wynkoop) "Kinetics in Tubular Flow Re-
actor-Hydrogenation of Ethylene Over Copper-
Magnesia Catalyst", Chem. Eng. Prog. 46, 233-244
f. "Fluidization of Masses of Particles", Research
a. (with R. K. Toner) "Kinetics and Equilibria", Ind.
Eng. Chem. 43, 1898
b. (with S. Valentine) "The Fluidized Bed-Transition
State in the Vertical Pneumatic Transport of Parti-
cles", Ind. Eng. Chem. 43, 1199-1203
a. (with P. F. Deisler) "Diffusion in Beds of Porous
Solids-Measurement by Frequency Response Tech-
niques", Ind. Eng. Chem. 45, 1219
b. "Rate Theory and Homogeneous Reactions", Ind. Eng.
Chem. 45, 894
c. "Rate Processes in Chemical Reactors", Chem. Eng.
Prog. 49, 150-54
"Rate Theory and Homogeneous Reactions", Ind. Eng.
Chem. 46, 880
(with P. F. Deisler and K. W. McHenry) "Rapid Gas
Analyzer Using Ionization By Alpha Particles", Anal.
Chem. 27, 1366
a. (with J. F. Wehner) "Boundary Conditions of Flow
Reactor", Chemical Engineering Science 6, 89-93
b. (with T. J. Hanratty and George Latinen) "Turbulent
Diffusion in Particulately Fluidized Beds of Particles"
AIChE Journal 2, 372-380
c. (with J. M. Prausnitz) "Turbulent Concentration Fluc-
tuations through Electrical Conductivity Measure-
ments", The Review of Scientific Instruments 27, 941-
d. (with F. A. Cleland) "Diffusion and Reaction in
Viscous-flow Tubular Reactor", AIChE Journal 2, 489-
a. (with J. M. Prausnitz) "Turbulent Concentration Fluc-
tuations in a Packed Bed", Ind. Eng. Chem. 49, 978
b. (with L. Lapidus and J. B. Rosen) "Flame Propaga-
tion Rates. Chemical Nature of Attachment Surface",
Ind. Eng. Chem. 49, 1181

c. (with K. M. McHenry) "Axial Mixing of Binary Gas
Mixtures Flowing in a Random Bed of Spheres",
AIChE Journal 3, 83-91
a. (with Fritz Fetting and A. P. Roy Choudhury) "Turb-
ulent Flame Blow-Off Stability Effect of Auxiliary
Gas Addition into Separation Zone", Combustion Inst.,
b. (with William Rice) "Surface Dynamics of Fluidized
Bed and Quality of Fluidization", AIChE Journal 4,
(with Frank Hill) "Radiative and Conductive Heat
Transfer in Quiescent Gas Solid Bed of Particles;
Theory and Experiment", AIChE Journal 5, 486-496
(with D. E. Lamb and F. S. Manning) "Measurement
of Concentration Fluctuations with an Electrical Con-
ductivity Probe", AIChE Journal 6, 682-684
(with Ruxton H. Villet) "Knudsen Flow-Diffusion in
Porous Pellets", Industrial and Engineering Chemistry
53, 837-840
"Progress Towards the a Priori Design of Chemical
Reactors", Pure and Applied Chemistry, 403-420
a. (with F. S. Manning) "Concentration Fluctuations in
a Stirred Baffled Vessel", AIChE Journal 9, 12-19
b. (with David E. Lamb) "Effects of Packed Bed Prop-
erties on Local Concentration and Temperature Pat-
terns", I&EC Fundamentals 2, 173-182
(with E. H. Blum) "A Statistical Geometric Approach
to Random-Packed Beds", AIChE-I Chem.E. Sympo-
sium Series, No. 4, 21-27
a. "Parametric Pumping: A Model for Active Transport",
Intracellular Transport (Symposia of the International
Society for Cell Biology, Volume V), Academic Press,
New York
b. (with W. A. Donohue, D. J. Valesano and G. A. Brown)
"Gas Absorption in a Stirred Vessel: Locale of Trans-
fer Action", Biotechnology and Bioengineering 8, 55-69
c. (with A. W. Rice and A. R. Bendelius) "Parametric
Pumping: A Dynamic Principle for Separating Fluid
Mixtures" I&EC Fundamentals 5, No. 1, 141-144
a. (with Alan W. Rice, Roger W. Rolke, and Norman H.
Sweed) "Parametric Pumping: A Dynamic Principle
for Separating Fluid Mixtures", I&EC Fundamentals
7, 337-349
b. (with N. H. Sweed) "Parametric Pumping; Separation
of a Mixture of Toluene and n-Heptane
a. (with N. H. Sweed) "Parametric Pumping: Separa-
tions via the Direct Thermal mode" I&EC Fundamen-
tals. 8, 221-231
b. (with R. W. Rolke) "Recuperative Parametric Pump-
ing" I. & E.C. Fundamentals 8, 235-246


Continued from page 19.
quently means that the difficulty of performing
calculations obscures the importance of the
chemistry. The examples which are chosen must
therefore involve some simplifying approxi-
mations. These simplifications may be introduced
by the instructor, but in many cases the key idea
in making the simplification can be suggested by
the students. For example, feedstocks for petro-
leum refining consist of a very large assortment
of paraffins, naphthenes, aromatics, and hetero-
aromatics. Reforming this complex mixture in-
volves many reactions. However, a grouping of
these reactions into paraffin isomerization, naph-
thene isomerization, paraffin dehydrogenation,
naphthene dehydrogenation, hydrocracking, and
dehydrocyclization permits a reasonable choice of
temperatures and pressures to be made on the
basis of thermodynamics. Students can calculate
the effect of temperature and pressure on
equilibrium constants for the reaction of repre-
sentative components in a naphtha stream.
Similarly, calculations for coal processing routes
cannot proceed quickly without significant
simplifications in the representation of the com-
position of coal. One of the most drastic sug-
gestions (which can be used most successfully) is
to consider coal to be essentially carbon. There are
pitfalls in these calculations, and the tentative
nature of this approach must be emphasized. How-
ever, such methods can successfully reinforce the
importance of quantitative reasoning. In addition,
they illustrate that crucial chemical insights con-
cerning composition and reactivity guide the cal-
Fundamental concepts of stoichiometry,
thermodynamics, kinetics, reactor design, and heat
and mass transfer provide a basis for performing
calculations. The calculations are not as detailed
as those encountered in a traditional design
course; rather, an emphasis is placed on "screen-
ing" calculations, the precursors of equipment
design calculations. These are in close juncture
with the processing chemistry. Thermodynamic
calculations are useful in evaluating stoichiometry
or predicting product distributions. An excellent
example is synthesis gas conversion. The stoi-
chiometry may be written to indicate methanol
production, but methane, higher alcohols, and
paraffins may all be present in significant amounts
under industrial conditions. The effect of tempera-
ture and pressure on these reactions can be easily

quantified. Similarly, the heat released in the
nitration of benzene can be determined; by com-
bining this information with activation energies,
students can demonstrate the potential for re-
action runaway. The heat released in this reaction
clearly illustrates the importance of heat transfer
and control in the creation of a process flowchart.
An illustration involving underground coal
gasification is given in the Appendix. This problem
is typical of systems involving feedstocks with
complicated chemical compositions. By using a
simplified composition, the stoichiometry of
several key reactions can be written. Since the
key gasification reactions at low temperature (me-
thane formation) are endothermic, other reactions
(such as combustion) are required to provide the
necessary energy. The students are asked to de-
termine whether it is feasible to gasify a particu-
lar coal. The open-ended nature of this question is
intentional. The results depend on the particular
choice of reactions. Similar thermodynamic
criteria may be introduced in ammonia synthesis
in demonstrating the effect of temperature or
pressure on equilibrium yields or in determining
the importance of carbon monoxide impurities.
The importance of the reverse water-gas shift
reaction in controlling the temperature in
methanol synthesis may also be clarified using
elementary thermodynamics.

The emphasis in the course on synthesis and
invention creates a sharp contrast with other
chemical engineering courses. Classroom activi-
ties are similarly distinct. We have scotched the
traditional lecture format for a much more in-
formal orientation. Central to the approach are
oral expression and exchanges, spontaneity,
rapid changes and creation, and competition. All
classroom techniques are designed to maximize
the students' participation in the conceptualiza-
tion of processes. The intent is to devise an en-
vironment more consistent with industrial ex-
perience. The style, language, and pace of in-
dustry are much different from those of the con-
ventional classroom, and we have attempted to
provide a transition to the industrial environment
in this course.
The case study approach is useful in promot-
ing this atmosphere. Typically, only one or two
periods are spent on a process. The case studies
are team taught, and student participation is en-
couraged by frequent alternations in faculty


leadership. All participating faculty are usually
present in the classroom; their presence is in-
valuable in posing questions about processes, in
suggesting alternatives, and in encouraging in-
Spontaneity is the most distinctive character-
istic of the classroom sessions. The classroom
functions as a model of the industrial "brainstorm-
ing" session. We attempt to create this atmosphere
in several ways. We encourage "off-the-top-of-the-
head" ideas; rapid-fire question-and-answer par-
ticipation is included. Ideas are sorted out and
recorded quickly to prevent the loss of any con-
cepts. Critical evaluation is particularly avoided
at some points in the course. Ideas are solicited
from all members of the class (making small
classes helpful). Eventually, the spontaniety must
be managed in a give-and-take atmosphere. The
instructor can adeptly allow "dull" ideas to be left
aside. And absolutely wrong ideas (e.g., fraudu-
lent thermodynamic reasoning) must be clearly
identified as being erroneous.
Competitive events are included-again re-
flecting an industrial flavor. Small groups are
frequently formed in the class, and these groups
typically present their ideas during the class.
The efforts of one group in a particular event are
shown in Figure 2. In this circumstance, groups
of about five students were required to create a
flowchart for reforming of naphtha. The creation
was to proceed by stages: there were three time
periods of about ten minutes during which the
groups could create and modify flowcharts. Each
of these periods was followed by an oral presenta-
tion of the flowchart and the basis for the reason-
ing. Critical evaluation by the students and
faculty was performed to determine key innova-
tions or errors. (This class period had been pre-
ceded by a discussion of the basic chemistry of re-
forming.) The results for most groups have been
very close to the actual flowsheet for reforming.
Other competitive events are also employed.
Alternative processes may be presented by faculty
advocates. Strong claims and contradictions can
be introduced. Students are then asked to present
their own evaluations. These roles can also be
reversed. Students can be asked to defend specific
aspects of given processing chemistry. One suc-
cessful event has dealt with rival choices for a
catalyst. Two instructors (each a "salesman" for
a catalyst) were permitted to answer specific
questions about their catalysts, but only questions
raised by the students. Thus, the students had to


leed stocks
teed aks Fixed Bed
exchange heavy ends

Feedstocks He --
-- Hydroprocessrmg
..H2 ....
HB benzene
Fxied Be
a a, toa'

hl ate
heat hydrocarbons
exchange heat exchange heal exfange

* Caalyst is helerogeneous Pl r-Aezy0
Overall reaction Is endolherml. requ ng a
substantial heat lnput
* Reformng is a net hydrogen producer

SHydropocesslng introduced to project uelormlng
catalyst Irom suflur and metals (can use Hnproduced)
* External heat exchange (or fixed bed reactor
possiblyy with produclsl
* Inlernmedale about 10 aim I hydrogen pressure
In reactor la balance between inhbitirng coke
tormatlon and an undesirable effect on aremaati(n
* catalyst regeneration is necessary because of coke
formation (not clear how to do Ihis)
SAddition of chlorinated hydro-carnns to reserve
acidity ol calalyst

* multiple rixed bed allows better control oI
tempera ure Iin range ol 400-500 C
Sregeneaion coked caalys with small amounts
of oxygen Iurn oil coke lemperaure control
important to avoid slnl erng ol catalystn perhaps
.Ith a .w-nq reactor


cross-examine the instructors, and the successful
"purchase" of the catalyst at the end of the class
period depended on the students' ability to pose
the "right" questions concerning the catalytic


Patents are a helpful tool in generating class-
room discussion and in developing homework
problems. A brief introduction to patents is pro-
vided by local patent lawyers who are active in
the chemicals area. The basis for patentable work,
reading the patent literature, and patent law are
discussed. This portion of the course complements
the emphasis on innovation by stressing the
creation and recognition of new ideas.
Several industrial processes offer excellent
examples of innovations recorded in the patent
literature. Processes for converting synthesis gas
to methanol and higher alcohols have a long
patent history. We have used a short series of
these patents for discussing the chemistry and
engineering of these processes. The patents il-
lustrate the use of different metal components of
the catalyst to reduce the temperatures or pres-
sures required within the reactor. One patent also
introduces the ideas of using CO2 as a moderator
(via the water-gas shift reaction) to control the
reactor temperature. A controversial patent sug-
gests using water as a replacement for hydro-
gen in producing methanol. Students were
challenged to distinguish between the patents.


We have scotched the traditional lecture
format for a much more informal orientation. Central
to the approach are oral expression and
exchanges, spontaneity, rapid changes
and creation, and competition.

Thermodynamics can frequently be used to evalu-
ate the patents and, in some cases, kinetic informa-
tion is available.
Patents also provide a basis for homework
problems. A typical problem involves the creation
of a flowchart based on a patent for acetic acid or
ethylene oxide. Several longer term projects of
this nature have also been used.

Much emphasis in the course is placed on oral
performance and evaluation. This nontraditional
method results in a more spontaneous response
from students and a better appreciation of the
conceptualization process. Oral reports and
presentations by students are frequently used to
provide opportunities for classroom interactions.
Student contributions are further encouraged by
the sharing of industrial experience; frequently
a rather high proportion of the class has worked
in industry during previous summers. Graduate
students with industrial experience also partici-
pate in the course.
Midterm examinations are conducted orally.
Students are assigned a chemical compound and
are sent to the library to determine "everything
that is important" concerning the processing for
its manufacture. Students are assigned topics
sequentially to avoid overlap in the library. After
two hours, the student returns to give a half-hour
presentation to the faculty instructors. We have
found that this format works very well with two
or three faculty members present, rather than
just one; however, the faculty time commitment
becomes large. Students are usually able to find
relevant resource material and the basic pro-
cessing information. The majority of students
can also reason using basic thermodynamics
and kinetics. Of course, much more than this is
desired, and the questions posed by the faculty in-
variably lead the more able student to some
aspects of industrial chemistry which were un-
researched or which are unknown. These probing
questions are the most difficult and also the most
instructive. Some students are shaken by the ex-

perience-usually to their benefit! Generally,
students have responded favorably to this ap-
A final project is also used in which the
students are required to work independently of
known solutions. Students are organized into
groups, but each individual is required to submit
a written report. The projects typically involve
new or unconventional processes. Recent patents
are used, such as the production of vinyl acetate
from carbon monoxide and hydrogen. Another
example of a successful final project is the cleanup
of chemical dumps. Here the student is confronted
with a myriad of possible reactions in an un-
familiar environment (soil). Much discussion
centers on identification of the problem (com-
position of materials and possible reactions) and
many alternative solutions are possible. We have
also used projects dealing with natural products
chemistry, enzyme chemistry, and photochemistry
-again, areas relatively unfamiliar to the

While we have found that several references
are useful in providing background information,
no single book is suitable as a text. Class notes are
available for some portions of the course. Students
refer frequently to their previous chemistry and
engineering textbooks. The following is a list of
references which provide some helpful material:
1) "Kirk-Othmer Encyclopedia of Chemical Tech-
nology," Second Edition, John Wiley & Sons, New
York, 1963. (Volumes of Third Edition, 1978, ap-
provides a good discussion of general processing
technology and equipment, some flowsheets pro-
vided, limited discussion of processing chemistry.
2) A. L. Waddams, "Chemicals from Petroleum,"
Third Edition, John Wiley & Sons, New York, 1973.
good overview of petrochemical industry, pro-
vides good overall flowcharts, some processing
3) P. Wiseman, "An Introduction to Industrial Organic
Chemistry," John Wiley & Sons, New York, 1976.
good perspective on chemistry involved in im-
portant organic processes, orientation toward
British routes (coal given larger discussion).
4) K. Weissermel and H.-J. Arpe, "Industrial Organic
Chemistry," Verlag Chemie, New York, 1978.
good collection of industry-wide flowcharts,
alternative chemical routes suggested with some
processing chemistry included.
5) H. A. Wittcoff and B. G. Reuben, "Industrial
Organic Chemicals in Perspective," Parts I and II,
John Wiley & Sons, New York, 1980.
provides a broad orientation for the chemistry


for industrial process, little discussion of flow-
charts or engineering requirements.
6) S. A. Miller, Ed., "Ethylene and its Industrial
Derivatives," Ernest Benn, London, 1969.
E. G. Hancock, Ed., "Propylene and its Industrial
Derivatives," Halsted, New York, 1973.
much of the petrochemical industry is described
in these books.
7) J. T. Maynard, "Understanding Chemical Patents,"
Americal Chemical Society, Washington, D.C., 1978.
an excellent, short introduction to patents.
8) E. Mansfield, J. Rapoport, A. Romeo, E. Villani, S.
Wagner, and F. Husic, "The Production and Appli-
cation of New Industrial Technology," W. W.
Norton, Inc., New York, 1977.

Although these references provide some use-
ful information, we have found that a very effec-
tive method for introducing an industrial orienta-
tion is through industrial speakers. These practi-
tioners have had extensive experience with a par-
ticular process. Typically, they are industrial
organic or inorganic chemists and possess an
excellent understanding of industrial technology,
economics, the availability of raw materials, and
the changing nature of chemical markets. These
interactions are also stimulating to classroom dis-

Problem Statement
Underground coal gasification is typically per-
formed at lower temperatures (for example, 700
K) than the Lurgi process which was discussed in
class. The product from such an operation is re-
ferred to as "high Btu" gas. The gasification is
performed by injecting oxygen and steam into
coal beds which have been drilled or fractured.
Typical annual consumption rates for a 100 billion
cubic feet plant are as follows:

Gas produced
Coal consumed
Oxygen consumed
Water consumed
Drill holes

100 BCF
5.05 million metric tons
1.53 million metric tons
1.54 million metric tons

A typical Wyoming coal ("Thunderbird coal")
has the following analysis:



The heat of combustion of this coal is 1.33 x
104 Btu/lb (138 kcal/mole, based on the formula
Determine if it is feasible to gasify this coal.

Solution Comments:
In order to solve this problem, the student
must determine the meaning of the criterion
"feasible". The previous classroom discussion pro-
vided an important clue: in a Lurgi gasifier, there
are regions of drying and devolatilization, com-
bustion, and synthesis gas production. Similarly
in this problem, coal must be combusted to pro-
vide the heat required by the endothermic re-
actions. Perceptive students also realize that there
is a significant heat demand due to drying the
coal and to heating the coal bed and the surround-
ing environment; some estimates of these heat
effects can be made.
In order to determine the grams of coal which
must be combusted per gram of coal converted to
produce gas, it is necessary to first determine the
stoichiometry of the reactions. The composition
of this coal may be expressed as CH1.150o.3.
N0.0146So.oo51 (providing the formula weight for
heat of combustion). The student may be
tempted at this point to write a large number of
reactions of this "species" with 02, H20, CO, CO2,
H2, H2S, etc. This approach is further frustrated
by the difficulty of obtaining thermodynamic
properties for the coal and related coal-derived
intermediates: such an approach would be useful
only if this data were available.
In previous classroom discussion, the
possibility of considering coal to be essentially
carbon was discussed. (Frequently, this "dis-
covery" is suggested by a student.) The students
are much more likely to be able to evaluate this
process if such an approximation were made.
The chemical species which therefore are con-
sidered to be involved in the gasification process
are C, 02, H2O, CO, CO,, H, and CH4. Four inde-
pendent reactions may be written for these


C + 2H2 ; CH,
C + HO CO + H2
CO + H2O ~ CO, + H2
C + 02 CO2

The equilibrium composition of the product gas
may be determined from thermodynamic calcula-
tions. However, an additional simplifying assump-


tion may be used. The fourth reaction involves
combustion of the carbon to carbon dioxide. Com-
plete combustion of other product gases also
occurs in the presence of oxygen at these tempera-
tures. The equilibrium constants for these re-
actions are very large; similarly, rate data which
was presented in class indicated that the rate
constants for reactions between carbon and
oxygen were very large. The fourth reaction
might be excluded from consideration since the
presence of excess oxygen at specific regions in
the bed would likely lead to complete combustion
products. The heat derived from such reactions, of
course, is necessary to balance the endothermic
reactions involved in the production of high Btu
The composition of the product gas may be de-
termined by thermodynamic calculations. At 700
K, the equilibrium constants are K, = 22.6, K2 =
1.60 x 10-3 and K, = 7.31. The corresponding gas
phase composition is ycH4 = 0.20, YH,o = 0.44,
Yco2 = 0.25, YH, = 0.094, and yco = 0.0074.
The results of the calculation indicate that
the heat required due to the overall endothermic
nature of the gasification reactions is relatively
low. Substantial amounts of energy are required
to heat the coal to 700 K and vaporize the moisture
in the coal bed. Students are also capable of pro-
viding a rough estimate of the heat loss to the
surrounding environment. The effect of approxi-
mating the coal composition is relatively small.
The results of the calculations vary as a result of
the particular assumptions made, but roughly
15-20% of the coal would have to be burned. O

Continued from page 27.
curious mixture of brevity and detail. A great
deal of straightforward algebraic detail is given
in many places while in others mastery of far
more difficult concepts is assumed. For example,
knowledge of eigenvalues and eigenvectors of
matrices is assumed in the earlier part of Chapter
3, but is reviewed in some detail later in the
same chapter. Far more background in stochastic
processes is assumed than the typical chemical
engineer has. Although little theory of stochastic
processes is actually needed, the jargon is used
extensively. The reader is not given enough back-
ground for even a clear physical interpretation
of the results. Such a commonly used concept as
the expected value of a stochastic variable is

never even defined. Notation is also a problem in
places, with the same symbols being used for
different quantities or different symbols used for
the same quantities within the same chapter.
In spite of some shortcomings Ray's latest book
is highly recommended. It is by far the best book
available for a graduate-level level course in
modern chemical process control. E

By J. Horak & J. Pasek
Heyden & Son, Philadelphia, PA
Reviewed by Moin Ahmed
Union Carbide Corporation

This book is an addition to a large number of
books on the design of chemical reactors. How-
ever, this book differs from many other text books
by emphasizing the practical aspects, sometimes
at the expense of needed theory. The book touches
some subjects like analytical methods and sta-
tistical methods of data evaluation which most
books on reactor design do not address.
The book delves into useful qualitative dis-
cussion of many design principles, design methods
and reactor descriptions. However, in a number
of places a more mathematical and less empirical
approach would have been useful. Most of the
examples are of a qualitative nature, and there
are very few examples which emphasize more than
one principle at a time. At least on one occasion
the book is misleading, referring to free energy
of reaction as enthalpy of reaction. The trans-
lator has often used terms that are not familiar
to American readers (like technological properties
of a reactor).
The book is not well organized and is divided
into too few chapters. After an introductory
chapter, there is a chapter on Reactions in Solu-
tions which is actually a chapter on homogeneous
reactions. This chapter is followed by a chapter
titled, Types of Reactors. In addition to the title,
it describes data collection, treatment and regres-
sion of data, determination of specific heats and
heats of reaction, and a very good description of
scale-up techniques referred to by the authors as
Data Transfer. Chapter 4 deals with the catalytic
reactors. This chapter also presents the trickle
bed reactors, heat transfer media and construc-
tion of heat transfer loops-subjects very useful
to practicing which are neglected by most re-


actor design books. On the other hand, the authors
have barely touched upon catalyst deactivation,
and the discussion of intra particle diffusion limi-
tation and multiplicity in catalyst pellets is very
limited. The fifth and the last chapter is on Gas-
Liquid and Liquid-Liquid Reaction. The material
in this chapter is better presented than in most
other books and is very useful to the design engi-
The authors have completely neglected non-
catalyzed gas-solid reactions and a very common
class of reactors, namely slurry reactors.
The book can be useful as a reference book
for chemical engineers in industry but its utility
as an undergraduate or a graduate text seems very
limited. This is due not only to the lack of mathe-
matical approach but also because it contains no
problems for students to solve. Finally, the book
does not represent the state of the art since most
of the references are pre-1975. O

Continued from page 9.
in 1973 and 1980; and Pieter Stroeve and Dewey
Ryu, are the most recent additions.
While Joe has always emphasized to his col-
leagues that undergraduate education is a vital
part of the UC Davis chemical engineering pro-
gram, he has also promoted the ideal of strong
and varied graduate training. Although Joe is
planning a partial retirement to begin in the fall
of 1984, he expects to continue to teach and work
with graduate students and postdoctoral scholars
on chemical engineering research. In view of the
personal characteristics he has so far exhibited,
it is not anticipated that Joe will retire to a life
of leisure and abandon. His single-minded pursuit
of achievement in solving chemical engineering
problems is probably the critical factor in Joe's
success. Undoubtedly, the robust creativity and
inexhaustible energy have been important ingre-
dients as well, but his exacting commitment to
getting the job done and done well has made the
difference between routine and monumental ac-
In all the aspects of chemical engineering edu-
cation-exemplary research, the writing of text-
books, teaching classes, the guiding of students
(graduate and undergraduate) in their research,
and administration-Joe Smith has earned and
will continue to deserve his Davis title, "Mr. Chem-
ical Engineering". O

Continued from page 31.
tive control (ke = 3.53 volt/volt, TD = 0.1 sec-
onds), and proportional-integral-derivative con-
trol (ke = 3.53, Ti = 20, TD =0.1). Introduction
of integral action eliminates the offset (at much
longer time than shown in the graph), and less
oscillation is shown with P-I-D control.
Fig. 5 represents the effect of sampling time on
the system with only proportional control action.

0 10 20 3 40 50 60
TIME (Second.)

70 80 90

FIGURE 5. Effect of sampling time on system with
proportional control.
Increasing the sampling time resulted in higher
overshoot and more oscillation. What is interest-
ing is that the system which is first order starts
to act like a second order system with decreasing
damping coefficient as the sampling time in-
creases. Similar responses are obtained for P-I
and P-I-D control.
It should be pointed out that the present un-
dergraduate process control course at NJIT does
not cover direct digital control. With this experi-
ment, and the introduction of z-transforms, stu-
dents can get a very good understanding of dis-
crete sampling and direct digital control. E
Partial support for equipment was provided
by Exxon. The author acknowledges encourage-
ment by Prof. E. C. Roche, and appreciates the
work done by Mr. S. C. Chuang and Mr. S. Chari.
1. P. B. Deshpande and R. H. Ash, "Computer Process
Control With Advanced Control Application." Instru-
ment Society of America, 1981.
2. W. L. Luyben, "Process Modeling, Simulation and
Control for Chemical Engineers," McGraw-Hill, 1973.


Continued from page 5.
industrially subsidized stipends to pursue full-
time graduate study, preferably at the PhD level.
Both new programs have already had impact on
enrollment; thus, the 1981-82 enrollment of grad-
uate students in the department was twice that of
the previous year.
Faculty research interests are divided along
traditional lines: F. C. Alley (PhD, North Caro-
lina, 1962), environmental pollution control; W.
B. Barlage, Jr. (PhD, North Carolina State, 1960)
and D. D. Edie (PhD, Virginia, 1972), rheology
and polymer processing; J. N. Beard, Jr. (PhD,
LSU, 1971), industrial energy conservation and
process control; W. F. Beckwith (PhD, Iowa
State, 1963), transport phenomena and kraft-
pulping process; R. C. Harshman (PhD, Ohio
State, 1951) and R. W. Rice (PhD, Yale, 1972),
chemical reaction kinetics and catalysis; C. H.
Gooding (PhD, North Carolina State, 1979), sep-
aration processes; S. S. Melsheimer (PhD, Tulane,
1969), mass transfer, mathematical modeling, and
process control; J. C. Mullins (PhD, Georgia Tech,
1965) and J. M. Haile (PhD, Florida, 1976), ther-
modynamics and statistical mechanics.

Sheltered in the southeastern foothills of the
Appalachian Mountains, Clemson enjoys a mild,
four season climate. The environs of the univer-
sity provide unlimited opportunities for outdoor
recreation, including all types of fresh-water
sports, hiking, camping, cross-country running,
and hang-gliding. Cultural activities can be pur-
sued in Atlanta-a two hour's drive from Clemson.
Furthermore, there is the summer arts festival,
Spoleto, held annually in Charlestown, SC. This
is an offshoot from Gian-Carlo Menolti's festival
held in the Umbrian hill town of Spoleto, Italy,
and has evolved into a musical and operatic event
of some significance.
The combination of mild climate, moderately
priced electrical power, low taxes, and non-union-
ized labor is attracting important new industry
to upstate South Carolina. The cotton mills of the
first part of the century have long disappeared
from the region and have been replaced with plants
producing nylon, polyester, and polypropylene.

Clemson and surroundings.
Today, literally scores of textile mills dot north-
west Carolina and eastern Georgia. Moreover,
high technology electronic manufacturing com-
panies are moving into the area.
The future of Clemson University also seems
particularly promising. The decision of the ad-
ministration not to expand the student body nor
the physical plant during the student boom years
of the 1970's seems, in retrospect, to have been
a wise decision indeed. The demand for entry into
the university, as well as chemical engineering,
is higher than ever. Private and industrial support
for the university seems to be increasing substan-
tially. In 1981 alone, several important contribu-
tions were made to the university: Abney Foun-
dation gave one million dollars for the endow-
ment of a chair in Economics, National Cash Reg-
ister and Digital Equipment Corporation gave
generous research grants to the College of En-
gineering, and Senator Strom Thurmond donated
his public papers to Clemson. Plans are being
made for the creation of a Strom Thurmond Cen-
ter for Excellence in Government and Public Serv-
ice that will include a library to house Thurmond's
papers, a performing arts center, and a continuing
education facility.
During its ninety year history Clemson has
matured from a local agricultural college to a re-
gional university of some stature. Now, the visions
of both the university and its Department of
Chemical Engineering are beginning to extend
beyond the region to problems and opportunities
of national significance. This further maturing
process, while not without growing pains, prom-
ises to enhance and deepen the educational envi-
ronment at Clemson University. D



Departmental Sponsors: The following 141 departments contributed

to the support of CHEMICAL ENGINEERING EDUCATION in 1983 with bulk subscriptions.

University of Akron
University of Alabama
University of Alberta
Arizona State University
University of Arizona
University of Arkansas
Auburn University
Brigham Young University
University of British Columbia
Brown University
Itucknell University
California State Polytechnic
California State U. at Iong Beach
California Institute of Technology
University of California (Berkeley)
University of California (Davis)
University of California (Santa Barbara)
Carnegie-Mcllon University
Case-Western Reserve University
University of Cincinnati
Clarkson College of Technology
Clemson University
Cleveland State University
University of Colorado
Colorado School of Mines
Columbia University
University of Connecticut
Cornell University
University of Dayton
University of Delaware
U. of Detroit
Drexel University
University of Florida
Florida Institute of Technology
Georgia Technical Institute
University of Houston
Howard University
University of Idaho
University of Illinois (Urbana)
Illinois Institute of Technology
Institute of Paper Chemistry
University of Iowa
John Hopkins University
Iowa State University
Kansas State University
University of Kentucky
Lafayette College

Lamar University
Laval University
Lehigh University
Loughborough University
Louisiana State University
Louisiana Tech. University
University of Louisville
University of Maine
University of Maryland
University of Massachusetts
Massachusetts Institute of Technology
McMaster University
McNeese State University
University of Michigan
Michigan State University
Michigan Tech. University
University of Minnesota
University of Missouri (Columbia)
University of Missouri (Rolla)
Monash University
Montana State University
University of Nebraska
New Jersey Inst. of Tech.
University of New Hampshire
New Mexico State University
University of New Mexico
City University of New York
Polytechnic Institute of New York
State University of N.Y. at Buffalo
North Carolina State University
University of North Dakota
Northeastern University
Northwestern University
University of Notre Dame
Nova Scotia Tech. 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
University of Queensland

Queen's University
Rensselaer Polytechnic Institute
University of Rhode Island
Rice University
University of Rochester
Rose-Hulman Institute
Rutgers U.
University of South Carolina
University of Saskatchewan
South Dakota School of Mines
University of South Alabama
University of South Florida
University of Southern California
Stanford University
Stevens Institute of Technology
Syracuse University
Teeside Polytechnic Institute
Tennessee Technological University
University of Tennessee
Texas A&M University
Texas A&I University
University of Texas at Austin
Texas Technological University
University of Toledo
Tri-State University
Tufts University
Tulane University
University of Tulsa
University of Utah
Vanderbilt University
Villanova University
University of Virginia
Washington State University
University of Washington
Washington University
University of Waterloo
Wayne State University
West Virginia Inst. Technology
West Virginia University
University of Western Ontario
Widener College
University of Windsor
University of Wisconsin (Madison)
Worcester Polytechnic Institute
University of Wyoming
Yale University
Youngstown State University

TO OUR READERS: If your department is not a contributor, please ask your department chairman to write CHEMI-
CAL ENGINEERING EDUCATION, c/o Chemical Engineering Department, University of Florida, Gainesville, Florida




Braun & Co
is a world leader
in the engineering and
construction industry. For
more than 70 years, we have
provided a wide range of services to the
process and power industries.
Our principal fields of activity are
chemical and petrochemical plants, oil
refineries, ore processing plants, coal
gasification facilities, and power

ating sta-
tions. Many of our
projects have been first-
of-a-kind, utilizing new
processes never before employed
on a commercial scale. We also have been
involved in the emerging synfuels
Our rapid growth has opened up many
challenging opportunities and
assignments for professional growth.
Positions are available at our engineering
headquarters in Alhambra, California, and
at our eastern engineering center in
Murray Hill, New Jersey.





Full Text