Chemical engineering education

EDITORIAL AND BUSINESS ADDRESS:
Chemical Engineering Education
Department of Chemical Engineering
University of Florida * Gainesvile, FL 32611
PHONE and FAX : 352-392-0861
e-mail: cee@che.ufl edu

EDITOR
Tim Anderson

ASSOCIATE EDITOR
Phillip C. Wankat

MANAGING EDITOR
Carole Yocum

PROBLEM EDITOR
James 0. Wilkes, U. Michigan

LEARNING IN INDUSTRY EDITOR
William J Koros, Georgia Institute of Technology

- PUBLICATIONS BOARD --

* CHAIRMAN *
E. Dendy Sloan, Jr.
Colorado School ofMines

* MEMBERS *
Pablo Debenedetti
Princeton University
Dianne Dorland
Rowan University
Thomas F. Edgar
University of Texas at Austin
Richard M. Felder
North Carolina State University
Bruce A. Finlayson
University of Washington
H. Scott Fogler
University of Michigan
Carol K. Hall
North Carolina State University
William J. Koros
Georgia Institute of Technology
John P. O'Connell
University of Virginia
David F. Ollis
North Carolina State University
Ronald W Rousseau
Georgia Institute of Technology
Stanley I. Sandier
University of Delaware
Richard C. Seagrave
Iowa State University
C. Stewart Slater
Rowan University
Donald R. Woods
McMaster University

Chemical Engineering Education

Volume 38

Number 1

Winter 2004

D EDUCATOR
2 Chuck Eckert of The Georgia Institute of Technology,
William J. Koros

> DEPARTMENT
8 The University of Alabama,
C.S. Brazel, D.W. Arnold, G.C. April, A.M. Lane, J.M. Wiest

> LABORATORY
14 A Fluidized Bed Adsorption Laboratory Experiment,
Pamela R. Wright, Xue Liu, Benjamin J. Glasser
34 Nanostructured Materials: Synthesis of Zeolites,
Steven S.C. ( /........ Bei Chen, Yawu Chi, Abdelhamid Sayari
38 The Fuel Cell: An Ideal ChE Undergraduate Experiment,
Jung-Chou Lin, H. Russell Kunz, James M. Fenton, Suzanne S. Fenton

> CLASSROOM
22 On the Application of Durbin-Watson Statistics to Time-Series-Based
Regression Models, Thomas Z. Fahidy
26 Teaching Electrolyte Thermodynamics,
Simdo P. Pinho, Eugenia A. Macedo
54 Top Ten Ways to Improve Technical Writing,
John C. Friedly
64 Use of ConcepTests and Instant Feedback in Thermodynamics,
John L. Falconer
68 Rubric Development for Assessment of Undergraduate Research:
Evaluating Multidisciplinary Team Projects,
James A. Newell, Heidi L. Newell, Kevin D. Dahm
74 Teaching Engineering Courses with Workbooks,
Yasar Demirel

D RANDOM THOUGHTS
32 Changing Times and Paradigms, Richard M. Felder

> CLASS AND HOME PROBLEMS
48 Incorporating Green Engineering into a Material and Energy Balance
Course, C. Stewart Slater Robert P Hesketh

D LEARNING IN INDUSTRY
60 UOP-Chulalongkorn University Industrial-University Joint Program,
Santi Kulprathipania, Ann Kulprathipanja

21 Positions Available
31 Book Review

CHEMICAL ENGINEERING EDUCATION (ISSN 0009-2479) is published quarterly by the Chemical Engineering
Division, American Society for Engineering Education, and is edited at the University of Florida. Correspondence
regarding editorial matter, circulation, and changes of address should be sent to CEE, Chemical Engineering Department,
University of Florida, Gainesville, FL 32611-6005. Copyright � 2004 by the Chemical Engineering Division, American
Society for Engineering Education. The statements and opinions expressed in this periodical are those of the writers and not
necessarily those of the ChE Division, ASEE, which body assumes no responsibility for them. Defective copies replaced if
notified within 120 days of publication. Write for information on subscription costs and for back copy costs and availability.
POSTMASTER: Sendaddress changesto ChemicalEngineeringEducation, ChemicalEngineeringDepartment., University
of Florida Gainesville, FL 32611-6005. Periodicals Postage Paid at Gainesville, Florida and additional post offices.

Winter 2004

educator

Chuck Eckert

of
The Georgia Institute
of Technology

WILLIAM J. KOROS
The Georgia Institute of Technology
Atlanta, GA 30330
any people refer to Chuck Eckert as the "father of
modem supercritical processing technology." His
work over three decades ago on solvation and re-
action fundamentals under supercritical conditions helped re-
awaken chemical engineers to the opportunities within the
supercritical state. This reawakening has blossomed into a
rich subdiscipline that now encompasses much more than
reaction and partitioning processes. Indeed, many of the most
exciting topics now involve tailoring control of morphology
of complex solids such as pharmaceuticals and polymers-
items not initially envisioned even by Chuck.
Because he has had such a professional impact in chemical
engineering, I was surprised by Chuck's answer to my ques-
tion, "What do you consider your most important contribu-
tion?" With a twinkle and wink of his eye, he pointed to a
chart on his office wall. It comprised a "family tree" of indi-
viduals he has worked with through the years and who he felt
he had positively affected. He said that the list symbolized
his real life contribution-much better than any article or
discovery could. He noted that most practical developments
in the supercritical area were due to his students and their
students and post docs long after they had left his direct su-
pervision. The "family tree" that Chuck pointed to was pre-
pared on the occasion of his selection as winner of the 1995
ACS Murphree Award. The award dinner, where the family

tree was presented to him, brought together many of Chuck's
former students, post docs, and colleagues who celebrated a
career that had focused on coupled technical and personal
mentorship for many individuals. This coupled contribution
is truly his "signature" characteristic.
OVERVIEW
Chuck's 39 years in academia include 24 years at the Uni-

� Copyright ChE Division of ASEE 2004

Chemical Engineering Education

versity of Illinois and 15 at Georgia Tech. During this period
of time, he supervised an impressive 76 PhD dissertations
and 65 additional MS theses. This pace continues, with an-
other 10 PhD's still in progress. The names of his students
are shown in Table 1.

While numbers don't tell the full story, they underline the
truth in Chuck's perception that "people have been his proud-
est product." Ken Cox, a senior researcher at Shell Oil, has
said, "There is no individual, outside my family, who has
had such a major impact on my life. Strange thing is ... he
really is family! Many of the alumni from his research
groups at Illinois and Georgia Tech form a close family
for... Pappa Chuck!"

Another dimension of this picture is revealed by under-
standing the academic branches in the "family tree." Eigh-
teen current or retired academics have worked with Chuck as
either graduate students or post docs. Moreover, a probably
incomplete list shows six "academic grandchildren" who have
been educated by Chuck's direct academic descendents and
who should also be added to the list to bring this academic
branch up to at least 24.

Keith Johnston (UT Austin) says that "Chuck is totally dedi-

cated to the careers of his students." Similar sentiments come
from Barbara Knutson (U. Kentucky): "Chuck develops both
intellectual skills and people skills in his graduate students.
He has acted as my coach, my mentor, and a cheerleader long
after graduation, but most importantly, he is my friend. Chuck
has succeeded in creating a close academic family." Joan
Brennecke (Notre Dame), who won the 2001 ACS Ipatieff

Chuck accepts his "Family Tree: at the 1995 Murphree
award dinner.

TABLE 1
Chuck Eckert's Graduate Advisees

K. F. Wong (1969)
R. A. Grieger (1970)
S.P. Sawin (1970)
B. E. Poling (1971)
R. B. Snyder (1971)
F. G. Clark (1973)
J. H. Byon (1973)
J. R. McCabe (1973)
C. R. Hsieh (1973)
J. S. Smith (1975)
M. E. Paulaitis (1976)
B. A. Newman (1977)
G.L. Nicolides (1977)
P. G. Glugla (1977)
C. W. Graves (1977)
R. R. Irwin (1978)
A. Huss, Jr. (1978)
K. R. Cox (1979)
T. C. Long (1979)
P. K. Lim (1979)
E. R. Thomas (1980)
K. Kondo (1981)
M. M. Alger (1981)
K. P. Johnston (1981)
T. Stoicos (1982)
D. H. Ziger (1983)
P. C. Hansen (1984)
T. K. Ellison (1985)
J. C. Van Alsten (1985)
C. T. Lira (1986)

W. T. Chen (1986)
S. W. Gilbert (1986)
B. S. Hess (1987)
M. M. McNiel (1987)
H. H. Yang (1987)
W. J. Howell (1989)
D. M. Trampe (1989)
J. F. Brennecke (1989)
A. R. Hansen (1990)
A. M. Karachewski (1990)
M. P. Ekart (1992)
D. L. Tomasko (1992)
M. J. Hait (1992)*
D. B. Trampe (1993)
B. L. Knutson (1994)
D. Suleiman (1994)
D. L. Boatright (1994)*
F. L. L. Pouillot (1995)
K. P. Hafner (1996)
J. Berkner (1996)*
F. Deng (1996)*
A. Dillow (1996)
B. L. West (1997)
D. M. Bush (1997)
M. Vincent (1997)*
K. Chandler (1997)*
J. Jones (1998)*
N. Brantley (1999)*
Z. Liu (2000)*
J. Brown (2000)*

H. Lesutis (2000)*
K. West (2000)*
C. Wheeler (2001)*
K. Griffith (2001)*
V. Wyatt (2001)*
T. Ngo (2001)*
S. Nolen (2001)*
J. Hallett (2002)*
J. McCarney (2002)*
X. Xie (2003)*
T. Chamblee (in progress)*
M.Lazzeroni (in progress)*
R. Jones (in progress)*
N. Maxie (in progress)*
C. Thomas (in progress)*
J. Aronson (in progress)*
M. Janakat (in progress)*
R. Weikel (in progress)*
C. Pondey (in progress)*
L. Drauker (in progress)*
E. Giambra (in progress)*
J. Grilly (in progress)*
E. Newton (in progress)*

*Joint with C. L. Lzotta

K. F. Wong (1967)
R. A. Grieger (1968)
L. D. Clements (1968)
S. P. Sawin (1968)
L. G. Schornack (1969)
J. R. McCabe (1969)
F. G. Clark (1970)
J. H. Byon (1970)
C. R. Hsieh (1971)
K. P. Slaby (1971)
J. S. Smith (1972)
D. W. Wood (1972)
P. E. Walter (1972)
R. H. W. Powell (1973)
A. I. Ness (1974)
P. G. Glugla (1975)
C. W. Graves (1975)
A. Huss, Jr. (1976)
R. R. Irwin (1976)
B. A. Scott (1976)
T. C. Long (1977)
K. R. Cox (1977)
L. A. Halas (1977)
P. K. Lim (1977)
D. P. Deschner (1979)
E. R. Thomas (1979)
T. T. Oberle (1979)
K. P. Johnston (1979)
M. R. Anderson (1980)
T. Stoicos (1980)
D. H. Ziger (1980)

S. P. Brinduse (1981)
W. T. Chen (1982)
T. K. Ellison (1982)
P. C. Hansen (1983)
S. P. Singh (1983)
C. T. Lira (1983)
J. G. Van Alsten (1984)
S. W. Gilbert (1984)
M. M. McNiel (1984)
R. L. Matuszak (1985)
M. J. Hait (1985)
H. H. Yang (1985)
J. H. Cordray (1986)
W. J. Howell (1986)
D. M. Trampe (1987)
J. F. Brennecke (1987)
A. R. Hansen (1987)
A. Karachewski (1987)
S. R. Alferi (1989)
M. P. Ekart (1989)
D. L. Tomasko (1989)
P. Katsikopoulos (1990)
R. K. Denton (1990)
K. J. Hay (1991)
D. Suleiman (1992)
K. Chandler (1995)
R. Thompson (1996)
B. Eason (2001)*
D. Kass (in progress)*
D. Taylor (in progress)*

PhD Students

MS Students

Winter 2004

Prize, said of her Ipatieff award symposium that ". . . it was
the first time (and probably the only time) that the three most
important men in my life were all in one room- my fa-
ther, my husband, and Chuck! I think the continued care
and mentoring is why the Eckert academic family has so
many close ties."

In addition to the mentor in Chuck, however, there is a ma-
jor scholar who has produced well over two hundred archi-
val journal articles, coauthored two books, and contributed
twenty-one book chapters. One of his well-known coauthors,
and his PhD research mentor, John Prausnitz (Berkeley) ob-
served that, "Chuck communicates very well and encourages
others by his enthusiasm and optimism. He thoroughly ap-
preciates the importance of computers in research and edu-
cation. In 1967, it was primarily his enthusiasm that convinced
me to write with him (and two other graduate students) an
early monograph on the use of computer calculations for
multicomponent vapor-liquid equilibria-it was Chuck's fore-
sight and drive that accelerated the use of computers for ap-
plied thermodynamics in industry and education".

Chuck's contagious enthusiasm, tempered by a solid un-
derstanding of thermodynamics and thoughtful insights on

education, have made him attractive as an consultant and ad-
visor. Moreover, strategically placed ex-students, knowing
his catalytic capabilities have engaged him for services rang-
ing from conventional analysis to the motivational aspects of
education as well as research and its performance.
Chuck's current research interests include
* Molecular thermodynamics and solution theory
* Phase equilibria
* Supercritical fluid properties
* Applied chemical kinetics and catalysis
* Separation processes
* Environmentally friendly chemistry and processes
* Creation of novel materials
Many of Chuck's successes have resulted from his interest
in "crossing the street" and collaborating with chemists. His
work related to high-pressure reaction theory, the development
of solvency models and development of new spectroscopic ap-
proaches typify this characteristic. In many respects, the chem-
istry aspects of problems are the greatest attractions for him.
Chuck's approach involves a close coupling of experimen-
tal and theoretical attacks on problems. Prediction of limit-
ing activity coefficients in water using a modified separation

Chemical Engineering Education

Chuck with Amyn Teja and Ron Rousseau at a Georgia Tech
reception in honor of Chuck's 1999 Walker Award.

of cohesive energy density coupled with actual measurement of these limiting coef-
ficients illustrate the approach. The above work has provided important contribu-
tions to the understanding of "ordinary" liquids related to petrochemistry and even
liquid metals. Another related, but still independent interest involves Chuck's focus
on spectroscopic techniques to study hydrogen-bonding systems-this initiative
touches many areas in thermodynamics.
While the above work is well-known and highly note-worthy, probably Chuck's
best-known contributions relate to the gas-liquid critical region with particular refer-
ence to supercritical extraction and processing. With regard to the supercritical field,
Pablo Debenedetti (Princeton University) notes that, "Since 1983, Chuck has, with
unmatched regularity, made the key experimental observations and asked the truly
important questions that other researchers in the field need to answer". Indeed, in
1983, Chuck pioneered the measurement of solute partial molar volumes at infi-
nite dilution in supercritical solvents. In addition to its practical importance,
this ignited a large theoretical thrust across the field aimed at interpreting the
provocative results he reported.
In 1988, Chuck introduced the use of spectroscopic techniques to study solvation
in supercritical solvents. This pioneering work provided the first direct insights into
the nature of solute-solvent interactions and the mechanisms of solvation under
supercritical conditions. Focusing attention on short-range effects due to molecular
asymmetry was a key advance. This theme has been developed by a huge number of
subsequent researchers around the world
Still later, Chuck's identification of the role of cosolvents in separations and
supercritical processing marked another major contribution. The ability to design a
solvent for a specific reaction or separation application through manipulation of pro-
cess conditions or cosolvent type opened new possibilities and again stimulated many
studies within the field.
His broad and deep contributions to the chemical engineering literature were rec-
ognized in 1999 by the William Walker Award. Chuck is shown in the photo above
at an informal reception at Georgia Tech in his honor following his selection for
the Walker Award.
CONTRIBUTIONS TO THE COMMUNITY
Chuck's contributions to his home institutions are discussed later, but his profes-
sional contributions to the broader community also deserve mention. In addition to

. . . Chuck

notes [that]

"Research

is perhaps

the best

instructional

tool that

professors

have at

their

disposal -

the

one-on-one

creative

interaction

of real,

unsolved

problems

is the

best

method of

teaching

and

learning."

Winter 2004

membership in the American Institute of Chemical Engineers
and The American Chemical Society, he is active in the In-
ternational Association for the Advancement of High Pres-
sure Science and Technology, the Association of Environ-
mental Engineering Professors, and The International Soci-
ety for Advancement of
Supercritical Fluids. He
has served on many Na-
tional Science Foundation
and National Research
Council committees
aimed at defining future
directions in the thermo-
dynamics area-espe-
cially for high pressure
applications. Current and
past service on the Edito-
rial Boards of the AIChE
Journal, Industrial and
Engineering Chemistry
Research, Journal of
Supercritical Fluids, and
Fluid Phase Equilibria, Son Ted and daughter Lyj
guarantees a steady flow Murphree
of manuscripts to his desk
to review, which I sometimes find him pouring over when I
visit his office.
In addition to presenting over 300 invited lectures, he has
served in an almost-endless list of service capacities to our
community. They range from the technical (Chairmanship
of the International Symposium on Supercritical Fluids)
to the time-consuming (AICHE, ABET Accreditation
Committee), but all are aimed at enabling the functioning
of our community.

A MIDWESTERNER
EDUCATED ON BOTH COASTS
Chuck grew up in St. Louis and attended MIT for his
bachelor's and master's degrees, which he received in 1960
and 1961, respectively. He then crossed the country and earned
his PhD from the University of California at Berkeley in
1964. He also did a postdoctoral stint in France, which
began a lifelong affinity for that country that still results
in frequent visits.

A DYNAMIC CAREER AT ILLINOIS
Chuck joined the faculty at the University of Illinois,
Champaign-Urbana, in 1965 as an Assistant Professor. Ris-
ing through the ranks with promotions to Associate Profes-
sor (1969) and full Professor (1973), he was recognized for
both his research and teaching contributions.
Chuck was also one of the pioneers in using computers for
interactive education. He developed a number of educational

nn (
Awa

programs on the "Plato" system focused on this interactive
concept-well ahead of most of the chemical engineering
community. In 1973, he received the Alan P. Colburn award
and in 1977, the ACS Ipatieff Prize. In 1983, Chuck was
elected to the National Academy of Engineering for his "Out-
standing contributions
leading to the selection of
liquid metals and
supercriticalfluids as sol-
vents in chemical reactors,
and to improved under-
standing of the extreme
conditions in such reac-
tors. "He has also received
awards for distinguished
teaching and leadership
reflecting his contribu-
tions to diverse curriculum
and strategic planning.
Chuck served at the
Head of the Department
at Illinois from 1980-86.
asey) with Chuck at 1995 Moreover, service to the
rd Dinner. community on ABET
and numerous Steering
Committees made the years in the middle and late 1980s
extremely busy.
Chuck recognized that poor communication skills were at
least as serious a handicap for a typical B S ChE as not being
able to solve complex equations. Again ahead of much of the
community, he developed a highly successful "Chemical En-
gineering Communications" course dealing with oral as well
as written technical communications skills. He "crossed the
street" once more, this time to the English Department where
he was able to assemble a team to deal with the full range of
communications needs. Such courses are now fairly common,
but twenty years ago, this initiative was viewed as "unusual"
at best. His selection for an Alumni Professorship in 1985
reflected recognition for his innovations to deal with the full
range of student needs.

A HUGE IMPACT AT GEORGIA TECH
Chuck moved to Georgia in 1989 and began a new super-
charged career. He holds the J. Erskine Love, Jr., Chair in the
School of Chemical and Biomolecular Engineering. He also
holds the title of "Institute Professor," which is reserved for
individuals who have had significant impact beyond their
individual School bounds. Chuck serves as the Director
of the Specialty Separations Center, which has a cross-
disciplinary vision and goals to connect activities across
the Tech campus.
Clearly, in the move to Georgia Chuck brought with him
his ideas regarding the importance of excellence in research

Chemical Engineering Education

and teaching, and he has found a receptive envi-
ronment at Tech. He was attracted by the I left
Institute's collegiality, its opportunities for mo
multidisciplinary work and partnerships with in-
dustry, and the opportunity to help promote the
rapidly emerging program at Tech. He notes that,
"The reality has far exceeded my expectations"
with regard to the above opportunities.
From my own observations, and the comments of colleagues
here at Tech, it is fair to say that the same sentiment is shared
with regard to the payoff on expectations.
Arnie Stancell, a faculty colleague at Tech says, "I have
had the pleasure of working with Chuck for ten years, and
his enthusiasm for educating students is infectious. He is al-
ways working on ways to engage students in learning. He
personally took on presenting a seminar course for freshman
to introduce them to chemical engineering. He developed in-
teresting problem sets illustrating applications of chemical
engineering. He brought in speakers to discuss current soci-
etal problems that the chemical engineer can help solve. Chuck
did not have to do this-he has won many honors and is
highly respected. He did it because of a genuine passion
for educating students and seeing them grow in their
knowledge and understanding."
Indeed, Chuck's enthusiasm is infectious. His latest initia-
tive is to promote research opportunities for undergraduates.
Besides his full complement of graduate students, Chuck has
opened his lab and made time to meet with undergraduates.
Although always a part of his vision, the significantly ex-
panded activity to involve undergraduates has caught the at-
tention of faculty and administrators alike. The President's
office at Tech has encouraged a broader participation by un-
dergraduates and cited Chuck's "ahead-of-the-curve" lead-
ership as exemplary. In his own words, Chuck notes, "Re-
search is perhaps the best instructional tool that professors
have at their disposal - the one-on-one creative interaction
of real, unsolved problems is the best method of teaching
and learning." The motivations for such a program are many,
and include

* Teaching fundamentals in ways that are more meaningful
than contrived textbook problems, or sanitized cookbook
laboratory experiments.
* Providing motivation, as the students are able to see the
impact i. 11 efforts on the real world. Students gain
enthusiasm and self-confidence.
* Putting the students in close contact with PhD students,
postdoctorals, and other high-level processionals; it
demonstrates teamwork and motivates students to seek
leadership roles in:/ .... " ' .. *
* Providing a framework that permits students to gain more
from their coursework.
* Providing a focus for students 'understanding of the

with the feeling that this duo could cook up
re than enough ideas to keep a full industrial
research center actively engaged if they were
aimed at any particular problem.

profession, and helps them formulate meaningful plans for
their futures - practice of the profession or graduate study.
* Fostering creativity, where traditional courses tend to
discourage it.
In 2000, Chuck received a State of Georgia Regent 's Award
for his leadership in this regard.

THE ECKERT-LIOTTA TEAM
In addition to the institutional issues that helped attract
Chuck to Georgia Tech, an important personal connection
also encouraged the move. Charlie Liotta, an internationally
well-known organic chemist in the Tech School of Chemis-
try, jokes that they built the School of Chemistry around him,
since he has been there for 39 years. Chuck and Charlie be-
came personally acquainted during numerous interactions as
consultants for DuPont. Their hosts at DuPont would often
team them together during consulting visits, and Chuck and
Charlie eventually realized that there must be a message there.
Indeed, their mutual technical interests and strengths were
extremely complementary, and possibilities for collaboration
were often discussed but never acted upon-until the oppor-
tunity for Chuck to move to Georgia Tech materialized. Ron
Rousseau, Chairman of Chemical and Biomolecular Engi-
neering at Tech, enlisted Charlie's active participation in re-
cruiting Chuck in 1989, and the "dynamic duo" has been in-
separable ever since. Together, they have published over fifty
papers in the past fourteen years. Moreover, all of the most
recent and current PhD and MS students that Chuck and
Charlie supervise in both chemical engineering and chemis-
try are done jointly.
I have been lucky enough to participate in one of their
weekly high-energy group meetings, and the intellectual in-
tensity there was impressive. I left with the feeling that this
duo could cook up more than enough ideas to keep a full
industrial research center actively engaged if they were aimed
at any particular problem.
Chuck indicates that much of the focus of their current re-
search is on sustainable development and environmentally
benign processing. This includes a variety of phase transfer
catalysis-related projects, under supercritical and near-criti-
cal conditions. These topics integrate three long-time fa-
vorite subjects of Chuck's: phase equilibrium, high-pres-
sure reactions, and supercritical partitioning. Based on
the past experience, this will be a good area to expect
future developments! O1

Winter 2004

EM department

The

University ofAlabama

C.S. BRAZEL, D.W. ARNOLD, G.C. APRIL, A.M. LANE, J.M. WEST
The University ofAlabama Tuscaloosa, Alabama 35487-0203
Sunny fall weekend in Alabama conjures up
images of the storied traditions of The Uni-
versity ofAlabama (UA): the aroma of South-
ern barbecue fills the air; alumni and students, as well
as many others, descend on campus for a three-day
tailgating party; many pay homage to the past by vis-
iting the Paul "Bear" Bryant Museum, and crowds
gather at Bryant-Denny stadium to cheer on the famed
Crimson Tide. When the weekend passes, the visitors
return to their normal lives in Tuscaloosa (home city
to UA) and elsewhere, and the excitement of the big
game is replaced by activities of the 20,000 students.
Set at the southern end of the Appalachians and bor-
dered by the Black Warrior River, UA's campus was
established in 1831 and has seen many historic mo-
ments. Several buildings on campus survived the U.S.
Civil War, and Governor Wallace's stand in the school-
house doorbrings to mind a more ignominious past.
Today, The University of Alabama provides a
breadth of educational options for a diverse stu-
dent body- from liberal arts and business to law,
science, and engineering.

LIVING IN WEST CENTRAL ALABAMA
Tuscaloosa's metropolitan area of 125,000 bustles
with more than just University activities. About an
hour's drive west of Birmingham, Tuscaloosa is
nestled in a forested area dotted with numerous rec-
reational lakes. The spring and fall seasons are espe-
cially long and pleasant, inviting the outdoor enthusi-
ast to participate in any number of pastimes.
Tuscaloosa's sister city of Northport is an active arts
center that hosts the annual Kentuck festival each fall
and numerous music and performing arts activities
Denny Chimes, one of the most recognizable features of the UA year-round. Local industries that employ our gradu-
campus, framed by a dogwood tree in full bloom. ates include JVC America Inc., Hunt Oil Co.,
@ Copyright ChE Division of ASEE 2004

Chemical Engineering Education

RadiciSpandex Co., Southern Heat
Exchanger Corp., and Mercedes- UA's Cher
Benz US International Inc.
active ro
The University of Alabama is cen-
tral to the city of Tuscaloosa in both strip
geography and spirit. It has an aes-
thetic appeal, with large grassy malls,
tree-lined sidewalks, and campus
buildings with stately Southern grace.
Sitting on the opposite side of campus from Bryant-Denny
stadium, the Chemical Engineering Department is
housed in the Tom Bevill Building, one of the more
recent additions to campus. It houses modern research
laboratories, faculty offices, conference rooms, and
interactive classrooms.

HISTORY AND GROWTH OF ChE AT UA
The College of Engineering at UA is the third oldest
continuously operating engineering program in the coun-
try. Created in 1837, just six years after the formation of
the University, the College remains an active and vital
part of the University's higher education mission and
solidifies the institution as the capstone for higher edu-
cation in the State of Alabama. With nearly 15,000 un-
dergraduate and 5,000 graduate students, UA is one of
seven major PhD-granting institutions in Alabama. The
campus is made up of eight colleges, with the College of
Engineering representing about ten percent of the stu-
dent population, but thirty percent of the honors students.
Established in 1910, the Chemical Engineering Depart-
ment, like many others in the nation, originated out of a
need for a degree that emphasized industrial aspects of
chemistry. Its establishment was just one year after the
inception of the American Institute of Chemical Engi-
neers. The first UA chemical engineering degree was
awarded in 1914.
During the early years, a professional degree was avail-
able to students in addition to the traditional BS and MS
degrees. Then, in the early 1960s, the College of Engi-
neering developed its PhD degree programs in response
to the arrival of NASA and other research-intensive
organizations in northern Alabama. The department
awarded the first two PhD degrees in the College of
Engineering in 1964.
Throughout the years, the changing face of the chemi-
cal industry has been reflected within UA's chemical en-
gineering degree program. From highly practical BS and
MS degree programs through the '60s and '70s, the de-
partment has evolved to keep pace with changes in in-
dustry and made sure that its ChE degree has retained
relevance as student career choices have become more
diverse. The mission of the Department has always been

nical Engineering Department maintains an
le in the national curriculum reform efforts,
ving to balance the important core concepts
at the heart of chemical engineering with
changing and emerging technologies.

UA chemical engineering graduates of 2003 stand along the
stately stairs of the President's Mansion, one of a handful
of buildings at UA to have survived the Civil War.

and remains to educate young professionals as translators of fun-
damental knowledge into viable solutions to problems that are
technically, environmentally, sociologically, economically, and
globally significant.
Today, UA's chemical engineering department comprises 230
undergraduate and 30 graduate students, along with a full-time
staff of 18, including 12 professors. The program offers BS
(since 1910), MS (since 1910), and PhD (since 1964) degrees

Winter 2004

and annually graduates more than 40 undergraduates
and eight graduate students.
UA students find employment in all areas of indus-
try, from fine chemicals and consumer products to poly-
mers and petrochemicals, or they pursue advanced study
in graduate school, medical/dental school, or law
school. Many undergraduates opt for minors or depart-
mental certificates in areas such as business or environ-
mental engineering. With more than thirty percent of its
students graduating with honors, chemical engineering
is a leader in the College and University for its diversity
(more than forty percent women and fifteen percent mi-
norities), its leadership, and its quest for excellence.
As one astute alumnus observed during a campus
visit, although the Department's image has been trans-
formed throughout the years, "the fundamental parts
that made a chemical engineer in the 1960s remain as
important for the chemical engineer in the new millen-
nium." While this assessment shows the continued
strength of a core chemical engineering degree, the
Chemical Engineering Department continues to evolve
to accommodate the new technologies that are just be-
coming visible on the horizon.

ChE FACULTY
There are currently 12 full-time, tenured, or tenure-
track faculty in the Department. They include four full
professors, three associate professors, and five assis-
tant professors. Griffin serves as the Southeastern
NIGEC Director and the State of Alabama EPSCoR
Director. All faculty members are fully engaged in the
instructional and research programs at the undergradu-
ate and graduate levels. Collectively, the department
has averaged more than $2 million of externally funded
awards over the last five years, resulting in a top-35
ranking for expenditures for chemical engineering re-
search as compiled by NSF for the last three years
(1999-2001). In addition, ASEE has consistently ranked
the department among the top 50 chemical engineer-
ing BS-degree-granting institutions.

UNDERGRADUATE PROGRAMS
From a student's perspective, the Chemical Engineer-
ing Department offers several unique opportunities. Un-
dergraduates get to know all of their professors during
their four years on campus. As freshmen, the students
take a one-hour introduction to chemical engineering
course that focuses on informing students about career
options, preparing them for problem solving, and build-
ing the camaraderie that grows between students dur-
ing their time on campus. The AIChE student chap-
ter actively involves the students in its meetings and
outreach activities.

Chemical Engineering Faculty
at The University ofAlabama

Gary C. April, Department Head
University Research Professor
Ph.D., Louisiana State University, 1969
large system modeling * biomass conversion
David W. Arnold
Professsor, Undergraduate Coordinator
Ph.D., Purdue University 1980
coal-water fuels * soil remediation
Christopher S. Brazel
Assistant Professor
Ph.D., Purdue University 1997
molecular design of polymer systems
* drug delivery Eric Carlson

Associate Profesor
Ph.D., University of Wyoming, 1986

f l numerical modehng ofpe
Peter E. Clark
Associate Professor
Ph.D., Oklahoma State University, 1972
rheology ofnon-Newtonian fluids

rmeable
media

RODert A. 'Grifin i
Cudworth Professor; Director, Environmental Inst.
Ph.D., Utah State University, 1973
environmental * soil remediation
Duane T. Johnson
Assistant Professor
Ph.D., University of Florida, 1997
interfacial phenomena * magnetic dispersion technology
* nonhlinear dynamics

UE

Tonya M. Klein
Assistant Professor
Ph.D., North Carolina State University, 1999
chemical vapor deposition for electronics
Alan M. Lane
Professor
Ph.D., University of Massachusetts, 1984
catalysis * colloids
Stephen M.C. Ritchie
Assistant Professor
Ph.D., University of Kentucky, 2001
advanced membrane structures for
environmental separations
C. Heath Turner
Assistant Professor
Ph.D., North Carolina State University, 2002
chemical reaction simulations
Mark L. Weaver
Adjunct Associate Professor
Ph.D., University of Florida, 1995
microstructural characterization and
tribology of bulk and thin films
John M. Wiest
Associate Professor
Ph.D., University of Wisconsin, 1986
molecular rheology * transport phenomena

Chemical Engineering Education

l

I

~--'C--

ur. Alein (ngntj runs a cnemicai vapor aeposimon
experiment with researchers in her laboratory.

The students form the heart of the department, and their
enthusiasm for UA chemical engineering shows at times such
as E-Day, where the students take the lead role in preparing
tours, demonstrations, and discussions for prospective en-
gineering students from high schools across Alabama. The
AIChE group also has a tradition of hosting a friendly pic-
nic with the AIChE student chapter from one of our rival
schools, Mississippi State.
As students progress through the curriculum, they can take
advantage of numerous educational opportunities. Nearly
thirty percent of the students are involved in cooperative
education. Involvement in undergraduate research has in-
creased significantly in the past five years, with more than
one-third of the students working in a chemical engineering
research lab.
The chemical engineering curriculum is centered around
the traditional chemical engineering courses in material and
energy balances, thermodynamics, and reaction and trans-
port phenomena. The students also take advanced elective
courses, two of which are technical-an advanced chemis-
try and an advanced chemical engineering course. The avail-
ability of engineering electives in chemical engineering has
increased substantially with the influx of new assistant pro-
fessors in the past five years. Four new junior/senior/gradu-
ate student electives have been taught for the first time at

ur. Lane (aiso Known as me oiues guitarist noole
'Doghouse' Wilson) gets his class involved in
the Reynolds' Rap.

UA since 2000. Two additional electives can be selected from
nearly anything offered on campus; students simply have to jus-
tify their selection by describing how the course will aid their
careers. With the wide availability of courses at UA, many choose
to fill these electives with business classes, biology courses,
foreign languages, environmental engineering classes, or un-
dergraduate research.
Summer Lab
One of the unique educational experiences at UA comes in
the early summer after completion of the junior year. "Summer
lab" is a five-credit-hour course that is perhaps the most intense
unit operations laboratory in the country. Lab is in session from
8 a.m. to 5 p.m., Monday through Saturday, for five weeks. It is
taught in May to early June each year to avoid scheduling con-
flicts and distractions for the students. If you were to ask an
undergraduate about summer lab, you would likely get one of
two answers: "It's scary, the time commitment is overwhelm-
ing," or "It was the most significant event during my time at
UA." The first statement represents what summer lab looks like
to the freshmen, sophomores, and juniors, while the attitude shifts
as seniors realize that the intense working environment not only
pulls together the theory they have learned in other chemical
engineering courses, but also prepares them for their careers.
By working in teams of three-to-five students, the students
gain valuable experience with team dynamics while they work
on five different experiments led by three to four professors.
The experiments change from year to year. Teams receive short
assignments composed of one-paragraph statements at the first
lab meeting on the first Saturday. After an extensive safety re-
view, they are released to write proposals, determine equipment
to be used, and perform preliminary work. The students must
prepare a proposal that is approved by the faculty for each ex-
periment, followed by two days to build and run the experi-

Winter 2004

. . the department has evolved to keep pace with changes in industry
and made sure that its ChE degree has retained relevance
as student career choices have become more diverse.

ment, to compile and sub-
mit a technical report, and
to present their results.
During the work, each i
group meets with the in-
structor to discuss experi-
mental strategies and give
progress reports. These
meetings are designed to
simulate an industrial set-
ting; they are informal and
may last as long as two
hours. Team members an-
swer questions on all as-
pects of the experiment at
the proposal meeting. The
challenge to create an ac- The Tom Bevill building, hon
ceptable proposal rests on
the team and often re-
quires several drafts. Great emphasis is placed on the pro-
posal so the students understand what they are doing in lab
and can get meaningful results. The instructors are heavily
involved in supervision of the experiments.
Undergraduate Honors Program
A relative newcomer to the undergraduate curriculum is an
honors program specifically for chemical engineering stu-
dents. The requirements to join it match that of the Univer-
sity Honors College, and the courses carry through the junior
and senior year. This curriculum requires a total of twelve
hours of honors classes, with at least six hours in chemical
engineering. Honors forum classes are taught at two levels:
sophomore level (beginning of ChE curriculum) and junior/
senior level. The forum subject rotates from semester to se-
mester, with different instructors delving into recent devel-
opments in chemical engineering, such as "Engineering the
Hydrogen Economy" and "Bionanotechnology." The honors
co-op and internship program allows advanced students to
work with industrial mentors and to earn honors credit upon
presenting project findings to faculty. Industrial recruiters have
shown marked enthusiasm about the honors co-op program,
and we will learn more as UA's chemical engineering honors
program matures.

GRADUATE EDUCATION AND RESEARCH
The department has offered graduate degrees in chemical
engineering since 1914. The emphasis has shifted over the

ie to

last decade from masters
to doctoral degrees. This
has been accompanied by
an increase in externally
funded research from just
under $1 million to more
than $3 million in 2003.
The laboratories and
graduate student offices
were custom designed by
-- - tthe faculty when the
building was constructed
in 1994.
A hallmark of our re-
search program is collabo-
ration with chemists,
chemical engineering at UA. physicists, biologists,
mathematicians, and other
engineers in a variety of
campus-wide research centers. The Center for Materials for
Information Technology (MINT) was established in 1990 in
response to JVC's 1986 decision to locate a magnetic tape
manufacturing facility in Tuscaloosa, as well as a large con-
centrationof the data storage industry in the Southeast. Chemi-
cal engineering faculty (Arnold, Johnson, Klein, Lane,
Weaver, Wiest) joined other faculty in science and engineer-
ing to earn an NSF Materials Research Science and Engi-
neering Center grant in 1994 (the first ever in the South-
east) with renewals in 1998 and 2002. The emphasis is on
developing new materials for high-density data storage
and spintronics.
Mercedes-Benz located their only US-based production
facility in Tuscaloosa in 1993, manufacturing the M-class
SUV here. Honda, Hyundai, Nissan, Toyota, and the support-
ing industrial suppliers followed soon after, making the re-
gion a center for automobile manufacturing. UA supports this
industry through the Center for Advanced Vehicle Technol-
ogy, in which the multidisciplinary fuel cell research group
plays a leading role. With a focus on materials, chemical en-
gineering faculty (Lane, Wiest, Turner, Klein, Ritchie,
Weaver) are developing new catalysts for hydrogen produc-
tion and fuel cells.
A microelectromechanical systems (MEMS) laboratory was
established in 2002. Initial work by Klein and collaborators
focuses on the microfabrication of gyroscopes. They recently
won an NSF grant to incorporate MEMS technology into the
undergraduate program.

Chemical Engineering Education

A sophomore demonstrates complex viscosity properties to
high school students on E-Day.

Charlotte Nix runs a demonstration of environ-
mental hazards of oil contamination for Project
ROSE. The audience included high school
students and their parents who were visiting the
UA campus for E-Day.

A long-standing departmental emphasis on environ-
mental research is now complemented by the
university's Center for Green Manufacturing. Major
projects have included waterborne magnetic inks (Lane,
Arnold), biomass conversion (April), soil remediation
(Arnold), and benign solvents and additives for the
polymer industry (Brazel).
The mining and petroleum industries remain a vital
part of the Alabama economy and are served by Carlson
(subsurface modeling) and Clark (complex rheology).
Clark was recently honored as a Society of Petro-
leum Engineers Distinguished Lecturer. He pre-
sented invited lectures throughout the U.S. during
the 2002-2003 academic year.
The department is particularly proud of its NSF
CAREER award recipients. Mark Weaver has been
studying multilayer thermal barrier coatings since 1999,
addressing the influence of thermal exposure on the
interfacial microstructure. Tonya Klein began her work
in the fall of 2003 on plasma-enhanced, atomic layer
deposition, which is an advancement of traditional
chemical vapor deposition.
strong collaborations among chemical engineering faculty,
colleagues across campus, and the industries we serve result
un and exciting atmosphere in which to conduct truly cut-
;dge research.

*REACH PROGRAMS
long the various outreach activities of the Department, Project
E (Recycled Oil Saves Energy) stands out in both statewide
ct and longevity. Project ROSE, under the direction of Gary
, has been running successfully for 27 years. It involves both
lic awareness arm and activities to aid local communities in
ima in collecting used motor oil for reclamation and recycle.
ach to school groups includes environmental models to ex-
the effects of point source and non-point source contamina-
on ecosystem management. Project ROSE is run by two
ical engineering staff members: Ms. Sheri Powell and Ms.
lotte Nix, who conduct demonstrations throughout the
Project ROSE recently celebrated its active presence in
7 Alabama counties.

FUTURE
's Chemical Engineering Department maintains an active role
national curriculum reform efforts, striving to balance the
rtant core concepts at the heart of chemical engineering with
;ing and emerging technologies. We are forging new rela-
hips with the biological sciences department on campus and
nue to expand our research programs through collaborations
n and beyond the Tuscaloosa campus. Ultimately, our com-
ent to education is expressed in the opportunities afforded
udents and the careers of our graduates.
LL TIDE ! E

Winter 2004

S91^ laboratory

A FLUIDIZED BED ADSORPTION

LABORATORY EXPERIMENT

PAMELA R. WRIGHT,* XUE Liu, BENJAMIN J. GLASSER
Rutgers University * Piscataway, NJ 08854

here are a variety of pedagogical and motivational ad-
vantages in exposing students to real process equip-
ment in a laboratory course.[1] There is also a need,
however, to use simple laboratory experiments in order to
help students better understand basic principles learned in
their coursework. Therefore, it is often advantageous to start
students off with simple experiments where the connection to
basic principles is obvious and then move on to more challeng-
ing and complex systems that resemble real-world situations.
A fluidized bed adsorption process provides a somewhat
unique opportunity for students to carry out a series of ex-
periments (on one piece of apparatus) that steadily approaches
the real process equipment. The series starts with a study of
bed expansion in a fluidized bed, goes on to residence time
distribution measurements, and ends with a study of a
bioseparation in a fluidized bed. This allows students to build
upon ideas they have already learned in fluid mechanics, mass
transfer, separations, and reaction engineering. The experi-
ment was developed in the Department of Chemical and Bio-
chemical Engineering at Rutgers University and forms part
of the Process Engineering Laboratory course for seniors.

PROCESS OVERVIEW
Advances in biotechnology have resulted in the produc-
tion of a multitude of therapeutic proteins by mammalian,
bacterial, and yeast fermentations. The global market for
therapeutic proteins used in the treatment of cancer and AIDS,
as well as growth factors and monoclonal antibodies for di-
agnostic applications is rising. Current work on genomics and
proteomics is likely to make it easier to discover new thera-
peutic proteins, which will in turn lead to an increase in the
production of proteins.
At the same time, primary recovery and purification of the
protein from the fermentation broth continues to be a signifi-
cant limiting factor in the overall economics of therapeutic
protein production. Therefore, bioseparations is a critical step
both from a processing and research point of view. In fact, as

*Address: Centocor Inc., 200 Great. -, , Parkway, Malvern, PA 19355

much as 80% of the production costs for many proteins can
be incurred during product isolation and purification.[2] For
example, therapeutic proteins such as interferons and
interleukins are considered high-value proteins with a price
of $1,000,000 per gram or more.[3] Product concentrations in
a typical feed stream are low, between 10-2 and 10-6 mg/L,
and much of the high manufacturing costs can be attributed
to recovery time and product losses across each step of the
purification process.[4] In addition, the final purified product
must often be greater than 99.9% pure, with less than 10 pg
per dose of nucleic acids and endotoxins.151
In the biotechnology and pharmaceutical industries, ion
exchange chromatography (IEC) is the most widely used
operation for purification of proteins. The operation typically
involves a packed bed of resin particles or adsorbent beads
that selectively adsorb the target protein. After the resin par-

Pamela R. Wright received her BS from the
University of Maryland, her MS from Stevens
Institute of Technology, and her PhD from
Rutgers University She is currently a Director
at Centocor Inc., where she works in the area of
biotechnology

Xue Liu received his BS and MS from
Tsinghua University (China). Currently he is
a PhD student at Rutgers University His re-
search is in the field of gas-particle flows in
fluidized beds and risers.

Benjamin J. Glasser is Associate Professorof
Chemical and Biochemical Engineering at
Rutgers University. He earned degrees in
chemical engineering from the University of the
Witwatersrand (BS, MS) and Princeton Univer-
sity (PhD). His research interests include gas-
particle flows, granular flows, multiphase reac-
tors, and nonlinear dynamics of transport pro-
cesses.

� Copyright ChE Division of ASEE 2004

Chemical Engineering Education

tiles become filled with protein, the feed
to the column is stopped and an eluent buffer . .
is passed through the column in an elution with simp
step. This leads to the product being re- p
leased into the eluent buffer, and the end
result is that the product is typically con-
centrated 10X to 40X.
Generally, the fermentation broths contain
suspended solids, e.g., cells or cell debris that would clog a
packed bed. To prevent this, feedstocks are usually clarified
by filtration or centrifugation before the chromatographic
separation in order to remove the cell debris. Fluidized or
expanded bed adsorption has increasingly become an alter-
native method of interest for adsorption of proteins from feed-
stocks containing cells.[6,71 In this process, a bed of adsorbent
beads is expanded or fluidized by the upflow of liquid, lead-
ing to large voids between the adsorbent beads and al-
lowing cells and cellular debris to pass through the bed
without becoming trapped. As a result, fluidized bed ad-
sorption eliminates the need for the expensive operations
of filtration and centrifugation.
Another advantage that fluidized bed adsorption has over
a packed bed is enhanced mass transfer, which can lead to
increased process yields.E81 This means that for a given pres-
sure drop across the bed, the fluidized bed can in principle
achieve a higher rate of protein removal. For these reasons,
this technology is increasingly being applied as a downstream
separation technique in the pharmaceutical and biotechnol-
ogy industries. At the present time, the technique has been
used for the recovery of recombinant proteins from mamma-
lian cell culture and E. coli fermentation broths.19-11]
Karau, et al., E[21 defined expanded bed adsorption as a sub-
set of fluidized bed adsorption that specifically addresses situ-
ations with low superficial velocities close to the minimum
fluidization velocity. For most resins, the expression "ex-
panded bed adsorption" is applicable only to bed expansions
of less than two times the settled bed height. In this article,
adsorption is investigated at bed expansions of two to four-

Figure 1. Schematic of normal operating mode of
fluidized bed adsorption process.
Winter 2004

. it is often advantageous to start students off
le experiments where the connection to basic
inciples is obvious and then move on to more
challenging and complex systems that
resemble real-world situations.

and-one-half times the settled bed height. Thus, the expres-
sion "fluidized bed adsorption" is used to emphasize that we
are investigating protein adsorption for a large range of bed
expansions, including high expansions.
The basic process of fluidized bed adsorption includes the
application of feed through the bottom of a column filled with
resin, as illustrated in Figure 1. Initially, the resin is settled,
but the upward feed flow results in suspension or fluidiza-
tion of the resin bed. Product in the feedstock adsorbs to the
resin while nonproduct solid material (e.g., cell debris) washes
out with the spent feed. Subsequent washing with a buffer
further removes nonproduct solid material that may remain
associated with the resin. Product is then recovered by intro-
ducing an eluent buffer (salt solution) through the top of the
column. To minimize process volumes, elution is usually
conducted in the packed-bed mode where the product is con-
centrated 10X to 40X. After elution, the resin can be cleaned
and regenerated for repeated use.
To determine the bed expansion characteristics, study the
effects of liquid velocity and bed expansion on the flow hy-
drodynamics, and identify the dominant mechanistic features
in a fluidized bed adsorption column, the laboratory course
is divided into three parts: bed expansion characterization,
tracer studies, and adsorption of protein. Each of the three
experiments involved in this project requires approximately
four hours of work and is carried out in a single afternoon.
Experiments are finished in three weeks, and the project write-
up is due in the fourth week. Before the first day of each lab,
students are required to read the introduction section from
the laboratory manual for that week's experiments as well as
related materials in the library.

EXPERIMENTAL
EQUIPMENT AND MATERIALS
The laboratory equipment consists of a Streamline 50 ex-
panded bed adsorption column (Pharmacia Biotech,
Piscataway, NJ), a peristaltic pump, an in-line UV sensor,
and a UV analyzer. A schematic of the experimental setup is
shown in Figure 2 (next page), with the principal compo-
nents listed in the caption. The column is constructed of a
borosilicate glass tube, 5 cm in diameter and 100 cm long.
The normal operating pressure is less than 0.5 bar, but the col-
umn can withstand pressures up to 1 bar. The column should
not be operated above 1 bar pressure or without liquid.
The column is supported by a stainless steel mounting for

STARTER BUFFER

LUITON - -
UFFER

H- I/Ho=3

SH/Ho=2
-- SETTLED BED
WASH HEIGHT, Ho
BUFFER
S---- ELUATE

protection and contains an adsorbing resin.
The minimum resin loading is 200 mL or 10
cm settled height; the maximum loading is
600 mL or 30 cm settled height. The resin is
retained by a stainless steel 60-mesh screen
at the base of the column. A peristaltic pump
is used to pump fluid into the base of the col-
umn through a stainless steel distributor plate
with 12 equally spaced 1-mm holes. The dis-
tributor plate is mounted in the base of the
column below the screen and it and the screen
are held in place with rubber gaskets. The col-
umn is equipped with a moveable rod piston
fitted with a 60-mesh screen to retain the resin
at high flow rates or high expansions. Dur-
ing operation, the piston is moved just above
the expanded bed height to minimize head
space. Spent charge is pumped out through
the piston and fed to an in-line UV sensor
(Wedgewood Technology, San Carlos, CA).
The signal from the sensor is analyzed by
an UV analyzer at 280 nm. The resin used in
the experiment is Streamline SP (Pharmacia
Biotech, Piscataway, NJ), which is a cation
exchange resin with a particle radius range
from 45 to 178 jim. A Malvern Mastersizer
X was used to determine that the average
particle radius is 89 pjm, with a particle-size
distribution that is approximately Gaussian
with a skewness of 0.878. Streamline SP has

Fluid Flow
Figure 2. Fluidized bed
adsorption column.
1. Top flange
2. Adapter rod piston
3. Adapter distributor and net
4. Stainless steel mount
5. Glass column
6. Bottom flange
7. Column distributor and net
8. Stand

been used previously in several fluidized bed adsorption ap-
plications, and its hydrodynamic and expansion properties
are well characterized.110,13,14] The average particle density is
1.18 g/mL. Each particle is composed of a crystalline quartz
core, covered by 6% cross-linked agarose. The dynamic bind-
ing capacity reported by the manufacturer is 70-85 mg/mL
for most proteins. Bound proteins inside the particle remain
attached at one adsorption site until they are eluted.
The protein lysozyme (EC 3.2.1.17, Sigma Chemical Com-
pany, St. Louis, MO) was selected as an adsorbing species
since it is relatively inexpensive, well-characterized, and eas-
ily assayed by spectrophotometric methods. Most importantly,
it adsorbs and desorbs readily from Streamline SP resin.
Lysozyme is a globular protein with hydrolytic enzyme
properties. It is nearly spherical, with dimensions of 4.5 x 3 x
3 nm.[15] The molecular weight is 14,600 and the isoelectric
pH is 10.7 to 11.3.[161 This high isoelectric pH allows adsorp-
tion by cation exchange resins at a wide range of pH values.
A point worth mentioning is that the use of protein is not, in
principle, necessary for this experiment. One could do a much
less expensive experiment by changing the protein adsorp-
tion into an ion exchange experiment-for example, exchang-
ing Na+ from a NaCl solution. We believe, however, that stu-

binding capacity, the adsorbent should always remain
wet -by no means should it ever be isolated via
filtration. Quickly pour the slurry into the column.
Resuspend any adsorbent remaining in the container
with deionized water and pour this into the column. If
aggregates of air-adsorbent remain floating on the liquid
surface, they need to be removed or pushed down into
the liquid. Allow the resin to settle and add more resin if
necessary to obtain desired settled bed height. Fill the
column to the rim of the glass tube with deionized
water.
* When the column is secure in the steel mounting
assembly, carefully tilt the adapter and insert it into the
column so that one side of the gasket on the adapter net
is in the water-filled column. Without trapping air under
the net, carefully put the adapter into a vertical position.
Slowly push the adapter down until the gasket can be
seen under the upper flange. When the adapter is firmly
seated in the column, push down the lid and replace the
washers and domed nuts. Fill the space above the
adapter with deionized water.
* To lower or raise the adapter, pump deionized water
into the column side connector (above the adapter) or
into the base of the column at a pump setting of 2 (150

Chemical Engineering Education

dents benefit from being exposed to a
bioseparation and working with a real pro-
tein and a commercial resin.

EXPERIMENTAL PROCEDURE
Column Setup * Before experiments, stu-
dents are required to familiarize themselves
with the standard operating procedure for
operating the Streamline 50 expanded bed
adsorption column. The procedure is
* The first step is to remove the adapter
from the column. The purpose of the
adapter is to minimize the head space
above the resin particles during fluidiza-
tion. To push out the adapter from the
column, use the hydraulic pump to pump
water into the base of the column at a
pump setting of 2 (150 mL/min). The
adapter rises. Stop pumping when the
adapter sits in the upper flange at the top
of the column. Then remove the domed
nuts and washers on the lid, raise the lid,
and remove the piston and adapter plate.
Once the piston and adapter have been
removed, reverse the pump to decrease the
level of water in the column to approxi-
mately 30 cm.
* Prepare an adsorbent-water slurry with
deionized water. To maintain the dynamic

mL/min). Stop the pump when the adapter is at the
desired height in the column. Once the resin is in the
column and the adapter height has been set, the column
is ready for operation.
Bed Expansion Characterization * The first step prior to
starting adsorption is to characterize the bed expansion as a
function of linear velocity and viscosity in a nonadsorbing
system with 200 mL of resin in the column. Viscous and non-
viscous fluids are pumped into the base of the column at four
different linear velocities. The expanded bed height is mea-
sured at each velocity to obtain expansion plots and
Richardson-Zaki plots.[17 This information is used to com-
pare fluidization conditions with published results and also
to identify desirable conditions for adsorption studies.
In this experiment, students are divided into three groups
and each group carries out experiments with a fluid of differ-
ent viscosity. The groups share their data at the end of the
experiment in order to increase the amount of data each group
has to analyze. Group A performs experiments using a 0.05
mol/L sodium acetate buffer solution with 0% glycerol and a
sodium acetate buffer solution with 5% glycerol. Group B
performs experiments with a 0.05 mol/L sodium acetate buffer
solution with 0% glycerol and a sodium acetate buffer solu-
tion with 15% glycerol. Group C uses a 0.05 mol/L sodium
acetate buffer solution with 0% glycerol and a sodium ac-
etate buffer solution with 30% glycerol. The fluid viscosity
is measured by a viscometer. The experimental procedure is
* Record the pump setting and allow ten minutes for
the bed to stabilize. The flow rate is determined by the
volume collected per unit time (mL/min). Once the flow
rate is known, the linear superficial velocity is just the
flow rate divided by the cross-sectional area.
* After ten minutes has passed, read off the stabilized
expanded bed height. The McCabe equation below
determines the fluidized bed porosity:[is]

H _ (1- ) (1
Ho (1- )
where o is the voidage of the particles in settled bed
mode, Ho is the settled bed height, H is the expanded
bed height, and e is the expanded bed porosity. A value
of Fo = 0.4 was measured for the particles in settled bed
mode.
* After the experiment, students can plot the logarithm
of the linear superficial velocity versus the logarithm of
the expanded bed porosity. The slope of this line is the
Richardson-Zaki coefficient.[171
Tracer Studies * To characterize the internal flow hydro-
dynamics and axial mixing of Streamline SP resin, tracer stud-
ies are performed using a 0.25% acetone pulse to determine
the dispersion and residence time distribution (RTD) charac-
teristics of the system as a function of bed expansion. The

[This] experiment also provides an
opportunity for students to carry out a
series of experiments that increases in
complexity and approaches the
real process equipment.

acetone is added into the sodium acetate buffer as well as the
various percentage glycerol buffer solutions. The acetone at
the column outlet is monitored by the UV analyzer at 280 nm
for a given degree of bed expansion, which is determined by
the liquid velocity corresponding to each fluidized bed height.
Students can obtain this information from the Bed Expansion
Characterization. A positive step signal is used to obtain resi-
dence time distributions by the F-curve method.[191 Measure-
ments associated with the positive step signal lead to an F-
curve. The data in the F-curve is then differentiated to obtain
the C-curve. Values for the variance (o ) of the C curve are
used to calculate the mean residence time in the expanded
bed, axial dispersion coefficient (Dx), and the number of theo-
retical plates (N). In the interest of saving time, only one run
per a given flow rate is carried out. The experimental proce-
dure is
* After recording such information as pH, temperature,
flow rate, and the characteristics of the solution,
students should move the adapter approximately 1 cm
above the desired expansion height. A large gap (or
large head space) above the resin may lead to a region
of pure fluid above the resin, and this will affect the
residence time distribution measurements. Start the
recorder/UV-monitor and allow it to warm up for 20
minutes or more. Prior to expansion, two 20-L carboys
need to be set up. One should be filled with sodium
acetate buffer solution and the other should be filled
with tracer (0.25% acetone in sodium acetate buffer
solution). Air bubbles should be evacuated from the
lines before expansion. Once the adapter is in position,
bed expansion can be started by introducing the buffer
solution. When the bed is fully expanded at the test
flow rate, note the expanded bed height from the
calibrated column and continue pumping buffer. At this
time, zero the UV sensor. After this is done, unclamp
and bleed the tracer line and clamp the buffer line.
* At the instant tracer is introduced, begin to record the
time and UV readings from the sensor. UV recordings
should be taken every 30 seconds in the beginning,
until an increase in activity is noticed, at which point
readings should be taken every 15 seconds. Continue to
take readings approximately 5-10 minutes after the UV
readings have leveled off.
* Then clamp the tracer line and re-open the buffer
line. Record this time and continue to record UV
readings in 15-30-second intervals until the readings go

Winter 2004

down to approximately zero.
* Every group should do three expansions that include
2X, 3.3X, and 4.5X the settled height.
Adsorption of Protein * After examining particle fluidiza-
tion and axial dispersion characteristics of the resin, dynamic
adsorption capacities are measured for the resin to assess
mass-transfer effects under different hydrodynamic condi-
tions. To identify the dominant mechanistic features of the
fluidized bed adsorption system, the fluidization studies
should be designed to isolate mass-transfer effects from hy-
drodynamic effects. This can be accomplished by frontal
analysis of breakthrough curves to determine dynamic ad-
sorption capacity of the resin under varying conditions of
linear velocity, viscosity, and axial dispersion. The experi-
mental procedure is
* Prior to experimentation, several initial steps should
be performed. The resin should be washed with 10 L of
a 1-mol/L NaCl solution at a pump setting of 1.5 (100
mL/min). This expands the bed and allows for proper
cleaning of the resin. Following the salt solution wash,
20 L of deionized water should be introduced into the
column with a pump setting of 1.5 (100 mL/min). This
removes the salt as well as other impurities that are
introduced while the resin is sitting immobile in the
column. The conductivity of the outlet should be
checked to ensure all the salt has been removed by
obtaining a conductivity reading of less than 5 mS. If
the conductivity is too high, continue washing the resin
with another 10 L of deionized water. Equilibrate the
resin with 20 L of a 0.05-mol/L sodium acetate buffer
solution at a pH of 5. If the resin is not equilibrated to
the buffer, inaccurate data will be obtained for the
adsorption. Prior to experimentation, additional buffer
solution (20 L) as well as protein solution (10 L) should
be prepared, and the UV sensor should be allowed to
warm up for 20 minutes to obtain accurate readings for
concentration. Then zero the UV sensor using 0.05-
mol/L sodium acetate buffer.
* Before starting the experiments, a sample of the
protein solution should be introduced into the UV
sensor to obtain an initial concentration reading. This is
the Co value. The desired breakthrough concentration
(usually 10 to 30% of initial concentration) is the
breakthrough percentage multiplied by the initial
concentration.
* For operation of the column, the following procedure
should be followed. From the Bed Expansion Charac-
terization, students have a direct correlation between
pump setting, linear velocity, and expanded bed height.
Due to the expense of the protein, only one adsorption
is carried out for each group. Group A uses a 2X
expansion, Group B uses a 3.3X expansion, and Group
C uses a 4.5X expansion. The lines to the column

should be bled prior to introducing any fluid into the
column, and the lines from each solution must be void
of air bubbles. The buffer solution should be introduced
first in order to obtain a stable bed height.
* Once this is achieved, the protein solution can be
introduced. Record UV readings at 1-minute intervals
until increased activity in the UV output is noticed.
Then take UV readings at 30-second intervals until C/
Co of 0.15 has been reached. This point is defined as
column breakthrough, which is the point of reduced
binding capacity. In most commercial applications, the
adsorption is discontinued at a point where the exit
concentration is 10% to 15% of the inlet feed concen-
tration, to prevent unacceptable product losses. In this
study, 15% has been used. Once breakthrough is
achieved, the time should be recorded as well as the
buffer volume.
* After the above procedure has been finished,
unclamp the buffer solution line and clamp the protein
solution line. At this point, 10 L of a 1-mol/L NaCl
solution at pH 5 should be introduced into the column
at a pump setting of 1.5 (100 mL/min) to recover the
protein. After that, 20 L of deionized water should be
introduced into the column at a pump setting of 1.5

(a)
410

S310

. 210

110

2 3 4 5
H/Ho

(b)
1000

100 -

S10
-0.16 -0.12 -0.08 -0.04
log(voidage)
Figure 3. The characteristics of the bed expansion.
(a) Plot of H/Ho versus linear velocity in the buffer solution
without glycerol (*), and with 30% glycol (A).
(b) Richardson-Zaki parameter plots in the buffer solution
without glycerol (4) and with 30% glycerol (A).

Chemical Engineering Education

(100 mL/min) to rinse the column and resin.

RESULTS AND DISCUSSION
Bed Expansion Characterization * The effects of fluid
velocity and viscosity on the bed expansion can be seen in
Figure 3. As would be expected, an increase in viscosity leads
to a larger expansion for a given superficial velocity (see Fig-
ure 3a). Richardson and Zakil1l1 observed that if the log of the
voidage was plotted versus the log of superficial fluidization
velocity, a linear relationship is obtained. A correlation was
developed and is generally called the "Richardson-Zaki equa-
tion," written as
Us (n+l) (2)
Ut
where n is the Richardson-Zaki number, u, is the superficial
velocity, and ut is the particle terminal velocity, which is a
function of particle density, fluid density, particle diameter,
and fluid viscosity. In the fluidized bed system, ut canbe seen
as a constant. In order to compute the Richardson-Zaki num-
ber, n, one can plot the logarithm of linear velocity versus the

(a)

1 �

0.8 -

o 0.6 -

0.4 -

0.2
0 - - * � Z I
0 3 6 9 12 15 18
Time (min)

(b)
0.25

0.2-

S0.15-

0 0.1 -

0.05

0 5 10 15 20
Time (min)
Figure 4. Acetone tracer curves for Streamline SP at an
expansion of H/Ho = 2 in 50 mol/L NAOAC buffer
solution. (a) F-curve; (b) C-curve

logarithm of fluidized bed porosity. One should get a straight
line with slope n+1. The Richardson-Zaki number is a func-
tion of the ratio of particle diameter to column diameter. Since
the resin and the column are not changed during experiments,
the Richardson-Zaki number should be the same for the dif-
ferent buffer solutions, as can be seen in Figure 3 (b). Although
the fluid viscosity does not change the Richardson-Zaki num-
ber, it does affect the bed expansion, as shown in Figure 3 (a).
Tracer Studies * To characterize the internal flow hydro-
dynamics and axial mixing of Streamline SP, tracer studies
are performed at different bed expansions. Good reproduc-
ibility is generally obtained from three trials at each condi-
tion and the standard deviation is generally less than 5% for
each parameter. Figure 4 shows typical acetone tracer curves
for Streamline SP at an expansion of H/H = 2 in 0.05 mol/L
NAOAC buffer. Axial dispersion coefficients are obtained
from the variance, o2, in the C-curve as follows:[121

Dax = (usH /2) (3)
where H is the height of the fluidized bed and u is the superfi-
cial linear velocity. (o canbe calculated in the following way:[191

tmean [(0 tCdt)/(f1 CdtJ]

G2 ( -tmean)2 Cdt)/(fo Cdt)]

2 = (52/t2ean

where C is the concentration of the tracer at time t. These
quantities can be evaluated by making use of the following
numerical integration formulas:

tmean =[( CtAtl)/( CAtl)] (7)

02 = [((t- meann2 * CAtl)/( C,At)] (8)

where the data is divided into time intervals of At and C1 is
the concentration of tracer at time t.
Once the value of (c2 and D has been calculated, the Peclet
number and the number of theoretical plates can be deter-
mined from
Pe = (usH /Dax) (9)

N = 1 / o2 (10)
The axial dispersion coefficients for Streamline SP in buffer
without glycerol at the expansion of 2X and 3.3X are com-
puted to be 1.8 x 10-6 mVs and 7.27 x 10-6 mVs, respectively.
When 30% glycerol is added, axial dispersion is relatively
unchanged at H/Ho = 2, but lower linear velocities are re-
quired to obtain this same degree of expansion. For the fluid-

Winter 2004

ized bed system, the Peclet number, which is the ratio of the
convective transport to the dispersive transport in the expan-
sion, can be used to quantify the extent of deviation from
plug flow in the column.[18] In true plug flow, the Peclet num-
ber approaches infinity. For completely mixed flow, the Peclet
number approaches 0. In this study, the Peclet number ranges
from 40 to 80, indicating a small deviation from plug flow.
Adsorption of Protein * For these experiments, the frontal
analysis of breakthrough curves has been used to determine
the effect of axial dispersion on adsorption in an expanded
bed. The breakthrough curves are shown in Figure 5. To fa-
cilitate direct comparison of breakthrough, the adsorbed con-
centration, q, is normalized with respect to the equilibrium
capacity q. and plotted as q/qo versus C/C,. As discussed ear-
lier, breakthrough is defined as C/Co= 0.15 or at 15% of the
feed concentration, Co. Results from RTD and frontal analy-
sis are shown in Table 1 together with the q/q values at break-
through (i. e., the q/q. value corresponding to C/Co = 0.15). Here
the average residence time for each condition is defined as

SeH / s (11)

When the expanded bed height is 2 times the settled bed
height, the bed porosity, c, is approximately 0.7. Under these
conditions, the linear velocity is 168 cm/h, and q/q. is 0.97
at breakthrough. The addition of 30% glycerol resulted in an
increased bulk phase viscosity and a linear velocity of only
64 cm/h is required to expand the bed to twice the settled
height. Under this condition, breakthrough occurs at q/q. =
0.86 even though the residence time is significantly higher
than for the buffer-only case. When Streamline SP is expanded
to 3.3 times the settled height in buffer at 300 cm/h, q/qo de-
creases to 0.68 at breakthrough. The residence time does not
change, but the axial dispersion increases compared to the
case where H/Ho = 2. Therefore, since the residence time is
relatively constant, early breakthrough is likely due to in-
creased axial dispersion.
When 30% glycerol is added, the expanded bed height in-
creases to 4.5 times of the settled height at a reduced linear
velocity of 150 cm/h, and a longer residence time than that
for the H/Ho = 2 expansion in glycerol is obtained. Here,
breakthrough occurs even earlier at a q/qo value of 0.54 due
to a 6-fold increase in axial dispersion. The shape of the break-
through curves for Streamline SP resin under the conditions
presented here is of interest as well. The breakthrough curves
are all relatively sharp except for the condition of H/H0 = 4.5
with 30% glycerol. In this case, a gradual breakthrough curve
is obtained, indicating that a low level of lysozyme is bled
through the column before breakthrough is established. In an
actual application, this would amount to product loss.
These results suggest that a macroporous resin such as
Streamline SP is best used for low viscosity feedstocks ap-
plied at intermediate linear velocities since dynamic capaci-
ties are severely reduced with higher viscosity feedstocks. It

should be mentioned that the particles used for this study were
not elutriated, and so a wide particle size distribution was
used for all cases (as supplied by the resin manufacturer).
The effect of particle size distribution on breakthrough in flu-
idized bed adsorptions was investigated recently by Karau,
et al. [12] In their study, they found that particles with a wide size
distribution would reduce axial dispersion compared to a nar-
row particle size distribution. The work described here could
be extended by sieving the resin into narrow fractions and car-
rying out experiments to confirm the results of Karau, et al.
The results of this work also suggest that to maximize
throughput with minimal product losses, the operation could
be divided into two steps. Initially, one could operate at very
high expansions until the onset of breakthrough due to high
axial dispersion. At this point the particles are not saturated.
Thus, the linear velocity can be reduced to decrease the bed
height to a regime where only intraparticle or film mass trans-
fer effects dominate. Adsorption could continue at this smaller
expansion with a corresponding longer residence time and re-
duced axial dispersion until the point of breakthrough. Further
experiments could be carried out to confirm this hypothesis.

CONCLUSIONS
This paper describes an experiment that exposes students

0.21
0.18 -
0.15
o 0.12-
0 0.09 -

0
0.06 -

0 0.2 0.4 0.6 0.8 1
dynamic capacity (q/qo)
Figure 5. Breakthrough curves for Streamline SP
* H/Ho = 2, 0% glycerol and us = 168 cm/h
* H/Ho = 3.3, 0% glycerol and us = 300 cm/h
A H/Ho = 2, 30% glycerol and u = 64 cm/h
* H/H = 4.5, 30% glycerol and us = 150 cm/h

TABLE 1
Results of Frontal Analysis with Streamline SP

Buffer H/H, us D q/q% T
(% glyc) (cm/h) (m2/s) (min)
0% 2.0 168 1.80 x 10-6 1.00 0.70 5.0
0% 3.3 300 7.27 x 10-6 0.75 0.82 5.4
30% 2.0 64 1.08 x 10-6 0.86 0.70 13.1
30% 4.5 150 6.27 x 10-6 0.57 0.87 15.7

Chemical Engineering Education

to the basic principles of fluidized-bed operation and protein
adsorption. Feedback from students who have worked on the
laboratory experiment has been very positive. They have par-
ticularly enjoyed working with a real protein and a commer-
cial resin (that needs to be handled with care).

In the experiment, students study the relation of the linear
velocity and the buffer viscosity to the expanded bed height
by simple bed operation, the flow hydrodynamics of the bed
expansion system by tracer studies, and the protein adsorp-
tion characteristics by frontal analysis of breakthrough curves.
In this way they are forced to put together concepts they have
learned in separate courses in fluid mechanics, mass transfer,
separations, and reaction engineering. The fluidized bed labo-
ratory experiment also provides an opportunity for students
to carry out a series of experiments that increases in com-
plexity and approaches the real process equipment.

NOMENCLATURE
H fluidized bed height (cm)
e fluidized bed porosity
n Richardson-Zaki number
u superficial velocity (cm/h)
ut particle terminal velocity (cm/h)
N theoretical plate number
Da axial dispersion coefficient (m2/s)
t time (s)
T average residence time
Pe Peclet number
C concentration (mol/L)
q adsorbed concentration (mol/L)

ACKNOWLEDGMENTS

Funds for equipment were provided by the NJCST Particle
Processing Research Center. We are grateful to David Unger
and Deanna Markley for assistance and to Amersham
Pharmacia Biotech for donating the resins used in this work.

REFERENCES
1. Luyben, W.L., "A Feed-Effluent Heat Exchanger/Reactor Dynamic
Control Laboratory Experiment," Chem. Eng. Ed., 34(1), 56 (2000)
2. Datar, R.V, T. Cartwright, and C.G. Rosen, "Process Economics of
Animal Cell and Bacterial Fermentations: A Case Study Analysis of
Tissue Plasminogen Activator," Bio/Technology, 11, 349 (1993)
3. Bentley, W.E., H.J. Cha, and T. Chase, "Application of Green Fluores-
cent Protein as a Fusion Marker in Recombinant Pichia Pastoris Fer-
mentation: Human Interleukin-2 as a Model Product," AIChE Annual
Meeting, Miami Beach, FL (1998)
4. Fuchs, R.L., R.A. Heeren, M.E. Gustafson, G.J. Rogan, D.E. Bartnicki,
R.M. Leimgruber, R.F. Finn, A. Hershman, and S.A. Berberich, "Puri-
fication and Characterization of Microbially Expressed Neomycin
Phosphotransferase II (NPTII) Protein and Its Equivalence to the Plant
Expressed Protein," Bio/ .. . . 11, 1537 (1993)
5. Hammond, P.M., T. Atkinson, R.F. Sherwood, and M.D. Scawen,
"Manufacturing New Generation Proteins: Part 1. The Technology,"
BioPharm, 4, 16 (1991)

/" POSITIONS AVAILABLE "
Use CEE's reasonable rates to advertise.
Minimum rate, 1/8 page, $100;
Each additional column inch or portion thereof, $40.

UCLA
UCLA Chemical Engineering Department is seeking applicants for a
faculty position effective 2004/2005 academic year. Candidates must
have a Ph.D. degree in chemical engineering or a related field, and be
able to teach undergraduate and graduate courses and direct M.S. and
Ph.D. theses. All ranks will be considered and the research area is
open. At the assistant professor level we are looking for candidates
with distinguished academic records who will develop imaginative
research and teaching programs and who will become future leaders
in the profession. Associate and full professor candidates should be
nationally recognized for their accomplishments. Resumes, reprints
of selected publications, a statement of research plans, and a list of
four references should be forwarded to: Professor Vasilios
Manousiouthakis, Chair, UCLA Chemical Engineering Department,
Box 951592, Los Angeles, CA 90095-1592. UCLA is an equal oppor-
tunity/affirmative action employer.

6. Wright, PR., F.J. Muzzio, and B.J. Glasser, "Effect of Resin Charac-
teristics on Expanded Bed Adsorption of Proteins," Biotechnol. Prog.,
15, 932 (1999)
7. Wright, PR., and B.J. Glasser, "Modeling Mass Transfer and Hydro-
dynamics in Fluidized Bed Adsorption of Proteins," AIChEJ., 47, 474
(2001)
8. Chase H.A., and N.M. Draeger, "Affinity Purification of Proteins Us-
ing Expanded Beds," J. ( . . ..597, 129 (1992)
9. Thommes, J., M. Halfar, S. Lenz, and M.R. Kula, "Purification of
Monoclonal Antibodies from Whole Hybridoma Fermentation Broth
by Fluidized Bed Adsorption," Biotechnol. Bioeng., 45, 205 (1995)
10. Batt, B.C., MM. Yabannavar, and V. Singh, "Expanded Bed Adsorp-
tion Process for Protein Recovery from Whole Mammalian Cell Cul-
ture Broth," Bioseparation, 5, 41 (1995)
11. Chang, Y.K., and H.A. Chase, "Ion Exchange Purification of G6PDH
from Unclarified Yeast Cell Homogenates Using Expanded Bed Ad-
sorption," Biotechnol. Bioeng., 49, 204 (1996)
12. Karau, A., J. Benken, J. Thommes, and M.R. Kula, "The Influence of
Particle Size Distribution and Operating Conditions on the Adsorp-
tion Performance in Fluidized Beds," Biotechnol. Bioeng., 55(1), 54
(1997)
13. Chang, Y.K., and H.A. Chase, "Development of Operating Conditions
for Protein Purification Using Expanded Bed Techniques: The Effect
of the Degree of Bed Expansion on Adsorption Performance,"
Biotechnol. Bioeng., 49, 512 (1996)
14. Wnukowski, P., and A. Lindgren, "Characterization ofthe Internal Flow
Hydrodynamics in an Expanded Bed Adsorption Column," presented
at Recovery of Biological Products VI, Interlaken Switzerland (1992)
15. Whitely, R.D., R. Wachter, F Liu, and N.H. Wang, "Ion Exchange
Equilibria of Lysozyme, Myoglobin, and Bovine Serum Albumin: Ef-
fective Valence and Exchanger Capacity," J. Chromatogr., 465, 137
(1989)
16. Zubay, G., Biochemistry, 2nd ed., Macmillan Publishing Company,
New York, NY (1988)
17. Richardson, J.FE, and W.N. Zaki, "Sedimentation and Fluidisation: Part
1," Trans. Instn. Chem. Engrs., 32, 35 (1954)
18. McCabe, W.L., J.C. Smith, and P. Harriott, Unit. *
cal Engineering, 4th ed., McGraw-Hill, New York, NY (1985)
19. Levenspiel, 0., Chemical Reaction Engineering, John Wiley & Sons,
Inc., (1972) 5

Winter 2004

classroom

ON THE APPLICATION OF

DURBIN-WATSON STATISTICS TO

TIME-SERIES-BASED REGRESSION MODELS

THOMAS Z. FAHIDY
University of Waterloo * Waterloo, Ontario, Canada N2L 3G1

A fundamental tenet in (linear) regression analysis is
that errors associated with a model must be random
and independent from observation to observation in
an experiment, with expectation (or mean value) zero. Vari-
ous aspects of residual behavior are routinely discussed in
modem texts on probability and statistics. The distribution
of

ek =Yk Yk k = 1,...,n
should show a random scatter when plotted against

Xk, Yk, or Yk
as abscissa.
If the statistical experiment involves observations in a time
sequence, and the error at time instant tk is influenced by the
error at the immediately previous time instant tk-, the result-
ing "influential carryover"'1,2]violates the error-independence
criterion. The errors may be negatively or positively corre-
lated.
The technique introduced by Durbin and Watson[3] more
than fifty years ago is a popular and straightforward test for
the existence of autocorrelation in time-series analysis (e.g.,
in forecasting). Only a small number of textbooks on prob-
ability and statistics intended for engineering and natural sci-
ences treats this subject matter, however.

The purpose of this article is to demonstrate the applica-
tion of the Durbin-Watson (DW) technique to regression
analysis concerning chemical engineering processes where
the "regressor"14] sequence occurs as a time series. Regres-
sion problems of this kind appear routinely in reaction kinet-
ics/chemical reaction engineering, applied transport phenom-
ena, process control, and engineering economics and plant
design, thus touching all major domains of the undergradu-
ate curriculum.
� Copyright ChE Division

The DW technique is illustrated by two examples. The first
is related to decisions concerning the order of a chemical re-
action. The second illustrates its usefulness in determining if
a regression model is statistically admissible, and as such, is
of major interest to chemical (and other) engineers.

BRIEF THEORY
Given the general first-order autoregressive process15]

p-1
Yk =P O + kXk,l+ek

k= ,...,n

where the errors are assumed to obey the first-order
autocorrelation
ek =pek-1 +Uk (2)
with p < 1, and independent random uk belonging to the nor-
mal distribution with zero mean and variance p2. The regres-
sor set {x} contains observations obtained at consecutive
time instants t1, t2, ..., t . In the case of correlated errors, the
variance of each error term is given by

G2(ek) - (3)
1ip2

Chemical Engineering Education

Thomas Z. Fahidy is Professor Emeritus of
Chemical Engineering at the University of Wa-
terloo. He obtained his BSc and MSc degrees
at Queen's University and his PhD at the Uni-
versity of Illinois Urbana-Champaign. His ma-
jor research and teaching interests are in ap-
plied electrochemistry electrochemical engi-
neering, applied engineering mathematics, and
applied probability and statistics. He can be
reached at

The purpose of this article is to demonstrate the application of the Durbin-Watson
(DW) technique to regression analysis concerning chemical engineering
processes where the "regressor""41 sequence occurs as a time series.

and the covariance of adjacent errors is

o (ek ek-) p (ek) (4)
To test the null hypothesis H0 : p = 0 against an appropriate
alternative hypothesis H1, the Durbin-Watson statistic

(ek - ek-)2
D k=2 SSED (5)
n 2 SSE
,e
k=1

is computed and compared to upper (dr) and lower (dL)
limits of D, as a function of observation size, in critical
tables.E 56] The decision scheme is given in Table 1.
The D-statistic is related to the Lag 1 autocorrelation15,
coefficient of residuals defined asN5

n
Yekek-1
r = k=2n (6)
Ye2
eki
k=l

by the simple relationship
D = 2(1 - ri) (7)
which is particularly useful for n < 15 since critical tables
do not extend outside the 15 < n < 100 range. If the in-
equality |r, > 2/ln stands, the independence of errors is in
serious doubt. The size of observations in the first example is
sufficiently large to use critical tables, whereas tables cannot

TABLE 1
Decision Schemes in the DW Statistical Test
Note: Rejection ofH is a - . . . , stronger result
than failure to reject it.

Test Hypotheses Criterion Decision

Ho: p 0; H : p>0 D

D > du Fail to reject Ho

Ho: p 0; H : p <0 (4-D) < dL Reject Ho in favor of H,
(4 - D) > du Fail to reject Ho

dL < D < d, Inconclusive
dL < (4 - D) < dp Inconclusive

be used in the second example.

EXAMPLE 1
Kinetics of the Bromination ofMetaxylene
The rate equation written in terms of bromine concentration
de
- kcm (8)
dt
has the rate constant k = 0.1 (dm3/mol)12 min-1 and apparent
order m = 1.5 at 17 �C. 1M As can be seen from Table 2 (next
page), the errors do not appear to be correlated, since the
DW-statistic D is larger than d, values at levels of signifi-
cance uo.
If we assume for the sake of argument, however, that the
decomposition is first order (m = 1), the test results depend
on the selected level of significance. Since R2, Radj2, and the
residual distributions (not shown) are not appreciably differ-
ent, the model carrying m = 1.5 is a better fit.
This conclusion is also supported by the 95% confidence
intervals for the true regression parameter b0 : (-0.6494;
0.3079) when m = 1.5 and (-3.6478; -2.01306) when m = 1;
in the second case, the correct value of zero does not even
fall into the interval
What happens if the decomposition is assumed to be of
zero order? With m = 0 in Eq. (8), the bromine concentration
would be a linear function of time. The c = po + P31t + error
model would have the sample regression parameters
b0 = 0.25849 and bI = -0.004119, with R2 = 0.857 and
so2 = 0.00724 (including the t = 63.00; c = 0.0482 observation
pair, lost by the rate-averaging process discussed in Ref. 8).
Since SSE = 0.03558 and SSED = 0.02419, however, the
DW statistic D = 0.7 is less than the d, values shown in Table
2, indicating a positive correlation between errors. The residual
distribution also being parabolic (i.e., definitely non-random),
the postulation of zero-order kinetics wouldbe statistically most
questionable, apart from its physical improbability.

EXAMPLE 2
Effect of Temperature/Humidity Index
on the Level of Pollution
The level of pollution as a function of the temperature/hu-
midity index, recorded on ten consecutive days at a certain
locationM91 are shown in Table 3. The problem assignment in
Ref. 9 is to determine if the data are suitable for a linear re-

Winter 2004

gression analysis.

Table 4 illustrates that increasing the degree of the poly-
nomial is not particularly effective, inasmuch as the ad-
justed R2 values indicate that even at best, only about 65%
of the variations in the pollution index are explained by
variations in the temperature/humidity index. The error
variances are also very similar.

The residual distribution in all three cases is reason-
ably random, and the numerical values of the Lag 1
autocorrelation coefficient magnitude are well below the
numerical value of 2/110 = 0.632. The errors appear to
be unrelated.

It is instructive to note that the power relationship Y =
PoXl would not yield a better fit with a nonlinear R2 =
0.690 (linearization yields ln(b,) = -5.77981 and b1 =
1.52312; the residual distribution is quasi-random).

FURTHER COMMENTS ON
THE DURBIN-WATSON TECHNIQUE

If the DW-statistic falls into the inconclusiveness zone, "reme-
dial measures" for autocorrelation may be applied: addition of in-
dependent variables, transformation of variables, the Cochrane-
Orcutt procedure, and the Hildreth-Lu procedure. The discussion
of these techniques is beyond the scope of this paper and may be

TABLE 3
Pollution as Function of Temperature/Humidity Index
x - temperature/humidity index; Y - coded pollution level

Day k 1 2 3 4 5 6 7 8 9 10

x OF 77 95 30 45 85 50 65 60 63 82
Y 1.5 4.0 0.5 1.4 2.0 0.8 2.5 2.0 1.7 2.8

TABLE 2
Application of DWT to the Kinetics of Metaxylene Bromination.
Experimental data are taken from Ref 8, Table 3.1.

k 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

t 0 2.25 4.50 6.33 8.00 10.25 12.00 13.50 15.60 17.85 19.60 27.00 30.00 38.00
X 0.3150 0.2812 0.2555 0.2353 0.2153 0.1980 0.1852 0.1713 0.1566 0.1465 0.1295 0.1107 0.0942 0.0799

yk 16.44 13.56 11.48 11.68 9.11 8.00

tk : observation time (min)
xk : mean bromine concentration (mol/dm3)
Yk : mean rate of reaction - 103 Ac/At (mol/dm3min)

7.73 7.71 5.87 4.06

3.64 3.23 2.79 2.10

41.00 45.00 47.00 57.00
0.0736 0.0692 0.0615 0.0518
1.55 1.35 1.25 1.18

Y - 0, + Px + error

-2.830640
57.830640
0.977
0.976
0.5305
9.86520
8.48330
1.162
a 0.01: Not correlated
a-0.025: No conclusions
a-0.05: Borderline positive correlation

Chemical Engineering Education

b,
b,
R2
Radj2
s 2
SSED
SSE
D
Decision on errors

Y=0 + p3x 5 + error

-0.170746
94.494843
0.987
0.986
0.3054
8.54756
4.88721
1.749
Not correlated at
S-0.01; 0.025; 0.05

Critical values of the DW statistic at n = 19f,562
a d, d,
0.05 1.16 1.39
0.025 1.03 1.26
0.01 0.90 1.12

found elsewhere.[10]
The DW technique may not indicate autocorrelated errors
associated with a second-order autoregressive pattern

ek =Plek-1 +P2k-2 + Uk (9)

and hence it is not robust against incorrect model specifica-
tions.
Alternative tests of autocorrelation include the Theil-Nagar
procedure[10,111 and the Olmstead-Tukey, Mann-Kendall,
Hotelling-Pabst, and von Neumann tests summarized briefly
by Powell.[6] To the author's knowledge, the Durbin-Watson
technique is more widely used.

CONCLUSIONS
Owing to the relative ease of its use, the inclusion of the
Durbin-Watson technique in a probability and statistics course
is well advised for the undergraduate chemical engineering
curriculum. It is somewhat surprising that the technique is
treated only by a small number of engineering textbooks, no-
tably the ones cited in this paper. Routine teaching of the
technique would further emphasize for students the impor-
tance of error structure analysis and help counteract their of-
ten-demonstrated inclination to assign inflated significance
to the R2 parameter.

ACKNOWLEDGMENT
Useful discussions with Dr. Tom Duever of the Depart-
ment of Chemical Engineering are hereby gratefully acknowl-
edged.

NOMENCLATURE
bI sample regression parameters, i.e., least-squares
estimators of true regression parameters P,, i = 1,...,p
c concentration

TABLE 4
Application of DWT to the Pollution Problem of Example 2.
Data are taken from Ref 9

Simple linear model Quadratic model Cubic model

b0 -0.80347 0.36495 -6.62620
b, 0.041771 0.001023 0.395176
b2 - 3.2274 x 10' -6.470 x 10-3
b, - - 3.644 x 10-5
R2 0.684 0.700 0.766
R 2 0.644 0.614 0.649
s 2 0.364 0.395 0.359
SSED 6.4739 6.0833 3.7614
SSE 2.9144 2.7654 2.1568
D 2.221 2.200 1.744
Ir.1 0.110 0.100 0.128

D Durbin-Watson statistic (Eq. 5)
dL, do lower and upper level bounds, respectively, in critical
tables of the Durbin-Watson statistic
e error (or residual), defined as the difference between
the observed and regressed value of the dependent
variable
k rate constant (Eq. 8)
m reaction order (Eq. 8)
n length of the time series and size of the observation
set
p size of the regression polynomial (simple linear: 2;
quadratic: 3, etc.)
R2 coefficient of determination; Rd 2 its adjusted value,
defined as 1 - [SSE/(n-p)]/[SSTV(n-1)]
r1 Lag 1 autocorrelation coefficient (Eq. 6)
se2 sample error variance, defined as SSE/(n-p)
t time; tk the k-th instant in the time series
u random variable (Eq. 2)
x independent variable (regressor)
Y dependent variable; Y regressed dependent variable

Greek Symbols
a level of significance in hypothesis testing
31 true population regression parameters, k =
UO2 true (population) variance
p true (population) correlation coefficient

1,...,p

Special Symbols
SSE sum of the squared errors (Eq. 5)
SSED sum of the squared error differences (Eq. 5)
SST total sum of squares in regression theory

REFERENCES

1. Hogg, R.V., and J. Ledolter, Engineering Statistics, Section 7.3, p.
287, Macmillan, NY, and Collier, London (1987)
2. Hogg, R.V.,andJ i ...-1I. I . . . .. Ohysi-
cal Scientists, 2nd ed., Section 9.3, p. 364, Macmillan, New York;
Maxwell Macmillan, Toronto; Maxwell International, New York; Ox-
ford, Singapore, Sydney (1992)
3. Durbin, J., and G. S. Watson, "Testing for Serial Correlation in Least
Squares Regression," Biometrika, 38, 159 (1951)
4. Walpole, R.E., R.H. Myers, S.L. Myers, and K. Ye, Probability and
Statistics for Engineers and Scientists, 7th ed., Section 11.1, p. 350,
Prentice Hall, Upper Saddle River, NJ (2002)
5. Neter, J., W. Wasserman, and M.H. Kutner, AppliedLinear Statistical
Models, 3rd ed., Section 13-3, p. 491, IRWIN, Homewood, Illinois
(1990)
6. Powell, F.C., Statistical Tables for the Social, Biological, and Physi-
cal Sciences, Cambridge University Press, Cambridge, United King-
dom (1982)
7. Priestley, M.B., Spectral Analysis and Time Series: Vol. 1. Univariate
Series, Section 3.3, p. 106, Academic Press, New York, NY (1981)
8. Hill, Jr., C.G., An Introduction to Chemical Engineering Kinetics and
Reactor Design, illustration 3.1, p. 44, John Wiley and Sons, New
York, NY (1977)
9. Strait, PT., A First Course in Probability and Statistics with Applica-
tions, Section 14.1, p. 455, HBJ Inc., New York, NY (1983)
10. Ref. 5, Section 13.4, p. 494
11. Theil, H., and A.L. Nagar, "Testing the Independence of Regression
Disturbances," J. Am. Stat. Assoc., 56, 793 (1961) 5

Winter 2004

classroom

TEACHING ELECTROLYTE

THERMODYNAMICS

SIMAO P. PINHO,* EUGENIA A. MACEDO
Universidade do Porto * 4200-465 Porto, Portugal

Electrolyte solutions can be found in many natural and
industrial processes. Some examples are the absorp-
tion of acid gases, such as carbon dioxide, for removal
from effluent gas streams, avoiding atmospheric pollution;[1'
the fractional crystallization processes in which several salts
are separated as pure phases from a multicomponent mix-
ture; for the production of fertilizers such as ammonium phos-
phate, ammonium nitrate, or potassium sulfate;121 for extrac-
tive distillation using salt as the extractive agent;3] and for
precipitation of globular proteins from an aqueous solution
by the addition of salts.[4]
It is not surprising that during the last few decades, much
attention has been devoted to experimental and theoretical
studies in this area. At the undergraduate level, however, most
of the thermodynamics courses still do not consider these
types of mixtures, and as a result the students are not given
enough insight into the differences when compared to non-
electrolyte thermodynamics. Nevertheless, several authors
have recognized this gap, and recent editions of the books by
Prausnitz, et al.,1 and Tester and Modell[61 include chapters to-
tally devoted to the thermodynamics of electrolyte solutions.
Electrolytes are usually classified according to their de-
gree of dissociation in solution: those undergoing a total dis-
sociation into cations and anions are called strong electro-
lytes, while the ones that participate in different chemical
reactions, such as ion association, are called weak electro-
lytes. This classification has no definite boundaries because
the degree of dissociation depends on, among other things,
the type of solvent and solute concentration. For instance,
zinc iodide is a strong electrolyte in water only if the concen-
tration is lower than about 0.3 molal.1F
In this paper, the thermodynamic description of a strong
electrolyte solution is illustrated by calculations on the freez-
ing point depression of strong electrolytes in water, empha-
sizing the differences between electrolyte and nonelectrolyte
thermodynamics. In this way, students can gain some knowl-

* Institute Politecnico de Braganga, 5301-857Braganga, Portugal.

edge on the physical chemistry of electrolyte solutions.

THE IDEALITY IN ELECTROLYTE SOLUTIONS
Freezing point depression is a colligative property that de-
pends on the number of solute particles but not on its nature.
If we consider a solution of a solvent 1 in which a solute A is
dissolved, the freezing point depression is defined as the dif-
ference between the melting temperature of the pure solvent,
T, and the freezing temperature e of the mixture, T (AT = T
- Tf). This last temperature is lower than the melting point of
the pure solvent. It is interesting to observe how the freezing
point changes with the amount of solute added to the solvent.
The simplest equation for the freezing point depression, which
is familiar to the students in a chemical thermodynamics
course, can be written as[8'

RT2
AT = Tm - Tf A (1)
AHf(Tm)
where AHf is the enthalpy of fusion at T., R is the ideal gas
constant, and xA is the solute mole fraction.
The different performance obtained, using Eq. (1), in the

Simao P. Pinho graduated in chemical engi-
neering from the University of Porto in 1992and
received his PhD from the same University in
2000. He became Professor Adjunto at Escola
Superior de Tecnologia e Gestdo, Instituto
Politecnico de Braganga, in 2000. His research
interests are in chemical thermodynamics and
separation processes.

Eugenia A. Macedo graduated in chemical
engineering from the University of Porto in
1978 and received her PhD from the same
University in 1984. She became Associate
Professorin the Chemical Engineering Depart-
ment at the University of Porto in 1990. Her
research interests are in chemical thermody-
namics and separation processes.

� Copyright ChE Division ofASEE 2004

Chemical Engineering Education

calculation of AT for nonelectrolyte and electrolyte solutions
can be easily compared. The relative percent deviations ob-
tained for the representation of freezing point depression for
aqueous solutions of D-fructose, ethylene glycol, NaCl, and
AgNO3, can be seen in Figure 1. Despite the fact that the
maximum mole fraction is around 0.01, for the NaCI and
AgNO3 aqueous solutions the deviations are much more pro-
nounced than for the nonelectrolyte systems, with errors
higher than 4% even at very low concentrations ( 5 x 10-4).
It should be mentioned that for those calculations, AHf(Tm) =
6010.0 J/mol and Tm = 273.15 K were used.[9]
One main assumption in the derivation of Eq. (1) is that the
solute is very dilute and forms an ideal solution. When, for
instance, NaCI is dissolved in water, the solution essentially
contains sodium and chloride ions. At this point it is impor-
tant to call the students' attention to the different nature of
forces depending on the kind of
solutes: the ions interact with each
other through coloumbic poten- 6 .
tial, which varies as 1/r. For neu-
tral solute molecules (nonelectro- 14
lytes) such as D-fructose, the in- 12
teractions vary something like 1/
r6. So the interactionbetween ions 10
in solution is effective over a i
much greater distance than the in- * "
teraction between neutral solute .
particles and, unlike what happens 4 -
in nonelectrolyte solutions, even 2
in very dilute solutions the long-
range nature of the electrostatic 000 0002 0
forces between the ions is respon-
sible for strong deviation from
ideal behavior. Thus, while Eq. (1) Figure 1. Comparison
is widely used for nonelectrolyte deviations in the calct
solutions, it cannot give reliable point depression fo
and electron
results for electrolyte solutions
since they are ideal at concentra-
tions too low to produce a measurable AT.
Figure 1 is a fine way of showing students the different
perspective that should be taken regarding the concept of ide-
ality at high dilution in electrolyte and nonelectrolyte solu-
tions. Another important difference that arises in the thermo-
dynamics of electrolytes is the concentration scale used. In
electrolyte, it is common to use the molality scale instead of
the mole fraction scale. Moreover, in order to properly ac-
count for the number of solute particles in solution, due to
the dissociation of the electrolyte, the mole fraction of solute
A used in Eq. (1) should be calculated as

XA =- A (2)
vnA + n1
where nA and n1 are the solute and solvent mole numbers,
respectively, and v is the sum of the stoichiometric coeffi-

clients of the anion and the cation.

THE DEBYE-HUCKEL THEORY AS THE
PATH FOR NON-IDEALITY
IN ELECTROLYTE SOLUTIONS
So far, the students have learned that, for electrolyte solu-
tions, assuming ideality may introduce significant errors in
the calculation of the properties of the solution, even at high
dilution. Thus, in order to obtain trustworthy values of AT,
corrections to the ideal behavior should be introduced using
the activity coefficient. From the thermodynamic condition
for equilibrium and after some reasonable assumptions, it is
possible to obtain1'8

n 71x = 1-- 1 (3)
R Tm TI

U * * * *" *
c Ethylene Glycol I
Fructose
* NaCI
* AgNO3

004 0.006 0.008 0010
Mole fraction

of the relative percentage
elation of the ideal freezing

r aqu
yte s

where yl is the solvent activity
coefficient and x1 is its mole frac-
tion. Now, AT can be calculated
by solving Eq. (3) for Tf
Taking into account only the
electrostatic forces, assuming ions
to be charged points in a continu-
ous medium of uniform relative
permittivity, and using well-estab-
lished concepts from classical
electrostatics, Peter Debye and
Erich Hickel[11] derived the fol-
lowing expression for the mean
ionic molal activity coefficient of
an electrolyte (y *)

, A+z_- z
n - 1+ Bai-

eous nonelectrolyte In Eq. (4), A and B are parameters
olutions.9101 related to the density and dielec-
tric constant of the solvent,[5,12,13]
and a is the so-called distance of closest approach between
ions (usually taken as 4 A), z+ and z are the charges of the
cation and the anion, respectively, and I is the ionic strength
defined by

Nons
I=0.5 , mz2 (5)
1=1
being m1 the molality of the ion i and N the number of
types of ions in the solution.
The ionic strength is a very common measure of concen-
tration in electrolyte solutions. In fact, it takes into account
not only the concentration of the ion but also the magnitude
of its charge. A big difference comes from the fact that using
this model, the freezing point depression is now not only de-
pendent upon the solute concentration, but also on its charges.

Winter 2004

So the characterization of the electrolytes, in terms of its ions valei
is fundamental to establish differences that occur when applyint
proposed methodology for the study of the freezing point depre,
of different types of electrolytes. Depending on the charge of the
ion and the anion, the electrolytes can be classified as 1:1, 2:1,
2:2, etc. For example, a 2:1 type has a cation of double charge ar
anion of unit charge.
From Eq. (4), taking into consideration the Gibbs-Duhem equa
the activity of the solvent can be calculated as

tnylx1 = -MlVm(1 -Az+z_ - o(Ba-i))

where M1 is the molar mass of the solvent (kg/mol), and s(y) i
function

o(y)= -f(1+y-2tn(1+y)- 1-

The full understanding of the thermodynamic concepts that m
possible the derivation of Eq. (6) from Eq. (4) is far beyond the s
of this paper, but it is important to refer to some of the most rele
points such as the definition of the activity coefficients in diff
concentration scales, the standard states and the normalization o
activity coefficients, and the need for defining mean ionic propel
which are calculated based on the properties of the ions.-I'71 These
cepts introduce significant changes to the nonelectrolyte therm
namics and should be carefully discussed with the students.
Inserting the result for fn y1xl given by Eq. (6) into Eq. (3),
possible to obtain better estimates for AT in electrolyte solutions.
ingA = 1.130 kg0 5/mol0 5 and B = 3.246x109 kgo 5/(m mol05), obta
by using values of the solvent density and dielectric constant for N
at 273.15 K,[9] one can calculate, for comparison with the prev
results shown, new values of AT for aqueous NaCl solutions.
The errors obtained assuming ideal behavior and using the De
Hiickel equation are compared in Figure 2. Using this new meth(
ogy, the errors in calculated values of AT are only higher than 4�

10

6 -

4

0.000 0.002 0 004 0006 0008 0010
NaC1 mole fraction

Figure 2. Comparison of the relative percentage de-
viations in the calculation of the freezing point de-
pression: ideal behavior and the Debye-Hiickel
equation. NaCl/water system.9 10]

aces, x, around 0.05. In fact, the Debye-Hiickel theory gives
g the an exact expression for the activity coefficients of the
;sion electrolyte and of the solvent for very dilute solutions,
cat- and as can be seen, the errors for AT at very low solute
1:2, mole fraction are near zero.
id an In Figures 3 and 4, the freezing point depressions are
shown for different types of electrolytes at low molal-
tion, ity in water assuming ideality and using the Debye-
Hiickel equation. In all cases the assumption of ideal-
ity agrees only with the experimental values at very
low concentrations, and the molality range of applica-
s the ability of this equation decreases as the valences of the
ions increase. This is evident in Figure 3 since the ideal

(7) 2.0 F- .

makes -- Ideal
akes Debye-Htickel, type 1.1
cope 5 ......... Debye-Huckel, type 2.2
;vant + Type 11
rent E Ty Tpe 2:2
f the .0 ..
rties, < + ,.*
con- - ...-," F
ody- 0-

it is
00
Fix- 0.0 01 0.2 03 0.4 05
iined Molality
vater
ious Figure 3. Comparison of the freezing point depres-
sion for 1:1 (HNO,, LiC1, NaC1, NaBr, NaOH, NaNO3,
KC1, KBr, KI, KOH, KNO,, CH3COOH, NHC1, and
bye- AgNO) and 2:2 (MgSO4, MnSO, ZnSO,, and CuSO4)
)dol- electroclytesf9,10' ideal behavior and the Debye-Hiickel
ofor equation

3.0
-- Ideal
2.5 - - - - Debye-Hiickel, types 1:2 and 2:1
V Type 1 2
2.0 - Type 2:1
5 V

1.0
-,f

05 V

0.0
00 0.1 02 0.3 0.4 0.5
Molality
Figure 4. Comparison of the freezing point depres-
sion for 1:2 (Na 2CO3, Na2SO4, Na2S203,, K2CO3, K2SO,,
and (NH)2SO,) and 2:1 (BaC12, CsC12, MgC12, SrC12,
and CaC2) electrolytes:[9] ideal behavior and the
Debye-Hiickel equation.

Chemical Engineering Education

curve in terms of molality is the same for 1:1 and 2:2
electrolytes, while the experimental data are different.
The improvement observed upon using a simple model
such as the Debye-Hiickel model is even more evident
for 1:2, 2:1, or 2:2 electrolytes than for 1:1 electrolytes.
Nevertheless, the Debye-Hiickel model allows more
accurate calculation of the freezing point depression
to higher concentrations for all types of electrolytes.
This brief discussion alerts the students to the
changes that must be made for the description of elec-
trolyte systems. Also, there can be significant differ-
ences when comparing the behavior of aqueous solu-
tions of electrolytes of different valences, which is ex-
plored in the next section by extending the calcula-
tions to concentrated solutions.

25
- Debye-Hctck el
20 1 NaC1, b = 0.1013 kg/mol A
A LiC1, bj= 0.3315 kg/mol
S NaNO3, b= -0.1026 kg/mol

C-)

S 000 1.00 2.00 3.00 400 5.00
Molahty

Figure 5. Analysis of the Guggenheim equation in
the description of the freezing point depression of
aqueous solutions with 1:1 electrolytes.!9,101 Improve-
ment to the Debye-Hiickel equation.

50
- Debye-Hiickel
40 E K CO , b= 0.0608 kg/mol U
A Na2S203 b = -0 0628 kg/mol E-
30 C CaCl2, b = 02267 kg/mol

20

10

00 1.0 2.0 3.0 4.0 5.0
Molahty
Figure 6 Analysis of the Guggenheim equation in the
description of the freezing point depression of aque-
ous solutions with 1:2 or 2:1 electrolytes[91 Improve-
ment to the Debye-Hickel equation.

EXTENDING THE FREEZING POINT CALCULATION
FOR CONCENTRATED SOLUTIONS
The main assumption of the Debye-Hiickel theory is that deviations
from ideality are only due to electrostatic forces between the ions, which
is physically reasonable at high dilution but unreal when the ionic con-
centration increases so the ions more closely approach each other and
short-range forces become dominant. Guggenheim suggested the use
of a power series in electrolyte concentration to better describe the
physical chemistry of electrolyte solutions, leading to the virial expan-
sion models. To do so, Guggenheim added a new specific electrolyte
empirical interaction parameter (b ), proposing the following equation
for the mean ionic molal activity coefficient:[13]

S A+zz +Z1
fny =- 1 + bI (8)

From Eq. (8), the activity of the solvent is given by

nyx1 =-Mlvm[1-Alzz_+z o(-)+-b-j (9)

It is interesting for the students to evaluate how this change makes
possible a much better quantitative description of the freezing point
depression at high concentrations. Thus, using an experimental value
of the freezing temperature at a concentration around 1 molal, it is
possible to obtain a value for the empirical parameter b+. For in-
stance, the experimental value for an aqueous NaCl solution of 0.90
molal is Tf = 270.11 K; from this, b+ = 0.1013 kg/mol is calculated.
Now, combining Eqs. (3) and (9) makes it possible to study the
usefulness of the equation proposed by Guggenheim for calcula-
tion of the freezing point depression.
Figure 5 presents a comparison between the Debye-Hiickel and
Guggenheim equations for the estimation of AT in aqueous solutions
of electrolytes of type 1:1 at concentrations up to 5 molal. It can be
easily observed that the use of the Guggenheim equation, with a new
empirical parameter regressed from a unique experimental freezing point
measurement, introduces a significant improvement in the representa-

35

3.0 -

15 -
5, - - - - Debye-Hiickel
1.0
S , MnSO , b,- 0.0389 kg/mol
05 ] CuS04 , b� -0.0028 kg/mol
0 0 , , ,
0.0 0.5 1 0 1.5 20
Molality
Figure 7. Analysis of the Guggenheim equation in the descrip-
tion of the freezing point depression of aqueous solutions with
2:2 electrolytes.[9' Improvement to the Debye-Hickel equation.

Winter 2004

tion of AT for all systems shown. Compared with the previ-
ous results shown for the NaCl/water system, the application
of this equation results only in a percentage deviation higher
than 4% for solute mole fraction around 0.15 ( 5 molal),
which is a 3-times-higher concentration than the results
achieved using the Debye-Hfickel equation. The use of the
Guggenheim equation for the systems of water/LiCl and wa-
ter/NaNO3 shows an even
greater improvement over the
Debye-Hfickel equation. TA
In Figures 6 and 7, the same Comparison of Different
kind of comparison is pre- the Freezing Point Depr
sented for, respectively, 1:2 and
2:1, and 2:2 electrolytes in wa- Salt Data Maximum
ter. The results obtained pro- Type Sets Molalvy
vide a very reasonable correla- 1:1 14 5.1 1
tion of the experimental data 2:1 5 4.2 2
1:2 6 4.8 3
and only for the system of wa- 1:2 6 4.8
ter/K2CO3 are there big dis- 2:2 4 1.7 8
crepancies relative to the ex-
perimental results for solute mole fractions higher than 0.15
( 3.4 molal), which is, nonetheless, a very good result.
Moreover, the Guggenheim equation makes it possible to
calculate a freezing point depression up to 400C (CaCl2
system, Figure 6).
Table 1 summarizes the deviations obtained in the repre-
sentation of the freezing points of different aqueous electro-
lyte solutions. It gives a more comprehensive comparison
between all methodologies considered here and the type of
electrolyte. First, one sees that the deviations from ideality
increase as the valences of the ions increase. The Debye-
Hiickel equation introduces improvements for all types of
systems, which are especially evident for the 2:2 electrolytes.
In that case, the maximum molality is much lower than in the
other cases, and that is certainly a contributing factor in the
big improvements obtained. Finally, it is important to stress
that based solely on one experimental data point for each salt,
a simple model like the Guggenheim equation makes it pos-
sible to calculate the freezing point for all systems with aver-
age error of about 2.10%.
Since colligative properties depend on the number of par-
ticles in solution, the freezing point data can be analyzed in
terms of the physical chemistry of the electrolyte solutions.
That is, it might give indications of the degree of dissocia-
tion, solvation, and ion-pairing. The students can also be asked
to consider other hypotheses that could be made or improved
for electrolyte solutions in the development of the models
studies here, and further to consider more complex models
such as the Pitzer model in the representation of thermody-
namic properties of electrolyte solutions.

CONCLUSIONS
The differences that must be taken into account when study-

BL
App]
essio
ysten

'deal
0.00
2.51
4.55
1.85

Chemical Engineering Education

ing aqueous electrolyte systems rather than nonelectrolyte
systems have been pointed out in this paper. Specifically, we
have shown that even at very high dilutions, one must use the
Debye-Hiickel type limiting law to properly represent the
freezing point depression. In this way, the students can com-
pare the experimental data with values assuming the ideal
behavior and using the Debye-Hiickel equation. Finally, the
students are also challenged to
understand the need for more
E 1 elaborate expressions in the
roaches for Calculation of representation of that property
n in Aqueous Electrolyte at high concentrations. To do
ns. this, we suggest obtaining an
Error (%o) empirical parameter of the
Debye-Hiickel Guggenheim Guggenheim equation using
6.89 1.19 an experimental data of the
18.56 2.77 freezing point depression at a
13.59 2.71 concentration around 1 molal.
11.86 4.02
_11.86 4.02 This simple analysis of elec-

trolyte solutions is certainly a
nice starting point to motivate students to get some knowl-
edge of electrolyte thermodynamics. It can be introduced in
a thermodynamic or a physical-chemistry course, which could
be even more attractive if it can be combined with a labora-
tory experiment for measurement of the freezing point de-
pression of an aqueous electrolyte solution.

REFERENCES
1. Maurer, G., "Electrolyte Solutions," Fluid Phase Equilibria, 13, 269
(1983)
2. Thomsen, K., "Aqueous Electrolytes: Model Parameters and Process
Simulation,"PhD T , i . .0 1 . l. ,, i.li i. ,_,1, 1,1,,, Tech-
nical University of Denmark, Lyngby (1997)
3. Furter, W.F., "Extractive Distillation by Salt Effect," Chem. Eng.
Comm., 116, 35 (1992)
4. Prausnitz, J.M., "Some New Frontiers in Chemical Engineering Ther-
modynamics," Fluid Phase Equilibria, 104, 1 (1995)
5. Prausnitz, J.M., R.N. Lichenthaler, and E.G. Azevedo, Molecular Ther-
modynamics of Fluid-Phase Equilibria, 3rd ed., Prentice-Hall,
Englewood Cliffs, NJ (1998)
6. Tester, J.W., and M. Modell, Thermodynamics and Its Applications,"
3rd ed., Prentice-Hall, Englewood Cliffs, NJ (1977)
7. Robinson, R.A., and R.H. Stokes, Electrolyte Solutions, 2nd ed.,
Butterworths, London, UK (1970)
8. Sandler, S.I., Chemical and Engineering Thermodynamics, 3rd ed.,
John Wiley & Sons, New York, NY (1999)
9. Lide, D.R., (Ed.), CRCt . . , physics, 79th ed.,
CRC Press, Boca Raton, FL (1999)
10. Clarke, E.C.W., and D.N. Glew, "Evaluation Functions for Aqueous
Sodium Chloride from Equilibrium and Calorimetric Measurements
Below 154 C," J. Phys. Chem. Ref Data, 14, 489 (1985)
11. Debye, P., and E. Hfickel, "Zur Theorie der Elektrolyte I. Gefrier-
punktserniedrigung und Verwandte Erscheinungen," Phys. Z., 24, 185
(1923)
12. Pinho, S.P., "Phase Equilibria in Electrolyte Systems," PhD Thesis,
q 1 " 11, 1 , .." 1 _ I., I,, I i ,. _.1.,, 1.1,. University of Porto, Porto, Por-
tugal (2000)
13. Zemaitis, Jr., J.F, D.M. Clark, M. Rafal, and N.C. Scrivner, Hand-
book of Aqueous Electrolyte Thermodynamics: Theory and Applica-
tion," AIChE, New York, NY (1986) 5

Mnbook review

The Pilot Plant Real Book:
A Unique Handbook for the Chemical
Process Industry
by Francis X. McConville
Published by FXM Engineering and Design, 6 Intervale Road,
Worcester MA 01602 (2002)

Reviewed by
Ka M. Ng
Hong Kong University of Science and Technology

The pilot plant is indispensable in the development of
chemical processes. Yet it is seldom covered in a typical
chemical engineering curriculum, leaving it as one of the sub-
jects that the graduate is supposed to learn "on the job." The
author suggests that this omission is a failure of today's edu-
cational system. Given the importance of pilot plant, which
can be viewed as one of the four elements of process devel-
opment, [1] there is some truth in this assertion. At least this
omission forgoes an opportunity to show the students how
basic principles, experiments, know-how, experience, simu-
lations, literature data, workflow, etc., come together in the
development of products and processes.
If you are an educator, a process development chemist, or
engineer, who shares McConville's view that there is a gap
in pilot plant education and practice, this book may be just
what you want. It provides a lucid account of how chemical
processes are transferred from the lab to the plant. The infor-
mation often needed for pilot plant personnel is organized in
a logical and readily accessible manner. This book is named
a "Real Book"-McConville explains that just as young jazz
musicians had to master the "Real Book," a bootleg, photo-
copied collection of the great jazz standards with all the songs
anyone needed to know in one place, this book has admira-
bly achieved a similar objective for pilot plants, particularly
those for the pharmaceutical industry.
Chapter 1 sets the tone by describing the role of a pilot
plant. It contains a wealth of hints on factors to consider and
things to do and not to do in scale-up, which is one of pri-
mary functions of a pilot plant. Some of the terms and jargon
commonly used in pilot plant such as work-up, batch record,
campaign report, equipment qualification, cGMP, and others
are explained. Chapter 2 describes the key pieces of equip-
ment and their operations in a typical pharmaceutical pilot
plant. Consider the discussion on the reactor. It complements
a chemical reaction engineering textbook in which reactor
theory and kinetics is covered by focusing on the practical
issues such as reactor types and configurations, selection
criteria, raw material charging, sampling methods, reac-

) tor cleaning, etc.

Winter 2004

Chapters 3, 4 and 5 are concerned with liquid handling,
heat transfer, and electrical instrumentation, respectively, all
basic issues in a pilot plant. Solvents are covered in Chapter
6. It identifies the solvents useful for crystallization, and those
limited for pharmaceutical use, as well as their physical and
chemical properties. Binary azeotropes for some common
solvents are also listed. These data are important for pilot
plants because it is often possible to take advantage of them
to improve the efficiency of drying and solvent exchange op-
erations by distillation.
Compressed gases are covered in Chapter 7. Proper proce-
dures for handling compressed gases, metering gases, using
gas pressure regulators, installing a vacuum pump, etc., are
described. Chapter 8 provides data on the properties of com-
mercial acids and bases, and buffers. The aqueous solubility
of various inorganics and organic are also given.
Chapters 9 and 10 are concerned with chemical hygiene
and safety, and materials selection, respectively. Chapter
11 contains miscellaneous topics such as unit conversion
tables, sieve sizes, etc., that might come in handy in daily
pilot plant operations.
There are many books on process development, equip-
ment and chemical data,[2-61 but this book is unique. Captur-
ing the experience of a seasoned pilot plant practitioner, it
delivers what is wanted and needed in a compact package,
particularly for pharmaceutical pilot plant projects. The top-
ics selected are highly relevant, the extent of coverage is to
the point, the data chosen are consistent with what a chemist
and engineer might need, and the style of writing is direct
and concise. There is also an extensive bibliography in case
additional information is required on the various topics.
This beautiful book is highly recommended for pilot plant
personnel as well as people engaging in chemical processing
and research. Its contribution to the education of process de-
velopment is still limited, however. My suggestion is to in-
clude pilot plant case studies to illustrate how the informa-
tion and tools are used to complete a process develop-
ment project, thereby taking it one step closer to a truly
"Real Book."
References Cited
1. Ng, K. M., and C. Wibowo, "Beyond Process Design: The Emergence
of a Process Development Focus," Korean J. Chem. Eng., 20, 791
(2003)
2. Woods, D. R., Process Design and Engineering Practice, Prentice-
Hall, Upper Saddle River, NJ (1995)
3. Woods, D. R., Data for Process Design and Engineering Practice,
Prentice-Hall, Upper Saddle River, NJ (1995)
4. Mansfield, S. Engineering Design for Process Facilities, McGraw-
Hill, New York, NY (1993)
5. Sandler, H. J., and E. T. Luckiewicz, Practical Process Engineering,
XIMIX, Philadelphia, PA (1987)
6. Ulrich, G. D., A Guide to Chemical Engineering Process Design and
Economics, Wiley, New York, NY (1984) 5

Random Thoughts...

CHANGING TIMES AND

PARADIGMS

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

Colleagues at a large public university I recently vis-
ited are doing some excellent research on first-year
engineering students-what attracted them to engi-
neering, how they view engineering as a curriculum and ca-
reer, how they feel about their first-year courses (it isn't
pretty!), their confidence levels before and after those courses,
and why the ones who drop out do so. I sat in on one of their
weekly meetings, and one of them-an education professor-
expressed bewilderment and dismay that with so much known
about what makes teaching effective, engineering programs
persist in using the same old ineffective methods. She won-
dered if there was any point in continuing research directed
at improving a system that is this intransigent.
I've heard the same thing from others engaged in educa-
tional reform-it's definitely an uphill battle, and it's easy to
get discouraged whenyour focus is restricted to a single cam-
pus. Taking a broader view, though, things don't look that
bad. Engineering education went through a major sea change
once before, and the signs are that it is doing so again. I tried
to offer some words of encouragement at the meeting and
thought I'd repeat them here for readers engaged in similar
lonely battles.
First, a little history. From the late 19th century through the
1950s, engineering education was a combination of lecture
and hands-on instruction closely tied to industrial practice,
and the faculty consisted primarily of experienced engineers
and consultants to industry. Inthe mid-1950s, America seemed
to be falling behind Russia in the space program and calls
were issued for an increased curricular emphasis on the math-
ematical and scientific foundations of engineering. In the years
that followed, external funding opportunities for basic re-
search skyrocketed, faculty started to be hired primarily for
their potential as researchers, and most laboratory and field
experiences disappeared from the engineering curriculum to
be replaced by lectures on applied math and science. The para-

digm shift from practice to science was essentially complete
in most engineering schools by the early 1970s.
In the 1990s, a rising chorus of complaints from industry
about the inadequate preparation of new engineering gradu-
ates for industrial jobs started to be acknowledged inside the
academy. In addition, evidence began to emerge from both
cognitive science and empirical classroom research that the
prevailing instructional model ("I show derivations of for-
mulas in class, then you plug into the formulas and do simi-
lar derivations in assignments and on tests") was ineffective
for promoting learning and the acquisition of critical think-
ing and problem-solving skills. Teaching workshops began
to be heavily subscribed at engineering conferences and on
campuses around the country, and NSF-funded programs and
individual campus initiatives-such as Project LE/ARN at
Iowa State-began to involve hundreds of previously tradi-
tional engineering faculty in education reform. Another ma-
jor step was ABET's adoption of new accreditation criteria
that required engineering programs to address both technical
and social outcomes in their curricula, all but forcing them to
adopt nontraditional methods in their classroom instruction.
(You clearly can't equip students with the ability to work ef-
ficiently in multidisciplinary design teams or give effective
technical presentations by giving them a few lectures on those
topics.)

Richard M. Felder is Hoechst Celanese Pro-
fessor Emeritus of Chemical Engineering at
North Carolina State University. He received
his BChE from City College of CUNY and his
PhD from Princeton. He is coauthorof the text
Elementary Principles of Chemical Processes
(Wiley 2000) and codirector of the ASEE Na-
tional Effective Teaching Institute.

� Copyright ChE Division ofASEE 2004

Chemical Engineering Education

These developments have given rise to a national move-
ment toward a more active, cooperative, problem-based in-
structional model for engineering education. While the new
approach cannot yet be said to have become dominant and
some universities seem determined to resist it (and ABET) to
the bitter end, evidence of its eventual ascendancy is mount-
ing. In the remainder of this article I want to share some of
the evidence I've recently seen.
I've given teaching workshops on campuses around the
country since the late 1980s in which I dis-
cuss active and cooperative learning, and I
usually ask the participants to raise their hands
if they use those methods in their classes. Ten Eng
years ago, two or three hands would typically educc
be raised. Now, 25-50% of the participants thr
indicate that they use active learning and lower
but still significant percentages use coopera- major
tive learning. This trend was also indicated by onc
a 1997 survey of over 500 engineering fac- and th
ulty at eight schools who were shown to be that i
representative of their faculties in most im-
portant respects. Many of the respondents re- so a,
ported regularly using active learning, team- deve
based assignments, and other student-centered have
methods11 to a
I frequently see impressive instructional in-
novations on campuses I visit and learn about
others in the literature and at conferences, the a mo
most dramatic of which involve project-based cooJ
and problem-based learning. Extensive re- probl
search has shown that students leambest when instruct
they perceive a clear need to know the mate-
rial being taught. Project/problem-based learn- for en
ing (PBL) uses this principle by introducing ed
course material on a just-in-time basis in the
context of realistic engineering problems and
projects. This instructional strategy has been used for many
years at the Colorado School of Mines and McMaster Uni-
versity, and numerous published articles report its successful
adoption at other universities around the world. An outstand-
ing example is ChemEngine (www.chemengine.net), a stu-
dent-owned and operated consulting firm at Virginia Com-
monwealth University that tackles engineering problems for
industrial clients and has saved those clients millions of dol-
lars in its few years of existence.
PBL has become the foundation of some course sequences
and clusters and departmental curricula. Texas A&M and sev-
eral other schools in the Foundation Coalition have trans-
formed their freshman engineering programs, integrating the

basic science and math courses traditionally taught in isola-
tion and emphasizing their interrelationships and applications
to engineering problems. In the spiral curriculum in chemi-
cal engineering at Worcester Polytechnic Institute, traditional
content is taught on a just-in-time basis in a sequence of
project-based courses. In each year of the curricula of sev-
eral engineering departments at the University of Queensland
in Australia, one or two project courses are taught that antici-
pate and integrate the material taught in parallel traditional
courses. Several entire universities have taken one form or
another of PBL as the basis of all of their
curricula, including the University ofAalborg
ring in Denmark and Olin University in Massa-
i went chusetts.
h a This is not to say that engineering educa-
tion reform is a done deal. If you look into a
change random class at a random engineering school
fore, today, you are still likely to see a professor
gns are deriving equations on a board, or (worse)
doing flashing PowerPoint slides of derivations to
half-asleep students in a half-empty room,
.... and administrators abound who still argue
ments that this approach somehow promotes learn-
n rise ing (research evidence to the contrary not-
ional withstanding). It may indeed turn out that ten
years from now the old teacher-centered ap-
toward proach will still dominate engineering edu-
ctive, cation. I doubt it, though, considering (a) the
rtive, active, cooperative, and problem-based
based courses and curricula springing up at univer-
sities everywhere, the concurrent growth of
il model engineering-based programs that equip fac-
eering ulty and graduate students to implement those
ion. instructional strategies, and the new ABET
criteria that (if seriously enforced) will com-
pel their use, (b) the power of instructional
technology to provide stimulating interactive
lessons and the growing occurrence and effectiveness of its
use at both traditional and on-line institutions, and (c) an
awareness among high school graduates that alternative
methods exist and an increasing unwillingness on their
part to put up with the old approach (a point that clearly
came out in the study mentioned at the beginning of this
column). Again, these things are never certain, but with
all that going on it's clear to me that the new paradigm is
the horse to bet on.
References
1. Practices.pdf> EO

Winter 2004

inee
rtioi
oug
sea
e be
e si
t is
gaii
lop
give
nati
rent
re a
era
em-
ion'
gin
ucat

All of the Random Thoughts columns are now available on the World Wide Web at
http://www.ncsu.edu/effectiveteaching and at http://che.ufl.edu/-cee/

e 9 /laboratory

NANOSTRUCTURED MATERIALS

Synthesis of Zeolites

STEVEN S.C. CHUANG, BEI CHEN, YAWU CHI, ABDELHAMID SAYARI
The University ofAkron * Akron, OH 44325-3906

Z eolites are crystalline aluminosilicates whose princi-
pal constituents are aluminum, silicon, and oxygen.[1]
They were discovered by Baron Axel F. Cronstedt,
who coined their name using the Greek words zeo (to boil)
and lithos (stone) because they bubbled under heating.[2-4] The
fundamental building blocks of the zeolite framework are the
tetrahedral units: [SiO4] and [A104]-. The silicate [SiO4] unit,
shown in Figure la, consists of a silicon atom surrounded by
four oxygen atoms; [A104], in which Al replaces silicon at
the center of the tetrahedron, bears a negative charge. This
charge is balanced by that of positive metal ions, mostly al-
kali cations, sitting in the gaps of the framework.

During zeolite synthesis, the tetrahedral units are joined
together via a common oxygen atom to form rings or cage
structures, referred to as secondary building units (SBU). The
SBUs can be assembled in many ways to produce various
types of zeolites. For example, the so-called 5-1 ring SBU
(see Figure lb) generates either ZSM-5 or ZSM-11. Figure
Ic illustrates the construction of a continuous framework of
ZSM-5 using 5-1 SBUs. This zeolite (Figure Id) exhibits two
intersecting channels: one straight and the other zig-zag.

The extensive research in zeolites was initiated after rec-
ognition of the similarity in the composition of zeolites and
silica-alumina.[5] The latter was used as a cracking catalyst in
refineries in the 1950s. At present, zeolites, including ZSM-
5, are used to process over 7 billion barrels of petroleum and
other chemicals annually, producing tens of billions of dol-
lars per year in revenues.[6] ZSM-5, one of the most widely
studied zeolites, has dominated the patent literature in appli-
cations of nanostructured materials. In addition, the well-de-
fined pores or cavities in nanometer range give rise to unique
molecular sieving capabilities and high internal surface ar-
eas suitable for a wide range of applications in fields other
than industrial catalysis, e.g., ceramics, electronic materials,
drug release media, sorbents, and ion exchangers.

The syntheses of zeolites often require the use of small or-
ganic species such as quaternary ammonium ions (e.g., R4N)
as templates or structure-directing agents. Detailed mecha-
nistic studies of ZSM-5 synthesis suggest that the hydropho-
bic hydration sphere formed around TPA (i.e., tetrapropyl
ammonium ion) is replaced by inorganic species forming an
organic-inorganic nanocomposite. Aggregation of these spe-
cies results in nucleation and eventually crystal growth in a
layer-by-layer fashion.[78] In 1992, Kresge, et al., had the
clever idea of using supramolecular assemblies such as sur-
factants or polymer liquid crystals as templates, instead of
the individual molecules or cations currently used as struc-
ture-directing agents for the synthesis of zeolites.[5,9,10] The
ordered materials were obtained and referred to as mesoporous
molecular sieves (MMS). These materials are similar to zeo-
lites, with the notable difference being that their pore sizes
are much larger (i.e., from 2 to over 30 nm).
The evolution from the concept of structure-directing to

Steven S.C. Chuang is Professor and Chair of the Chemical Engineer-
ing Department at The University ofAkron. He received his PhD from
the University of Pittsburgh in 1985. He teaches chemical process con-
trol, materials science, and chemical engineering laboratory His research
interests are in catalysis and reaction engineering. @uakron.edu>
Bei Chen is currently a research associate at the Oak Ridge National
Laboratory She received her PhD in Chemical Engineering under the
direction of Professor Chuang at The University of Akron in 2003. She
received herBS from the East China University of Science and Technol-
ogy in 1996 and her MS from The University of Akron in 1999, both in
chemical engineering.
Yawu Chi received his PhD in Chemical Engineering under the direc-
tion of Professor Chuang at The University of Akron in 2000. He re-
ceived his BSChE from Dalian University of Technology and his MS in
physical chemistry from Dalian Institute of Chemical Physics, both in
China. He is currently a Research Engineer at Stone and Webster, Inc.,
a subsidiary of The Shaw Group, Inc.
Abdelhamid Sayari is Professor of Chemistry at the University of Ot-
tawa. His PhD degree is from the University of Tunis and the University
Claude Bernard, Lyon, France. His major research interests involve
heterogeneous catalysis, focusing on the synthesis, characterization and
applications of zeolites and nanostructured porous materials.

� Copyright ChE Division ofASEE 2004

Chemical Engineering Education

that of supramolecular assemblies has led to rapid develop-
ment in the synthesis of nanostructured materials. These novel
nanomaterials opened new opportunities in many areas, such
as biosensing, drug delivery, bioseparation, and heterogeneous
catalysis.[11-20] The simplicity in the concepts of template syn-
thesis, along with the complexity of interrelated factors in
zeolite synthesis, make ZSM-5 synthesis an excellent project
that allows students to integrate basic principles of
nanomaterials synthesis into reaction engineering.
This paper describes an experiment on ZSM-5 synthesis
that was performed in our juniors' chemical engineering labo-
ratory at the University of Akron. The objective of this ex-
periment is to provide hands-on experience for the students
that includes formation of working teams, performing litera-
ture searches, grasping basic concepts of nanostructured
material synthesis, experimental design, reactor operation,
infrared spectroscopic analysis, troubleshooting, and
learning assessment.

EXPERIMENTAL
Materials and Equipment * Sodium hydroxide (certified
A.C.S. grade), tetrapropyl ammonium bromide (i.e., TPA)
(98+%), and sulfuric acid (0.1 M standardized solution) were
obtained through Alfa Aesar; sodium aluminate (-8% H20,
99.9% Al) was purchased from Strem Chemicals; and Aerosil
silica was generously donated by Cabot Corporation. All
chemicals were used without further purification. The hy-

drothermal synthesis of zeolite was conducted in a 300 cm3
stainless steel autoclave (Pressure Products Inc). The samples
were analyzed by X-ray diffraction (XRD) [Phillips APD3700
X-ray diffractometer (Cu-K0 radiation)] and infrared spec-
troscopy (IR) [Nicolet Magna 550 Series II infrared spec-
trometer equipped with a DTGS (deuterated tri-glycine sul-
fate) detector].
Synthesis * The key steps involved in the hydrothermal syn-
thesis of ZSM-5 can be seen in Figure 2 (next page): prepa-
ration of solutions containing the Si and Al precursors along
with the structure-directing agent, mixing, aging, hydrother-
mal treatment, filtration, drying, and calcination. Solution A,
containing the Si precursor and the structure-directing agent
(TPA), was prepared by dissolving 11.1 g of Aerosil silica in
32 ml of 1.25 M NaOH solution and then adding 81 ml of 3.2
wt% TPA aqueous solution. Solution B was obtained by dis-
solving 0.6 g of sodium aluminate (NaAlO2) in 10 ml of H20.
Mixing Solutions A and B under vigorous stirring resulted
in a homogeneous gel at pH 13. The pH of the gel was then
adjusted to 11 by addition of 0.1 M H2SO4. The resulting gel
was aged for 2 h prior to hydrothermal treatment. Aging is
crucial to obtain the desired crystalline phase and to acceler-
ate crystallization. The aged gel was finally loaded in a stain-
less steel autoclave and heated at 1500C for 4 h. The hydro-
thermal treatment is a thermally activated process. Increas-
ing the temperature of the reactant solution above the boiling
point facilitates the crystallization process, i.e., supersatura-

Figure 1. Structure of ZSM-5.

Winter 2004

(d). SEM of ZSM-5

. (c). ZSM-5 Structures

(a). Tetrahedron

tion, nucleation, and crystal growth. To monitor the crystallization process,
8 ml of gel reactant/ZSM-5 product mixture was sampled via a sampling 1.6 g NaOH
valve every 30 min for analysis. The samples were filtered, dried, and then 32 g H20 2.5 g TPA-Br
pressed in the form of thin disks for XFD and IR analyses. 11.1 g silica 78 g H20

RESULTS AND ANALYSIS o 0.6 g NaA02
10 gH20
The appearance and evolution of an XRD pattern characteristic of ZSM-5 0o
zeolite is shown in Figure 3. The well-defined X-ray diffraction pattern
manifests the crystalline microstructure of XSM-5.1211 The IR spectra of
ZSM-5 samples, which exhibit two key bands at 450 cm-1 and 548 cm-1, are
shown in Figure 4. The former is due to the Si-O stretching of the tetrahe- Solution A Solution B
dral unit, whereas the latter is due to the double 5-1 ring (SBU) vibration.[22]
The peak intensity in the XRD pattern and the IR intensity of the 548 cm-1 I
band reflect the extent of the crystallization process.
The crystallinity of the samples was determined by comparing the XRD
peak intensity and the intensity ratio of the 548 to 458 cm-1 IR bands with
those of calibrated samples. Crystallinity calibration was carried out by mea- I
during the intensity of a number of standard samples (i.e., mixtures of pure En
ZSM-5 from Zeolyst and Aerosil silica) of known ZSM-5 concentrations. R ta si, Al: , i
The crystallinity determined here corresponds to the zeolite yield, which Na sources I precursors
is defined as the ratio of the amount of zeolite to the initial amount of sD.
SiO2 and NaAlO2.
The experimental ZSM-5 crystallization curve, which plots the crystallin- Y-. .. ..
ity of the zeolite versus the hydrothermal treatment time, is shown in Figure ,,.''
5. The parallel between the crystallinity as measured by XRD and by IR
indicates that the relative intensity of the 458 cm-1 band can serve as a reli- amorphlis.o i
able index of the ZSM-5 crystallinity and yield during its synthesis, allow-
ing the use of a low-cost infrared spectrometer to determine the ZSM-5
structure. The zeolite crystallization curve usually exhibits an S-shaped profile Figure 2. Hydrothermal synthesis of ZSM-5.
with an inflection point, which separates the induction period and the auto-
catalytic growth period. ZSM-5 synthesis at 1500C can be completed in 5.5
h with a final crystallinity near 100%. The key parameters governing the
zeolite crystallization include hydrogel molar concentration, alkalinity Crystallization ime
Crystallization Time
(i.e., pH), temperature, template, pressure, and seeding. The complex- (hr)
ity of the interactions of these factors makes zeolite synthesis an inter-
esting laboratory project that allows each team of students to design 50 1
their own experimental parameters and carry out the experiment at a
specific set of conditions.

DISCUSSION
In a two-hour lecture, the instructor covered nanomaterial synthesis and
applications, the basic principles of zeolite synthesis, and typical character-
ization techniques such as X-ray diffraction and infrared spectroscopy as 4
well as safety issues. A graduate assistant demonstrated the operation proce- 3
dure of the autoclave and the infrared spectrometer. A list of the tasks and
time needed to complete them can be found in Table 1. 0
A typical experiment team consists of four students, and a typical synthe-
sis procedure for ZSM-5, as shown in Figure 2, is given to them. The first
homework assignment is to use SciFinder Scholar to search for a ZSM-5 0 10 20 30 40 50 60 70 80
synthesis recipe from journal articles or patents and to compare the litera- 20
ture recipe with the given one. Experience with literature searches allows Figure 3. XRD pattern of as-synthesized XSM-
students to gain a better understanding of the process of translating scien- 5 samples vs. crystallization time at 1500C.

Chemical Engineering Education

I I I I 1
1400 1200 1000 800 600 400 200

Wavenumber (cmr )

Figure 4. IR spectra of as-synthesized XSM-5
samples vs. crystallization time at 1500C.

20 30
Crystallization Time (hr)

100

80

S60

40

20

TABLE 1
Experimental Tasks
* Formation of the team and distribution of tasks
* Literature search
* Selection of a synthesis process
* Designing and planning the experiment
* Implementation
1. Preparation of precursor solution (0.5 hr)
2. Hydrothermal treatment (7 h)
3. Filtration and drying (3 h)
4. Infrared analysis (2 h)
* Report preparation
1. Kinetics of zeolite syntheses. Are you able to derive a meaningful rate
expression and obtain reaction order and rate constant for zeolite
synthesis?
2. What are the factors governing the zeolite synthesis? Discuss the phase
behavior as well as heat and mass transfer in the autoclave during
zeolite synthesis.
3. Compare the results obtained with those in the literature.
4. Propose and design a novel nanostructured material based on the
concept oftemplated synthesis and self-assembly.
* Peer review:
1. Task distribution
2. Time management
3. Coordination
4. Quality of work
5. Objective accomplishment

Figure 5. Crystallizations (%) vs crystallinity time.

Winter 2004

tific discovery to practical technology. Students are strongly encouraged
to either modify the given recipe or use a literature or patented recipe to
design and implement their experiment for zeolite synthesis. Extra bonus
points are given to teams that use literature recipes. The team that chooses
a literature recipe must submit its recipe to the instructor to ensure the
safety and availability of required chemicals. Of the seven teams in the
2000 class, only two opted for a literature recipe.
Experimental planning involves selection of hydrothermal synthesis
conditions and assignment of tasks to each student on the team. Each stu-
dent is responsible for a specific task. Students who are not involved in a
specific assignment are required to observe and understand their team-
mates' tasks. The total time needed for the experiments is 12.5 h.
To help students increase their understanding of the interrelationship
between reaction engineering and nanomaterials synthesis, as well as to
promote their ability to link experimental observations to fundamental
concepts, we posed several questions, which can be found in Table 1.
These questions provide a framework for students to prepare their
reports and for the instructor to evaluate the students' understanding
and creativity.
Our 1999 learning survey revealed that the typical problem encoun-
tered in zeolite synthesis was plugging of the sampling valve. Peer review
and close supervision of the students' performances revealed that the ma-
jority (90%) of the students accomplished the assigned tasks. Peer review
also pointed out the problems students encountered in coordinating the
experimental work. The final grade of individual students was obtained
by adjusting the team report grade based on each student's contribution
Continued on page 47.

S9,1 laboratory

THE FUEL CELL

An Ideal ChE Undergraduate Experiment

JUNG-CHOU LIN, H. RUSSELL KUNZ, JAMES M. FENTON, SUZANNE S. FENTON
University of Connecticut * Storrs, CT 06269

here is much interest in developing fuel cells
for commercial applications. This interest is
driven by technical and environmental advan-
tages offered by the fuel cell, including high perfor-
mance characteristics, reliability, durability, and clean
power. A fuel cell is similar to a battery-it uses an
electrochemical process to directly convert chemical
energy to electricity. Unlike a battery, however, a fuel
cell does not run down as long as the fuel is provided.
Fuel cells are characterized by their electrolytes since
the electrolyte dictates key operating factors such as
operating temperature. The main features of five types
of fuel cells are summarized in Table 1.i11
The proton exchange membrane (PEM) fuel cell is
particularly amenable for use as an undergraduate
laboratory experiment due to safety and operational
advantages, including use of a solid polymer electro-
lyte that reduces corrosion, a low operating tempera-
ture that allows quick startup, zero toxic emissions,
and fairly good performance compared to other fuel
cells. A cross-sectional diagram of a single-cell PEM
fuel cell is shown in Figure 1. The proton exchange
membrane (Nafion�) is in contact with the anode
catalyst layer (shown on the left) and a cathode
catalyst layer (shown on the right). Each catalyst
layer is in contact with a gas diffusion layer. The
membrane, catalyst layers, and the gas diffusion
layers make up what is called the membrane-elec-
trode-assembly (MEA).
Fuel (hydrogen in this figure) is fed into the anode
side of the fuel cell. Oxidant (oxygen, either in air or
as a pure gas) enters the fuel cell through the cathode
side. Hydrogen and oxygen are fed through flow chan-
nels and diffuse through gas diffusion layers to the

TABLE 1
Summary of Fuel Cell Technologies

Fuel Cell
Alkaline (AFC)

Electrolyte
Potassium Hydroxide

Temperature
_fC Applications
90-100 Military
Space Flight

Phosphoric Acid Phosphoric Acid 175-200 Electric Utility
(PAFC) Transportation
Molten Carbonate Lithium, Sodium, and/or 650 Electric Utility
(MCFC) Potassium Carbonate
Solid Oxide Zirconium Oxide 1000 Electric Utility
(SOFC) Doped by Yttrium
Proton Exchange Membrane Solid polymer <100 Electric Utility

(PEMFC)

(poly-perfluorosulfonic acid)

Portable Power
Transportation

� Copyright ChE Division ofASEE 2004

Chemical Engineering Education

Jung-Chou Lin eamed his PhD from the University of Connecticut and his BS from the
Tunghai University, Taiwan, both in chemical engineering. After graduation he was
employed as an Assistant Professor in Residence to develop fuel cell experiments for
the undergraduate laboratory at the University of Connecticut. Currently, he is a senior
Research Engineer at Microcell Corporation in Raleigh, North Carolina.
H. Russell Kunz is Professor-in-Residence in the Chemical Engineering Department
at the University of Connecticut and Director of Fuel Cell Laboratories at the University
of Connecticut. An internationally recognized expert in fuel cell development, Dr. Kunz
was educated at Rensselaer Polytechnic Institute, receiving his BS and MS degrees in
Mechanical Engineering and his PhD in Heat Transfer
James M. Fenton is Professor of Chemical Engineering at the University of Connecti-
cut. He teaches transport phenomena and senior unit operations laboratory courses.
He earned his PhD from the University of Illinois and his BS from the University of
California, Los Angeles, both in Chemical Engineering. His research interests are in the
areas of electrochemical engineering and fuel cells.
Suzanne S. Fenton is the Assistant Department Head and Visiting Assistant Professor
of Chemical Engineering at the University of Connecticut. She received her BS degree
in Environmental Engineering from Northwestern University and her PhD in Chemical
Engineering from the University of Illinois. She teaches transport phenomena and se-
nior unit operations laboratory courses and provides innovative instruction for second-
ary school students.

catalyst on their respective sides of the MEA. Activated by the catalyst in the
anode, hydrogen is oxidized to form protons and electrons. The protons move
through the proton exchange membrane and the electrons travel from the anode
through an external circuit to the cathode. At the cathode catalyst, oxygen re-
acts with the protons that move through the membrane and the electrons that
travel through the circuit to form water and heat.
Since the hydrogen and oxygen react to produce electricity directly rather
than indirectly as in a combustion engine, the fuel cell is not limited by the
Carnot efficiency. Although more efficient than combustion engines, the fuel
cell does produce waste heat. The typical efficiency for a Nafion PEM fuel cell
is approximately 50%.

Figure 1. PEM fuel cell cross section.

C-

10.

- 1.0

of
'olarization "S
ss Dominates) .

0.5

0 10.0
1200 1400

0 200 400 600 800 1000
Current Density (mA/cm2)

Figure 2. Representative fuel cell performance curve at 25 �C, 1 atm.

Fuel cells can be used to demonstrate a
wide range of chemical engineering prin-
ciples such as kinetics, thermodynamics, and
transport phenomena. A general review of
PEM fuel cell technology and basic electro-
chemical engineering principles can be found
in the literature.1 -81 Because of their increas-
ing viability as environmentally friendly en-
ergy sources and high chemical engineering
content, fuel cell experiments have been de-
veloped for the chemical engineering under-
graduate laboratory as described in the re-
mainder of this paper.

OBJECTIVES
The objectives of the fuel cell experiment
are
* To familiarize students with the
working principles and performance
characteristics of the PEM fuel cell
* To demonstrate the effect of oxygen
concentration and temperature on fuel
cell performance
* Tofit experimental data to a simple
empirical model
Students will measure voltage and mem-
brane internal resistance as a function of op-
erating current at various oxygen concentra-
tions and temperatures; generate current den-
sity vs. voltage performance curves; and cal-
culate cell efficiency, reactant utilization, and
power density. Current density is defined as
the current produced by the cell divided by
the active area of the MEA. By fitting cur-
rent density vs. voltage data to a simple em-
pirical model, students can estimate ohmic,
activation (kinetic), and concentration (trans-
port) polarization losses and compare them
to experimental or theoretical values.

BACKGROUND
The performance of a fuel cell can be char-
acterized by its
1. Current density versus voltage plot as
shown in Figure 2
2. Ettic ienc v
3. Reactant utilization (ratio of moles of
fuel consumed to moles of fuel fed)
4. Power density (ratio of power produced
by a single cell to the area of the cell
(MEA)

Winter 2004

Anode: H,- 2 H + 2 e-
Cathode: V2 0, + 2 e- + 2 lH -> H20

Overall: H2 + 02 , - H20

Current Density- Voltage Characteristics
Since a fuel cell is a device that facilitates the direct con-
version of chemical energy to electricity, the ideal or best-
attainable performance of a fuel cell is dictated only by the
thermodynamics of the electrochemical reactions that occur
(a function of the reactants and products). The electrochemi-
cal reactions in a hydrogen/oxygen fuel cell are shown in
Eqs. (1) and (2).
Anode Reaction

H2 -> 2H + 2e (1)
Cathode Reaction

1-O2+2H +2e- -> H20 (2)
2
The reversible standard (i. e., ideal) potential E� for the H2/02
cell reaction is 1.23 volts per mole of hydrogen (at 25 �C,
unit activity for the species, liquid water product) as deter-
mined by the change in Gibbs free energy. Reference 1 pro-
vides a derivation of this potential. The reversible standard
potential for the hydrogen/oxygen cell is indicated on the
current density-voltage diagram in Figure 2 as the horizontal
line drawn at a voltage of 1.23. The Nernst equation can be
used to calculate reversible potential at "non-standard" con-
centrations and a given temperature. Equation (3) is the Nernst
equation specifically written for the H2/02 cell based on the
reactions as written.

E = O+ RT n PH)(Po)2 (3)
S nF (PH,20)

where
R gas constant (8.314 Joule/mol �K)
T temperature (�K)
F Faraday's constant (96,485 coulombs/equiv)
n moles of electrons produced/mole of H2 reacted
(n=2 for this reaction)
Eo reversible potential at standard concentrations and
temperature T (volts)
E reversible potential at non-standard concentrations
and temperature T (volts)
PH2'2,PH2 partial pressures of H2, 02, and H20, respectively
(atmn)
Note: 1 volt = 1 joule/coulomb

The Nernst equation cannot be used to make both tempera-
ture and concentration corrections simultaneously. To do this,
one must first apply Eq. (4) to "adjust" the standard potential
E� for temperature and then apply the Nernst equation to
adjust for concentration at the new temperature.[6]

-o -0 o AS (T 1 (4)
r -n2 o E-T nF t (4)

Subscripts 1 and 2 on E� denote "at temperatures T, and T2"

and AS is the entropy change of reaction (= - 163.2 J/0K for
the H2/02 reaction at 25 �C, unit activity for the species, liq-
uid water product).
When a load (external resistance) is applied to the cell, non-
equilibrium exists and a current flows. The total current passed
or produced by the cell in a given amount of time is directly
proportional to the amount of products formed (or reactants
consumed) as expressed by Faraday's law

mnF
IMt
sMt

where I (A) is the current, m (g) is the mass of product formed
(or reactant consumed), n and F are defined above, s is the
stoichiometric coefficient of either the product (a positive
value) or reactant (a negative value) species, M (g/mol) is
the atomic or molecular mass of the product (or reactant )
species, and t (s) is the time elapsed. Equation (5) is valid for
a constant current process. Faraday's law can be written in
the form of the kinetic rate expression for H2/02 cell as

I d(moles H20) -d(moles H2) -2d(moles 02) (6)
2F dt dt dt (6)
There is a trade-off between current and voltage at
nonequilibrium (nonideal) conditions. The current density-
voltage relationship for a given fuel cell (geometry, catalyst/
electrode characteristics, and electrolyte/membrane proper-
ties) and operating conditions (concentration, flow rate, pres-
sure, temperature, and relative humidity) is a function of ki-
netic, ohmic, and mass transfer resistances. The current den-
sity vs. voltage curve shown in Figure 2 is referred to as the
polarization curve. Deviations between the reversible po-
tential and the polarization curve provide a measure of
fuel cell efficiency.
Kinetic Limitations * Performance loss (voltage loss) re-
sulting from slow reaction kinetics at either/both the cathode
and anode surfaces is called activation polarization (,act,o and
,act,,). Activation polarization is related to the activation en-
ergy barrier between reacting species and is primarily a func-
tion of temperature, pressure, concentration, and electrode
properties. Competing reactions can also play a role in acti-
vation polarization.
Kinetic resistance dominates the low current density por-
tion of the polarization curve, where deviations from equi-
librium are small. At these conditions, reactants are plentiful
(no mass transfer limitations) and the current density is so
small that ohmic (= current density x resistance) losses are
negligible. The Tafel equation describes the current density-
voltage polarization curve in this region.
lact = B log|i| - A (7)
where qact is the voltage loss due to activation polarization
(mV), i is current density (mA/cm2), and constants A and B
are kinetic parameters (B is often called the Tafel slope).[6]

Chemical Engineering Education

As shown in Figure 2, the kinetic loss at the cathode, iot,o
(the reduction of 02 to form water) is much greater than ki-
netic loss at the anode, a.a, in the H2/02 cell.

Ohmic Limitations * Performance loss due to resistance to
the flow of current in the electrolyte and through the elec-
trodes is called ohmic polarization (ohm). Ohmic polariza-
tion is described using Ohm's law (V=iR), where i is current
density (mA/cm2) and R is resistance (fl-cm2). These losses
dominate the linear portion of the current density-voltage
polarization curve as shown in Figure 2. Improving the ionic
conductivity of the solid electrolyte separating the two elec-
trodes can reduce ohmic losses.

Transport Limitations * Concentration polarization (i1 .
and T )1 ) occurs when a reactant is consumed on the surface
of the electrode forming a concentration gradient between
the bulk gas and the surface. Transport mechanisms within
the gas diffusion layer and electrode structure include the
convection/diffusion and/or migration of reactants and prod-
ucts (H2, 02, H ions, and water) into and out of catalyst sites
in the anode and cathode. Transport of H ions through the
electrolyte is regarded as ohmic resistance mentioned above.
Concentration polarization is affected primarily by concen-
tration and flow rate of the reactants fed to their respective
electrodes, the cell temperature, and the structure of the gas
diffusion and catalyst layers.
The mass-transfer-limiting region of the current-voltage
polarization curve is apparent at very high current density.
Here, increasing current density results in a depletion of re-
actant immediately adjacent to the electrode. When the cur-
rent is increased to a point where the concentration at the
surface falls to zero, a further increase in current is impos-
sible. The current density corresponding to zero surface con-
centration is called the limiting current density (iQ), and is
observed in Figure 2 at approximately 1200 mA/cm2 as the
polarization curve becomes vertical at high current density.
The actual cell voltage (V) at any given current density can
be represented as the reversible potential minus the activa-
tion, ohmic, and concentration losses, as expressed in Eq.
(8).
V = E - (act,c + lact, a) - iR - (lconcc + .conca) (8)
Note that activation (aT,,, , acta) and concentration (1.con.,c
.on, a) losses (all positive values in Eq. 8) occur at both elec-
trodes, but anode losses are generally much smaller than cath-
ode losses for the H2/02 cell and are neglected. Ohmic losses
(iR) occur mainly in the solid electrolyte membrane. An ad-
ditional small loss will occur due to the reduction in oxygen
pressure as the current density increases. Current fuel cell
research is focused on reducing kinetic, ohmic, and transport
polarization losses.

Cell Efficiency
Fuel cell efficiency can be defined several ways. In an en-
ergy-producing process such as a fuel cell, current. ii ..... n ,. r'
is defined as

theoretical amount of reactant
required to produce a given current
actual amount of reactant consumed

In typical fuel cell operation, current efficiency is 100% be-
cause there are no competing reactions or fuel loss. Voltage
efficiency is

actual cell voltage V (10)
reversible potential E

The actual cell voltage at any given current density is repre-
sented by Eq. (8) and reversible potential by Eq. (3). Overall
energy etti iencv is defined as

Ce = f*Cv (11)

The H2/02 fuel cell of Figure 2 operating at 0.8 V has a volt-
age efficiency of about 65% (=0.8/1.23*100). The overall
efficiency at this voltage, assuming that the current efficiency
is 100%, is also 65%. In other words, 65% of the maximum
useful energy is being delivered as electricity and the remain-
ing energy is released as heat (35%).
A fuel cell can be operated at any current density up to the
limiting current density. Higher overall efficiency can be ob-
tained by operating the cell at a low current density. Low
current density operation requires a larger active cell area to
obtain the requisite amount of power, however. In designing
a fuel cell, capital costs and operating cost must be optimized
based on knowledge of the fuel cell's performance and in-
tended application.

Reactant Utilization
Reactant utilization and gas composition have major im-
pact on fuel cell efficiency. Reactant utilization is defined as

U Molar J1. . i.., ,, , - Molar flowratereactant, out
U-
M olar 1 '.1 . .. , , i

Mol H2 / s consumed (12)
Mol H2 / s fed

"Molar flow rate consumed" in this equation is directly pro-
portional to the current produced by the cell and can be cal-
culated from Eq. (6). In typical fuel cell operation, reac-
tants are fed in excess of the amount required as calcu-
lated by Faraday's law (i.e., reactant utilization < 1).
Higher partial pressures of fuel and oxidant gases gener-
ate a higher reversible potential and affect kinetic and
transport polarization losses.

Winter 2004

Power Density

The power density delivered by a fuel cell is the product of
the current density and the cell voltage at that current den-
sity. Because the size of the fuel cell is very important, other
terms are also used to describe fuel cell performance. Spe-
cific power is defined as the ratio of the power generated by
a cell (or stack) to the mass of that cell (or stack).

EQUIPMENT, PROCEDURE,
AND IMPLEMENTATION
The experiments presented here are designed to give the
experimenter a "feel" for fuel cell operation and to demon-
strate temperature and concentration effects on fuel cell per-
formance. The manipulated variables are cell temperature,
concentration of oxygen fed to the cathode, and current. Flow
rates are held constant and all experiments are performed at
1 atm pressure. The measured variables are voltage and re-
sistance, from which polarization curves are generated and
fuel cell performance is evaluated. A simple empirical
model can be fit to the data, allowing students to sepa-
rately estimate ohmic resistance, kinetic parameters, and
limiting current density. Table 2 summarizes the condi-
tions investigated in this study.
Many other experimental options are available with the
system described in this paper, including an investigation of
the effect of 1) catalyst poisoning, 2) relative humidity of the
feed gases, or 3) flow rate on fuel cell performance.

Equipment

A schematic diagram of the
experimental setup is shown Eq
in Figure 3. An equipment list
for in-house-built systems, in- Quant. Equipment/Supplies
cluding approximate cost and 1 Fuel cell load (sink and p
the names of several suppli- 1 Computer (optional)
ers, is provided in Table 3. 1 Data acquisition card (op
Completely assembled sys- 1 Single cell hardware w/h
teams can be purchased from 1 Membrane-electrode-asse
Scribner Associates, Inc. 5 Temperature controller: 0
(www. scr ib ne r. co m), 4 Heating element (heating
Lynntech Inc. (www.lynn- 5 Thermocouple
tech.com), ElectroChem Inc. 2 Humidifier (2" ID stainle
(www.fuelcell.com), and 2 Rotameter (0-200 cc/min for
TVN (www.tvnsystems.com). N/A Valves and fittings (stainl
Hydrogen, supplied from a 20 ft Tubing (1/4" stainless ste
pressurized cylinder, is sent 4 Regulator
through the heated anode hu- N/A Gas (H2, N2, Air, ('
midifier before being fed 1 Digital flow meter (for ca
through heated tubes to the Other
anode side of the fuel cell. TOTAL
Similarly, oxidant with any
desired composition (oxygen *List is not exhaustive

TABLE 2
Experimental Conditions: All at P=- atm

Anode Feed Cathode Feed
Dry basis Dry basis
Temp Flowrate Composition Temp Flow rate Composition
(C) (m/min) (Mole %) (C) (ml/min) (Mole %)

80 98 100% H2 80 376 100% 02
80 98 100% H2 80 376 Air-21% 02 inN2
80 98 100% H2 80 376 10.5% 02 in N2
80 98 100% H2 80 376 5.25% 02 in N2
18 98 100% H2 18 376 100% 02

Effluent

PEM Fuel Cell ,. Switch Valve

Fuel Cell Load <--> @ Rotameter
____I___Humidifier

Effluent N H2

Figure 3. Schematic of experwitmentch Valve
,, Rotameter
Humidifier

Figure 3. Schematic of experimental setup.

TABLE 3
uipment List for In-House-Built Systems

Approx. Cost Vendor*

ower supply)

tional)
eating element (5 cm2)
embly (5 cm2)
-100�C
tape)

ss pipes and caps)
H, fuel; 0-400 cc/min for oxidant)
ess steel)
el)

liberation of rotameter)

$2,000
$1,000
$1,000
$1,500
$200
$1,000
$400
$200
$200
$400
$1,500
$200
$1,000
$1,000
$500
$1,000

-$13,000

Scribner, Lynntech, Electrochem, TVN
Dell, IBM, Compaq
National Instruments
Electrochem, Fuel Cell Technology
Electrochem, Lynntech, Gore Associates
OMEGA
OMEGA
OMEGA
McMaster-Carr
OMEGA
Swagelok

Airgas
Airgas
Humonics

Chemical Engineering Education

in nitrogen) is supplied from a pressurized cylinder and sent
to the heated cathode humidifier before being fed through
heated tubes to the cathode side of the fuel cell. Constant
volumetric flow rates for anode and cathode feeds are manu-
ally controlled by rotameters. Humidification of the feed
streams is necessary to maintain conductivity of the electro-
lyte membrane. Heating of the humidifiers, the tubes leading
to the fuel cell, and preheating of the fuel cell is accomplished
using heating tape, and temperatures of the feed streams and
fuel cell are maintained using temperature controllers. To
avoid flooding the catalyst structure, the humidifier tempera-
ture is maintained at or slightly below the cell temperature.
The relative humidity of a stream exiting a humidifier can be
determined manually by flowing the stream across a tem-
perature controlled, polished metal surface and measuring its
dew point. Effluent from the fuel cell is vented to a hood for
safety purposes.
The PEM fuel cell comprises an MEA with an active area
of 5 cm2 (prepared at the University of Connecticut) and is
housed in single-cell hardware with a single-pass serpentine
flow channel. Our fuel cell load and data acquisition elec-
tronics are integrated in a single unit manufactured by Scribner
Associates. During a typical experimental run (constant flow
rate, oxidant composition, and temperature), the current is
manipulated/adjusted on the fuel cell load and the voltage
and resistance are read from built-in meters in the load. The
fuel cell load uses the "current-interrupt technique"131 to mea-
sure the total resistance between the two electrodes.

Procedure
A fuel cell with a prepared or commercial MEA is first
connected to the fuel cell test system. Before feeding the hy-
drogen and oxidant into the fuel cell, humidified nitrogen is

TABLE 4
Sample Flow-Rate Calculation

m Is
Faraday's Law: - = -mol / time
Mt nF
Hydrogen consumption in fuel cell = I/(2F) mol/time
Oxygen consumption in fuel cell = I/(4F) mol/time
To produce a current of I 1 Amp, H2 consumption is:
- I/(2F) - 1/(2 x 96485) - 5.18 * 10-6 mol/s
-3.11 * 10' mol/min
According to gas law: PV - NRT
At 80C and 1 atm, V/N - RT/P - 0.082*(273.15 + 80)/1 - 29 L/mol
So H2 consumption is: VH2 9.0 ml/min @ 1 Amp current
02 consumption is: V02 4.5 ml/min @ 1 Amp current
Corresponding V - 4.5/0.21 - 21.4 ml.min @ 1 Amp current
To convert the above numbers to vol flowrates at a desired current
density (amp/cm2), divide ml/min by 1 cm2 to get ml/min/cm2.
For desired 45% H2 utilization at 1 Amp/cm2 current density
U - moles consumed/moles fed - 0.45
H2 feed flow rate is: VH2 - 9.0/0.45 - 20 ml/min/cm2 @ 1 Amp/cm2
100 ml/min @ 1 A current with 5 cm2 MEA

introduced to purge the anode and cathode sides of the single
cell. During the purge (at 50 cc/min), the cell and humidifi-
ers are heated to their respective operating temperatures (e.g.,
cell, 80 �C, humidifiers, 80 �C). When the cell and humidifi-
ers reach the desired temperature, the humidified nitrogen is
replaced by humidified hydrogen and oxidant for the anode
and cathode, respectively. During experiments, fuel and oxi-
dant are always fed in excess of the amount required to pro-
duce a current of 1 A as calculated by Faraday's law (Eq.
5). The hydrogen and oxidant flow rates used in these ex-
periments are based on operating at 1 A/cm2 with an ap-
proximate reactant utilization of 45% for the hydrogen
and 30% for oxidant (based on air). A sample calculation
is provided in Table 4.
After introducing the fuel and oxidant into the cell, the open
circuit voltage (zero current) should be between 0.8 and 1
volt. Fuel cell performance curves are generated by record-
ing steady state voltage at different currents. Approximately
5 minutes is required to reach steady state for changes in cur-
rent at constant composition and temperature, but it might
take 20 to 30 minutes to reach steady state for a change in
either oxidant composition or temperature. The system should
be purged with nitrogen during shutdown. Short-circuiting
the fuel cell will destroy the MEA.

Implementation and Assessment
This experiment will be included as part of a three-credit
senior-level chemical engineering undergraduate laboratory.
The course consists of two 4-hour labs per week, during which
groups of 3 to 4 students perform experiments on five differ-
ent unit operations throughout the semester (e.g., distillation,
heat exchanger, gas absorption, batch reactor, etc.). Each unit
is studied for either one or two weeks, depending on the com-
plexity and scale of the equipment. Given only general goals
for each experiment, students are required to define their own
objectives, develop an experimental plan, prepare a pre-lab
report (including a discussion of safety), perform the experi-
ments, analyze the data, and prepare group or individual writ-
ten and/or oral reports.
The fuel cell experiment described above can easily be com-
pleted in one week (two 4-hour lab periods). Additional ex-
periments can be added to convert this lab into a two-week
experiment. Due to their similar nature and focus (genera-
tion of performance/characteristic curves and analysis of
efficiency at various operating conditions), the fuel cell
experiment could be used in place of the existing cen-
trifugal pump experiment.
Immediate assessment of the experiment will be based on
student feedback and student performance on the pre-lab pre-
sentation, lab execution, and technical content of the written/
oral reports. Existing assessment tools (End-of-Course Sur-
vey, Senior Exit Interview, Alumni Survey, Industrial Advi-

Winter 2004

sory Board input, and annual faculty curriculum review) will
be used to evaluate the overall impact of the experiment.

RESULTS AND DISCUSSION

Performance

Performance curves (voltage vs. current density) and mem-
brane resistance vs. current density at 80 �C with different
oxidant compositions (pure oxygen, air, 10.5% 02 in N2 and
5.25% 02 in N2) are shown in Figure 4. Measured open cir-
cuit voltage (Voo) canbe compared to reversible potential cal-
culated via Eqs. (3) and (4). These values are presented in the
legend of Figure 4. Students will observe that the actual open
circuit voltage is slightly lower than the theoretical maxi-
mum potential of the reactions. Activation polarization (ki-
netic limitation) is observed at very low current density (0-
150 mA/cm2). Kinetic losses increase with a decrease in oxy-
gen concentration. At low current densities, membrane resis-
tance ohmicc polarization) is nearly constant (about 0.14 f1-
cm2) and is independent of oxidant composition. Membrane
resistance begins to increase slightly with increasing current
density at 800 mA/cm2 due to dry-out of the membrane on
the anode side. Dry-out occurs at high current density be-
cause water molecules associated with migrating protons are
carried from the anode side to the cathode at a higher rate
than they can diffuse back to the anode. Mass transport limi-
tations due to insufficient supply of oxygen to the surface of
the catalyst at high current density is observed, especially for
gases containing low concentrations of oxygen. Limiting cur-
rents are evident at about 340 mA/cm2 and 680 mA/cm2 for
the 5.25% and 10.5% oxygen gases, respectively, but are not
obvious for pure oxygen and air. Limiting current density
can be shown to be directly proportional to oxygen content.

The effect of operating temperature (180C vs. 800C, both
at 100% relative humidity) on cell performance and mem-
brane resistance for a pure 02/H2 cell is shown in Figure 5.
Measured open-circuit voltage and reversible potential at 80�C
are slightly lower than the corresponding voltages at 180C.
This is due to higher concentrations of reactants when fed at
lower temperatures and 100% relative humidity. Elevated
temperatures favor faster kinetics on the catalyst surface and
lower membrane resistance, however, resulting in better cell
performance. Under fully hydrated environments (100% RH),
membrane resistance decreases with increasing temperature
due to increased mobility of the protons. Again, limiting cur-
rent density for pure oxygen is not obvious in this plot.

A linear relationship between current density and reactant
utilization (per Eq. 5) is clearly evident in Figure 6. Reactant
utilization decreases with increasing inlet oxygen concen-
tration (at constant flow rate) because of an increase in
the moles reactant feed.

Power density (W/cm2) delivered by a fuel cell is defined
by the product of current density drawn and voltage at that

current density. The effect of current density on power den-
sity for various oxidant compositions is shown in Figure 7.
For a given feed composition, maximum power density is
achieved approximately halfway between no-load and limit-
ing current densities. The selection of "optimal" operating

I.O 1.0
S-0- OPerformancewithOxygen (V .01 V,E=. 17V)
0.9 -- Resistance with Oxygen 0.9
-- Performance with Air (Voc0.96 V, E=1.14 V)
0 - Resistancc with Air 0S8
S-A-erfo..mance with 1.5% 02 in N2 (Voc=0.94V, E=1.13 V)
Resistan.- with 10.5 % 0,2 nNi N
SPer formance with 5.25 % 02 in N2 (Voc=0.92 V, E=1.12 V)
^0.Rsista e.. with 5.25 % 02 in N2 0.6
0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 T0.1
0.0 - 0.0
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Current Density (mA/cm2)
Figure 4. Effect of oxidant concentration on cell perfor-
mance and membrane resistance at 800C, 1 atm.

1.1 1.1
o1.01 - er-formance with Oxygen at 80 �C (V =1.01 V, E1.17 V) 1.0
0.9 Resistance with Oxygen at 80 C 0.9
SPerformance with Oxygen at 18C (Voc=1.04 V, E=1.24V) 0 '
0.8 ~0.8
El Resistance with Oxygen at8 C
,0.7 0.7
0.6 0.6
S0.5 0.5
> 0.4 0.4
0.3 0.3
0.2 )0.2
0.1 0.1
0.0 0.0
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Current Density (mA/cm2)

Figure 5. Effect of temperature on cell performance and
membrane resistance at 1 atm, pure 02'

0 400 800 1200 1600 2000
Current Density (mA/cm2)

Figure 6. Effect of current density and oxidant composi-
tion on reactant utilization at 800�C, 1 atm.

Chemical Engineering Education

conditions depends on how the fuel cell is to be used. For
example, for vehicular applications, higher power density is
required to minimize the weight of the car at the expense of
efficiency. For residential (non-mobile) applications, a cell
with higher efficiency would be preferred.

2000

800 1200
Current Density (mA/cm2)

Figure 7. Effect of current density and oxidant
composition on power density at 800C, 1 atm.

900
800
700
600
500
3 400

0 400 800 1200 1600
Current Density (mA/cm2)

2000 2400

Figure 8. Nonlinear regression fit of experimental
data at 800C, 1 atm.

TABLE 5
Best-Fit Values for Kinetic Parameters, Ohmic I
Transport Parameters Obtained Using Eq.
Compared to Values Calculated or Measured by (
Eq. (14):V - E + A - (B log(i)) - iR - w exp(zi)
Eq. (7):q. - B log|i| - A

Oxidant Temp E+A Bfq.14 BttoEq. RftWq.4 R.ad
Comp (C) (ml) (mV/dec) (mV/dec) (-cnd) (l-cnd)

Oxygen
Air
10.5% 02 in N
5.25% 02 in N2

Winter 2004

Empirical Model
Although comprehensive modeling of a fuel cell system is
beyond the scope of an undergraduate lab, a simple model
describing voltage-current characteristics of the fuel cell can
be introduced to the students and tested for 1) its ability to fit
the data, and 2) its usefulness as an analytical tool. The fol-
lowing empirical model describing the loss of cell voltage
due to kinetic, ohmic, and transport limitations was proposed
by Srinivasan, et al.: [9]

V = E- (B log (i)- A) -iR- w exp(zi) (13)

where E, B, A, R, w, and z are "fit" parameters. Lumping E
and A together gives

V = E+A -(B log (i)) -iR- w exp(zi) (14)

Equation (14) is modeled after Eq. (8) assuming the anode
polarization terms in Eq. (8) are negligible, that the kinetic
limitations of the cathode can be described by the Tafel Eq.
(7), and that mass transport losses can be fit using the param-
eters w and z. The purely empirical term, w exp(zi), in Eq.
(14) can be replaced with a more physically meaningful term

where ihm (mA/cm2) is the current density corresponding to a
zero surface concentration, and C (mV/decade) is a param-
eter related to the Tafel slope. Due to space limitations, how-
ever, the physical meanings and the accurate estimation of C
and ihm will be explained in a forthcoming publication.[10]
The model fit to experimental data using nonlinear regres-
sion software (Polymath) is shown in Figure 8. All curves
generated using this model have correlation coefficients in
excess of 0.999. The model therefore is excellent as a fitting
function for fuel cell performance curves from which values
canbe interpolated or extrapolated. This is particularly handy
for estimating limiting current
density in cases where the data
is insufficient.
Losses, and Values for the adjustable param-
. (14) eters [(E+A), B, R, w, z] calculated
theirr Means by the regression software are sum-
marized in Table 5. The "regres-
sion generated" values for R can
Correlation be compared to experimentally
Correlation
w Z Coefficient measured values (shown on the
(m) (cnm/mA) (RA2) right-hand scale of Figures 4 and
4.202 0.0020 0.999 5) and "regression generated" val-
0.018 0.0074 0.999 ues for B canbe compared to those
0.035 0.0133 0.999 predicted using theory. In this way
0.008 0.0297 0.999 the model can be tested for its
"analytical" capability.

* oxygen
-..- modeled line for oxygen
L air
k modeled line for air
[,A 10.5 % 0,2 in N,
A . ------ modeled line for 10.5 % 02 in N2
S�* 5.2 5% iO21nN 2
"\ _ --- modeled line for 5.25% 02 in N2

t U
5.2 %- 02 nN

moee \m fo -2-.02noee in o .5%0 nN

79 85 0.20
77 84 0.29
87 94 0.33
88 95 0.51

0.14-0.16
0.14 - 0.16
0.14-0.16
0.14-0.16

Contrary to experimental results, resistance calculated us-
ing Eq. (14) increases with decreasing oxygen concentration
and is 40%-200% higher than measured membrane resistance
(0.14 - 0.16 fl-cm2 measured by the current-interrupt tech-
nique). This suggests that R from Eq. (14) includes voltage
losses other than the ohmic resistance of the membrane and
that the model is not reliable in predicting true physical be-
havior of individual contributions to the polarization curve.
For instance, "model R" is assumed to be constant over the
entire range of current densities, but in actual fuel cell opera-
tion, R is a function of current density at high current density.
Theoretical Tafel slope, B, is equal to 2.303 RT/&aF where
R is the ideal gas constant, T is absolute temperature, F is
Faraday's constant, and a is a lumped kinetic parameter equal
to 1 for the oxygen reduction reaction occurring on the cath-
ode.6 According to this theory, the Tafel slope should be about
70 mV/decade at 800C. Table 5 shows the regression gener-
ated B is 20-36% higher than the value of 70 mV/decade.
Again, one might suggest some physical reasons for this dis-
crepancy, such as the existence of diffusion or resistive
losses in the cathode catalyst layer of the electrode. We
may argue, however, that the model is too "flexible" to
assign any physical significance to the values of the "fit"
parameters (i.e., a huge range of values for each param-
eter will yield a good fit).
Tafel slopes are more accurately obtained from raw data
using the Tafel equation, Eq. (7). In this case, B canbe found
by plotting iR-free voltage (V + iR) vs. log i (see Figure 9)
and measuring the slope of the line in the kinetically con-
trolled portion of the plot (at low values of log i). Values for
B found by using this technique have been included in Table
5. While those values found from Eq. (7) are more accu-
rate than those from Eq. (14), they still differ from the
theoretical value of 70.
The Tafel slope should not be a function of the oxygen
concentration at low current density, so the lines in Figure 9
should all be parallel. It is clear that mass transport does not
interfere with the calculation for the oxygen performance
(straight line over the full decade of 10 to 100 mA/cm2). The
5.25% oxygen curve, however, is linear only for two points,
10 and 20 mA/cm2, as mass transport resistances occur at
lower current densities.
The parameters w and z are intended to describe mass trans-
port limitations, but actually have no physical basis. One might
expect these parameters to be dependent on flow charac-
teristics in the cell that were not investigated in this study.
Therefore, the predictive or analytical usefulness of w and
z cannot be evaluated.

CONCLUSIONS
Fuel-cell based experiments embody principles in electro-
chemistry, thermodynamics, kinetics, and transport, and are

well suited for the chemical engineering curricula. Students
are given an opportunity to familiarize themselves with fuel
cell operation and performance characteristics by obtaining
voltage-versus-current-density data for the unit atvarying oxi-
dant compositions and temperatures.
A simple model can be used as a fitting function for inter-
polation and extrapolation purposes. Model sensitivity analy-
sis can be performed to evaluate its usefulness as an analyti-
cal tool. The lab can be completed easily in two 4-hour lab
periods. The experiment is also suitable for use as a demon-
stration in a typical lecture course or as a hands-on project
for high school students and teachers. The experimental sys-
tem is described, including cost and vendor information.

NOMENCLATURE
A kinetic parameter used in Eqs. (7), (13), and (14)
(mV)
B Tafel slope (mV/decade)
C parameter related to the Tafel slope (mV/decade)
E reversible potential at nonstandard concentration
at temperature T (V or mV)
E� reversible potential at standard concentration at
temperature T (V or mV)
F Faraday's constant = 96,485 (coulombs/equiva-
lent)

I
i
iim
M
m
n

N
PH2' 02 H20
R

current (A)
current density (mA/cm2)
limiting current density (mA/cm2)
molecular weight (g/mol)
mass of product formed or reactant consumed (g)
moles of electrons participating in the reaction per
mole of reactant (equiv/mol)
moles
partial pressures (atm)
electrical resistance (fl-cm2)

Figure 9. Tafel slope estimation using IR-free voltage plot
of experimental data at 800C, 1 atm.

Chemical Engineering Education

1.0
0.9
0.8
0 .7
UL 0.6
o 0.5
0.4
0.3
0.2
0.1
0.0

10 100 1000
Current Density (mA/cm2)

R universal gas constant = 8.31 (J/mol-K)
s Stoichiometric coefficient of the product (positive
value) or reactant (negative value) species
AS entropy change of reaction (J/K)
T temperature (K)
t time (s)
U reactant utilization (moles consumed/moles fed)
V voltage (V or mV)
w mass transport parameter used in Eqs. (13) and
(14) (mV)
z mass transport parameter used in Eqs. (13) and
(14) (cm'/mA)
a a lumped kinetic parameter equal to 1 for the
oxygen reduction reaction
e overall energy efficiency = current efficiency *
voltage efficiency
Cf current efficiency = theoretical reactant required/
amount of reactant consumed (g/g)
ev voltage efficiency = actual cell voltage/reversible
potential (V/V)
'lact ,a'act, activation polarization at the anode and cathode,
respectively (mV)
Sconc ,a1conc concentration polarization at the anode and
cathode, respectively (mV)

REFERENCES
1. Thomas, S., and M. Zalbowitz, Fuel Cells: Green Power Los Alamos
National Laboratory, LA-UR-99-3231 (1999); downloadable PDF file
available at
2. Larminie, J., and A. Dicks, Fuel C . .. . : Explained, John Wiley
& Sons, New York, NY (2000)
3. Hoogers, G., Fuel Cell Technology Handbook, 1st ed., CRC Press
(2002)
4. Hirschenhofer, J.H., D.B. Stauffer, R.R. Engleman, and M.G. Klett,
Fuel Cell Handbook, 5th ed., National Technical Information Service,
U.S. Department of Commerce, VA (2000)
5. Koppel, T., and J. Reynolds,A Fuel CellPrimer: The Promise and the
I .. . downloadable PDF file available at cgi-bin/fuelweb/view item/cat 18/subcat 19/product 20>
6. Prentice, G., Electrochemical Engineering Principles, Prentice Hall,
New Jersey (1991)
7. Bard, A.J., and L. Faulkner, Electrochemical Methods: Fundamentals
andApplications, 2nd ed., John Wiley & Sons, New York, NY (2000)
8. Fuel Cells 2000 Index Page, The Online Fuel Cell Information Center
at
9. Kim, J., S-M. Lee, and S. Srinivasan, "Modeling of Proton Exchange
Membrane Fuel Cell Performance with an Empirical Equation," J.
Electrochem. Soc., 142(8), 2670 (1995)
10. Williams, M.V., H.R. Kunz, and J.M. Fenton, "Evaluation of Polar-
ization Sources in Hydrogen/Air Proton Exchange Membrane Fuel
Cells," to be published in J. Electrochemical Society 5

Nanostructured Materials
Continued from page 37.
and results of peer evaluation.

CONCLUSION

ZSM-5 synthesis serves as an excellent example to intro-
duce students to the basic concepts of templated synthesis
and self-assembly that govern nanomaterials synthesis. This
experiment brings together a number of subjects that students
have learned from their previous courses: infrared spectros-
copy (from organic chemistry), kinetic analysis and reactor
operation (from reaction engineering), heat transfer (from
transport phenomena), and phase behavior (from thermody-
namics). The project also requires students to demonstrate
their creativity and innovation through the experimental de-
sign and implementation of a nanostructured material syn-
thesis.

ACKNOWLEDGMENTS

This work was supported by the NSF Grant CTS 9816954
and the Ohio Board of Regents Grant R5538.

REFERENCES
1. Gates, B.C., Catalytic ( . . . . John Wiley & Sons (1992)
2. Breck, D.W., Zeohte Molecular Sieves: Structure, ( . . . . and Use,
John Wiley & Sons (1973)
3. Dyer, A., An Introduction to Zeohte Molecular Sieves, John Wiley &

Winter 2004

Sons (1988)
4. Kerr, G.T., Catal. Rev - Sci. Eng., 23, 281 (1981)
5. Kerr, G.T., Sci. Am., 261, 100 (1989)
6. Thomas, J.M., Sci. Am., 266, 112 (1992)
7. Burkett, S.L., and M.E. Davis, Chem. Mater, 7, 920 (1995)
8. Kirschhock, C.E.A., V. Buschmann, S. Kremer, R. Ravishankar, C.J.Y.
Houssin, B.L. Mojet, R.A. van Santen, P.J. Grobet, P.A. Jacobs, and
J.A. Martens, Angew. Chem., Int. Ed., 40, 2637 (2001)
9. Kresge, C.T., M.E. Leonowicz, W.J. Roth, J.C. Vartuli, and J.S. Beck,
Nature, 359, 710 (1992)
10. Beck, J.S., J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D.
Schmitt, C.T.W. Chu, D.H. Olson, E.W. Sheppard, et al., J. Am. Chem.
Soc., 114, 10834 (1992)
11. Sayari, A., and S. Hamoudi, Chem. of Mats., 13, 3151 (2001)
12. Ying, J.Y, C.P. Mehnert, and M. S. Wong,Angew. Chem., Int. Ed. Engl.,
38, 56 (1999)
13. Corma, A., Chem. Revs., 97, 2372 (1997)
14. Konduru,M.V.,S.S.C. (.....- andX. Kang, . Phys. Chem. B., 105,
10918 (2001)
15. Monnier, A., F. Schuth, Q. Huo, D. Kumar, D. Margolese, R.S. Max-
well, G.D. Stucky, M. Krishnamurty, and P. Petroff, Science,, 261,
1299 (1993)
16. Huo, Q., R. Leon, P.M. Petroff, and G.D. Stucky, Science, 268, 1324
(1995)
17. Kim, J.M., and G.D. Stucky, Chem. Commun., 13, 1159 (2000)
18. Kresge, C.T., M.E. Leonowicz, W.J. Roth, J.C. Vartuli, and J.S. Beck,
Nature, 359, 710 (1992)
19. Konduru, M.V., and S.S.C. ( ....._ Catal., 196, 271 (2000)
20. Kruk, M., M. Jaroniec, V. Antochshuk, and A. Sayari, J. Phys. Chem.,
B, 106, 10096 (2002)
21. Treacy, M.M.J., and J.B. Higgins, Collection of SimulatedXRD Pow-
der Patterns of Zeolites, Elsevier (2001)
22. Coudurier, G., C. Naccache, and J. Vedrine, J. Chem. Soc., Chem.
Commun., 1413 (1982) 5

Fe R class and home problems

The object of this column is to enhance our readers' collections of interesting and novel prob-
lems in chemical engineering. Problems of the type that can be used to motivate the student by
presenting a particular principle in class, or in a new light, or that can be assigned as a novel
home problem, are requested, as well as those that are more traditional in nature and that eluci-
date difficult concepts. Manuscripts should not exceed ten double-spaced pages if possible and
should be accompanied by the originals of any figures or photographs. Please submit them to
Professor James 0. Wilkes (e-mail: wilkes@umich.edu), Chemical Engineering Department,
University of Michigan, Ann Arbor, MI 48109-2136.

Incorporating

GREEN ENGINEERING

Into a Material and Energy Balance Course

C. STEWART SLATER, ROBERT P. HESKETH
Rowan University * Glassboro, NJ 08028

Through the support of the US Environmental Protec-
tion Agency (EPA), a Green Engineering Project has
fostered efforts to incorporate green engineering into
the chemical engineering curriculum. Green engineering is
defined as the design, commercialization, and use of processes
and products that are feasible and economical while mini-
mizing generation of pollution at the source and risk to hu-
man health and the environment.
The Green Engineering Project has supported several ini-
tiatives, including development of a textbook, Green Engi-
neering: Environmentally Conscious Design of Chemical
Processes, [1] and dissemination through regional and national
workshops.[2] The latest phase of this project supports the de-
velopment of curriculum modules for various chemical engi-
neering courses.[3] This paper describes how the green engi-
neering topics are "mapped" into a material and energy bal-
ances course and presents a sample of the types of problems
that were developed for instructor use.
Green engineering principles should be familiar to and used
by all engineers, and the need to introduce the concepts to
undergraduates has become increasingly important.[4-6] The
most common method of incorporating it into the curriculum
has been through a senior/graduate elective course on envi-
ronmental engineering or pollution prevention.[7-9] Integrat-
ing green engineering principles into various chemical engi-
neering courses has been more challenging;[10] it is most of-

ten integrated into the design sequence.'11 Incorporating en-
vironmental issues into a material balance course has been
reported by Rochefort[121 by using a material balance module
developed by the Multimedia Engineering Laboratory at the
University of Michigan.[13] The uniqueness of the problem
module described in this paper is that it can be easily inte-
grated into a material and energy balances course and that it
maps many of the green engineering principles and underly-
ing concepts to topics covered at this level, thus providing
the basis for further integration of green engineering in
subsequent courses.
The introductory material and energy balances course is a
logical place to put basic terminology and concepts of green
engineering. The initial goal of this module was to "map"
some topics from the Green Engineering text to those taught

C. Stewart Slater is Professor and Chair of Chemical Engineering at
Rowan University He received his PhD, MPh, MS, and BS from Rutgers
University. His research and teaching interests are in the area of mem-
brane technology where he has applied these to fields such as specialty
chemical manufacture, green engineering, bio/pharmaceutical manufac-
ture and food processing.
Robert P. Hesketh is Professor of Chemical Engineering at Rowan Uni-
versity He received his PhD from the University of Delaware and BS from
the University of Illinois. He has made significant contributions to the de-
velopment of inductive teaching methods and innovative experiments in
chemical engineering and has done research in the areas of reaction en-
gineering, process engineering, and combustion kinetics.

@ Copyright ChE Division of ASEE 2004

Chemical Engineering Education

in the material and energy balances course, which predomi-
nately uses the text Elementary Principles of Chemical Pro-
cesses.[14] The curriculum module developed[151 has 25
problems (with solutions) that can be used by an instruc-
tor for in-class examples, cooperative learning, homework
problems, etc.
Two to four problems have been developed for each main
topic in material and energy balances and the majority of them
have multiple parts. Most require a quantitative solution, while
others combine both a chemical principle calculation with a
subjective or qualitative inquiry. The problems take a topic
from a particular subtopic/topic (section/chapter) and then
find a green engineering analog. Some cover specific termi-
nology, principle, or calculation covered in both texts, such
as in the calculation of vapor pressures of volatile organic
compounds (VOCs), while others introduce concepts only
covered in a green engineering text.
Presenting a topic found only in the green engineering text
is the most challenging integration of course material. For
example, the concept of occupational exposure is introduced
by having students perform a unit conversion with a dermal
exposure equation. In a similar way, workplace exposure lim-
its are introduced in the context of calculating concentration

TABLE 1
Conceptual Mapping of Green Engineer
in a Material and Energy Balances

Green Engineering Topic
* How green engineering is used by chemical engineers in the profession

* Various defining equations used in green engineering
* Typical method of representing concentrations of pollutants in a process (%, fractions, ppm, etc)
* * .. .11 .I ... I . i i ...." of a chemical manufacturing process
SBalances on recycle operations in green engineered processes
* Green chemistry in stoichiometry
* Combustion processes and environmental impact
* Use of various equations of state in green engineering design calculations for gas systems
* Pollutant concentrations in gaseous form

* 1 ,. ii . i. h ,,,. i. i i 1 . I... i 1 .. iii ........ I vapor recovery system s
. 1,, .h, , , . h '', l,, , , ,,.. I ,,,I ,,, . .,,,
* Representation of various forms of energy in a green engineering process
* Recovery of energy in a process-energy integration
* Use of heat capacity and phase change calculations
* Mixing and solutions issues in green engineering
* Energy use in green chemistry reactions, combustion processes
* Overall integration of mass and energy balances in green engineering on an overall plant design b
* Use of various simulation tools and specifically designed software for green engineering design
* R presentation oi ... ... I....,_ 11 I i 1. ..... . .'. '. ,o, _,.... ..'. ... , ' .. significar
* Industrial case studies of green engineered manufacturing processes
* From Felder & Rousseau14

using mole and mass fractions. This helps optimize time us-
age and course flow, since as prior papers on various subjects
have pointed out, "to put in X, you need to take out Y." By
taking basic material and energy concepts and designing a
problem to introduce a green engineering concept, a unique
integration of concepts occurs.
Some problems have additional questions that require stu-
dents to investigate the literature, go to a web site, or per-
form a more qualitative analysis of the problem. For example,
in the dermal exposure problem, the student must go to an
EPA or related web site to determine threshold limiting val-
ues and permissible exposure limits for other chemicals.
The level of green engineering material is quite elemen-
tary since the objective is to give students some familiar-
ity with concepts that would form the basis for more sub-
stantial green engineering problems in subsequent courses
such as transport, thermodynamics, reactor design, sepa-
rations, plant design, etc.
An overall conceptual view of green engineering topics
mapped to those in a material and energy balances course is
presented in Table 1. The mapping is done in a very generic
way so that an instructor can see the general outline of the
topics taught in a material and energy balances course and
some of the general ar-
eas of green engineer-
ing concepts. Not all of
ring Topics the concepts covered in
Course a material and energy
, .. balances course have a
Material and Energy Balances Topic*
green engmeenng ana-
Chap. 1: What Some ChEs do for a Living green engineering ana-
Chap log and vice versa.
Chap. 2: Intro. to Engineering Calculations That is why the EPA-
supported Green Engi-
Chap. 3: Process and Process Variables neering Project has
Chap 4: Fundamentals of Material Balances multiple modules de-
veloped for other
courses in the chemical
engineering curricu-
hap. 5: " lum. The material in
this module was devel-
Chap. 6: Multiphase Systems oped to be used at the
first-semester sopho-
more level and there-
Chap. 7: Energy and Energy Balances
fore integrates green
Chap. 8: Balances onNonreactive Processes engineering concepts
in a way that a student
Chap. 9: Balances on Reactive Processes starting a chemical en-
asis gineering program can
Chap. 10: Computer-Aided Calculations readily understand.
Several problems from
ice Chap. 11: Balances on Transient Processes the module have been
the module have been
Chap. 12-14: Case Studies presented below, fol-
lowing the order of in-

Winter 2004

corporation in the course. A full set of solved problems is
available at .

PROBLEM 1
Occupational Dermal Chemical Exposure Equation

Problem Statement
Undesired occupational exposure to chemicals contacting
the skin during sampling, splashing, weighing, transfer of
chemicals, process maintenance, etc., can be estimated as the
sum of the products of the exposed skin areas (cm2) and the
amount of chemical contacting the exposed area of the skin
(mg/cm2/event). The dermal exposure equation given below
can be used to estimate the exposure to a chemical absorbed
through the skin.
DA = (S)(Q)(N)(WF)(ABS) (1)
where
DA dermal (skin) absorbed dose rate of the chemical (mass/
time)
S surface area of the skin contacted by the chemical
lengthh)
Q quantity deposited on the skin per event (mass/length2/
event)
N number of exposure events per day (event/time)
WF mass fraction of chemical of concern in the mixture
dimensionlesss)
ABS fraction of the applied dose absorbed during the event
dimensionlesss)
Roberta Reactor, a process technician, is sampling a reactor
containing acrylonitrile. Unfortunately, she is not following
proper safety procedures for personal protection and is not
wearing the required gloves. As plant safety officer, you are
asked to estimate her dermal absorption rate (mg/workday)
for this unwanted exposure. Data from US EPA indicates that
batch process sampling yields between 0.7 and 2.1 mg/cm2
for the quantity Q in the dermal exposure equation.
a) Show that this equation is dimensionally homoge-
neous using the following units for the parameters:
DA (g/min); S (cm2); Q (mg/cm2event); N (event/
day).
b) Using the following data, determine DA in the units
of mg/workday for this exposure using the upper
limit of Q. During the workday, which is an 8-hour
shift, Roberta samples the reactor every hour and
exposes one of her hands. The mass fraction of
acrylonitrile in the reactor is 0.10 and the fraction of
the applied dose absorbed during the sampling is 1.0
(representing that all of the acrylonitrile contacting
the skin is absorbed).
c) What personal protective equipment must Roberta
wear?

(Problem can be used in Sections 2.2 and 2.6 of
Felder and Rousseau.)
Problem Solution
This problem introduces students to the concept of work-
place exposure to chemicals and methods for presenting the
associated risk. The parameters needed to solve the problem
are either given in the statement, found in the literature, or
must be measured. The surface area of the hand can be found
in texts-or for more fun, have the students trace their hands
on engineering paper and estimate the area, model the hand
as a trapezoid (palm) with cylinders (fingers), or use a
planimeter. This part of the problem gives the "hands-on"
characteristic to the learning experience.
To prove the equation is dimensionally correct, the student
inputs the units from the problem to show that they cancel on
the left-hand and right-hand sides of the equation. To solve
for the dermal absorption, the values are put into the equa-
tion and units are converted. A value of 325 cm2 for a student's
hand surface area is measured (literature value1] is 408.5 cm2
for median size of one adult woman's hand).

DA 325cm2 2.1 mg event 0.11.00 546mg (2)
cm2(event) day day

Information on the hazards associated with contact with this
chemical can be obtained by going to msds> and viewing a representative material safety data sheet
(MSDS) on acrylonitrile. Students will see that exposure to it
causes skin irritation, is harmful if absorbed through the skin,
may cause skin sensitization (an allergic reaction), that pro-
longed and/or repeated contact may cause defatting of the
skin and dermatitis, and that it is toxic in contact with skin.
They will also note from the web site that proper personal
protective equipment (gloves, safety goggles, and respirator)
must be used.
Students may also suggest that a method other than manual
sampling could be used to reduce risks to the technician and
avoid discharges into the workplace. This is a good practical
exercise and would help any student in a hazards and oper-
ability study (HAZOP) performed in subsequent laboratory
or project-based courses.

PROBLEM 2
Concentration Determination Using Threshold Limit
Value and Permissible Exposure Limits

Problem Statement
Two parameters that are used to establish workplace limits
for concentrations of chemicals are the Threshold Limit Value
(TLV) and Permissible Exposure Limits (PEL). TLV is the
level at which no adverse effect would be expected over a

Chemical Engineering Education

worker's lifetime. It is a guideline set by a nongovernmental
body, but the PEL is set by the U.S. Occupational Safety and
Health Administration (OSHA) and is considered the legal
limit in manufacturing facilities.
The solvent n-heptane is used in the manufacture of metal
components for washing the parts to remove oils used in the
cutting step. Several meters are used to monitor airborne con-
centration values in the plant. Your job as a process engineer
is to convert the data provided for TLV and PEL values for n-
heptane into the units used by the concentration meters shown
below.
a) Meter A: ppb
b) Meter B: mole fraction
c) Meter C: mass fraction
d) What are the consequences of an unwanted release of n-
heptane?
e) Suggest a more environmentally benign solvent for the
washing operation.
(This problem can be used in Section 3.3 ofFelder and
Rousseau.)

Problem Solution
This problem involves the concept of concentration and
incorporates the green engineering principle relating that con-
centration to workplace exposure limits of TLV and PEL.
The solution will involve the student first going to one of the
EPA-suggested websites and looking up the TLV and PEL
for n-heptane. By going to and
using the Mallinckrodt Baker MSDS for n-heptane, the val-
ues of TLV = 400 ppm and PEL = 500 ppm are obtained.
This problem can also involve students in learning how to
read an MSDS (which is shown later when they examine the
consequences of unwanted exposure). Next, students convert
to the desired units using conversions from ppm to ppb, mole
fraction, and mass fraction.
PEL Meter A

500 ppm 103 ppb 5.00 x 105 ppb (3)
ppm

PEL Meter B

500 ppm y- 5.00 x 10-4mol C7H16 / mol (4)
106 ppm

PEL Meter C
Choosing a basis of 100 moles and starting with the mole
fraction for meter B

To determine the risk associated with undesired release of
n-heptane in the plant workplace, students examine the MSD S
and see a health rating of 2, and for the section on hazards/
potential health effects they see the following for inhalation:
inhalation of vapors irritates the respiratory tract; it may pro-
duce light-headedness, dizziness, muscle incoordination,
loss of appetite, and nausea; and higher concentrations
can produce central nervous system depression, narcosis,
and unconsciousness.
In the last part of the problem, students investigate whether
an alternate solvent is more environmentally benign. Think-
ing of what solvents they might be using in a chemistry lab,
they might chose acetone, for which the same website would
give an overall health rating of 1, or slight, and PEL = 750
ppm and TLV = 750 ppm. So the solvent acetone is slightly
better environmentally than n-heptane to use. A listing of sol-
vents and their physical properties can be found using EPA's
free green chemistry expert system software.[16]

PROBLEM 3
Mass Balance on Reverse Osmosis Process for
Electroplating Waste Reuse and Recovery

Problem Statement
Reverse osmosis is a separation process used for pollution
prevention in many industries. It is an environmentally ef-
fective separation process since it can be used for material
recovery and recycle while it eliminates unwanted discharges
from a chemical manufacturing operation. In reverse osmo-
sis, a liquid feed stream under pressure passes across a semi-
permeable membrane filter that allows the passage of water,
but rejects organic and inorganic contaminants. In this op-
eration, the purified water stream produced is called the "per-
meate," and the stream of concentrated impurities is called
the "retentate."
You have been hired as a process development engineer
for Shiny Electroplaters, and your first assignment is to look
at the reduction of chromium discharge from its operation, as
shown in Figure 1. Considering the process to be a steady-
state continuous operation, determine
a) The permeate quantity (kg/hr) and chromium concen-
tration (mass fraction) being produced.
b) The potential uses for the permeate and retentate
streams in a "green" process design.
c) The advantages this process has over other pollution

5.00 x10-4molC7H16 100 mol 100.2g
mol mol

1.73x10-3gC7H16 /g

Winter 2004

5.00 x10-4molC7H16 100mol 100.2g (1-5.00x10-4)molAir 100mol 29g
mol mol mol mol

prevention techniques.
(This problem can be used in Section 4.3 ofFelder and
Rousseau.)

Problem Solution
This problem gives an example of a green manufacturing
process that uses a modem separation system such as reverse
osmosis for pollution prevention. It makes students think
about how the separation is used to make the manufacturing
operation "green." The problem is solved using a material
balance working on a continuous process at steady state. The
student performs a total mass balance and balance on chro-
mium over the process, yielding the following two relation-
ships:

1111 =1112+3 X1lll =X2l2 +X3m13

210kg/hr= 50kg/ hr+rit3 (0.10)(210kg/hr)=

(0.40)(50kg /hr)+X3(160kg /hr)

1113 = 160 kg/hr X3 = 0.00625

Students can brainstorm the potential uses of the permeate
and retentate to make this a "green" process by recycle and
reuse (see Table 2) and can then redraw the overall process to
show mass integration (Figure 2). Students speculate on the
advantages of this process from a green engineering stand-
point and find that it simultaneously produces a purified wa-
ter stream and concentrate with no phase change required-
energy savings: no by-products produced, no additional
chemicals required, operates at ambient temperature.

PROBLEM 4
Heating Value of Renewable Fuels

Problem Statement
Energy use, conservation, and the environmental impacts
of the production and use of fuels are important green engi-
neering topics. Currently available oil and coal reserves are
nonrenewable and have air-quality issues associated with their
use. Although there is no perfect fuel from an economic and
environmental perspective, there are alternatives that should
be considered.
Ethanol is considered a "green fuel" since it can be made
from renewable and sustainable resources and burns cleaner
than fossil fuels. The process to produce ethanol can use a
renewable resource such as domestically grown crops and

thereby lessens the need for importation of crude oil. Since
ethanol contains none of the carcinogenic compounds that
are found in fossil fuels, worker exposure risk is reduced. In
addition, when it is burned, ethanol generates fewer undes-
ired by-products than gasoline.
a) Investigate and draw a process flow diagram for the
production of ethanol from corn. Suggest methods of
mass and energy integration in this process to make it
more environmentally efficient
b) Calculate the higher heating value (HHV) and lower
heating value (LHV) of ethanol (kJ/mol).
c) How does this compare to the HHV of fuel oil
gasoline at 44 kJ/g? What are other comparisons of
fuel oil/gasoline combustion and ethanol combustion?
d) The use of hydrogen as a potential fuel of the future

TABLE 2
Potential Uses of the Permeate and Retentate to Make a
"Green" Process by Recycle and Reuse

Permeate Uses Retentate Uses
Process water Recovery of Chromium; send
concentrate to an electrolytic
cell
Wash water/rinse water Recycle to plating bath for
make-up of chromium losses
Water for dilution
Heat exchanging (energy integration)

Feed Permeate
Plating 1m v erse ri
Operation 210 kg/hr Osmo
0.10 kg Cr/kg
Retentate
1i2
50 kg/hr
0.40 kg Cr/kg

Figure 1. Process flow diagram of reverse osmosis for
the reduction of chromium discharge from
electroplating operation.

Figure 2. Process flow diagram showing the integration
of permeate and retentate streams.

Chemical Engineering Education

has received much recent attention. What is its HHV
(kJ/mol) and what are the environmental issues and
challenges related to its use?
(Problem can be used in Sections 9.4, 9.6, and( i-i "..
12-14 ofFelder and Rousseau.)

Problem Solution

This problem requires that students investigate the produc-
tion and use of ethanol fuel from a renewable and sustainable
resource. To find a suitable flow diagram for the production
of ethanol from biomass, students should be required to go
to the library and report the literature source used, such as a
biochemical engineering text or a technical encyclopedia.[17,18]
Students typically find the com-to-ethanol process uses fer-
mentation followed by various separations (including distil-
lation, membranes) that also show overall process integra-
tion of mass and energy.
Students next determine the heating values of ethanol yield-
ing HHV = 1366.9 kJ/mol and LHV = 1234.9 kJ/mol. A com-
parison of the heating values to gasoline is made and stu-
dents are asked to investigate other comparisons. From a green
engineering perspective, students are asked to investigate the
combustion products of gasoline and other fuel oils. They
will find that a 10% blend of ethanol reduces CO, CO2, VOCs
from evaporation, SO2, particulate matter, and aromatics com-
pared to burning gasoline.[19]
Finally, students are asked to examine hydrogen and deter-
mine heating values and other combustion issues. Here they
find that on a mole basis the HHV is 285.8 kJ/mol, but on a
mass basis, HHV is 141.5 kJ/g, which is higher than gasoline
or ethanol. They also see that H2 bums much more environ-
mentally efficiently since only water is produced as a com-
bustion product. A major issue in the use of hydrogen is its
source, which is typically a hydrocarbon.
Upon investigation, students will also see that it currently
costs more to produce hydrogen. Technology needs to be
developed to use it in the next generation of vehicles, and the
infrastructure to transport and dispense hydrogen fuels needs
to be developed.

CONCLUSIONS

Green engineering concepts can be integrated into a mate-
rial and energy balances course by using uniquely developed
examples and problems. These problems introduce terminol-
ogy and basic concepts that lay the groundwork for more
extensive incorporation of green engineering in subsequent
courses. Problems were developed within the framework of
a material and energy balances course and teach students about
topics such as workplace exposure routes/limits, recycle and
recovery processes, green chemistry, combustion, and mass
and energy integration. By using in-class examples or home
problems with a cooperative learning approach, students can

learn the concepts needed in both a material and energy bal-
ances course and green engineering.

ACKNOWLEDGMENTS
Support for work described in this paper originates from
the US Environmental Protection Agency, Office of Pollu-
tion Prevention and Toxics, and Office of Prevention, Pesti-
cides, and Toxic Substances X-83052501-1 titled "Green
Engineering in the Chemical Engineering Curriculum." Spe-
cial thanks go to Sharon Austin and Nhan Nguyen of the
Chemical Engineering Branch of the US EPA.

REFERENCES
1. Allen, D.T., and D.R. Shonnard, Green Engineering: E-w, ...M ,,
Conscious Design of Chemical Processes, Prentice Hall, Englewood
Cliffs, NJ (2001)
2.
3. Hesketh, R.P, M.J. Savelski, C.S. Slater, K. Hollar, and S. Farrell, "A
Program to Help University Professors Teach Green Engineering Sub-
jects in Their Courses," paper 3251, Proc. 2002 Am. Soc. Eng. Ed.
Ann. Conf, Montreal, QE (2002)
4. Bakshani, N., and D.T. Allen, "In the States: Pollution Prevention
Education at Universities in the United States," Poll. Preven. Rev,
3(1), 97 (1992)
5. Anon., "Chemical Companies Embrace Environmental Stewardship,"
Chem. & Eng. News, 77(49), 55 (1999)
6. Kuryk, B.A., "Global Issues Management & Product Stewardship,"
Proc. Global Climatic Change Topical Co . i- ' 2002 Spring Meet.,
New Orleans, LA (2002)
7. Abraham, M.A., "A Pollution Prevention Course that Helps Meet EC
2000 Objectives," Chem. Eng. Ed., 34(3), 272 (2000)
8. Grant, C.S., M.R. Overcash, and S.P. Beaudoi, "A Graduate Course
on Pollution Prevention in Chemical Engineering," Chem. Eng. Ed.,
30(4), 246 (1996)
9. Simpson, J.D., and W.W. Budd, "Toward a Preventive Environmental
Education Curriculum: The Washington State University Experience,"
J. Env Ed., 27(2), 18 (1996)
10. Gibney, K., "Combining Environmental Caretaking with Sound Eco-
nomics: Sustainable Development is a New Way of Doing Business,"
Prism, January (1999)
11. Brennecke, J.F., J.A. Shaeiwitz, M.A. Stadtherr, R. Turton, M.J.
McCready, R.A. Schmitz, and W.B. Whiting, "Minimizing Environ-
mental Impact of Chemical Manufacturing Processes,"Proc. 1999Am.
Soc. Eng. Ed. Ann. Conf, Charlotte, NC (1999)
12. Rochefort, W.E., "A Traditional Material Balances Course Sprinkled
with 'Non-Traditional' Experiences," Proc. 1999 Am. Soc. Eng. Ed.
Ann. Conf, Charlotte, NC (1999)
13. Montgomery, S., Multimedia Education Laboratory, University of
Michigan at
14. Felder, R.M., and R.W. Rousseau, Elementary Principles . .
Processes, 3rd ed., John Wiley & Sons, New York, NY (2000)
15. Slater, C.S., "Green Engineering Project: Material and Energy Bal-
ance Course Module," June (2003) greenengineering>
16. Green Chemistry Expert System (GCES), opptintr/greenchemistry/tools.html> US EPA, Office of Pollution Pre-
vention and Toxics, viewed 7/11/03
17. McKetta, J.J., and W.A. Cunningham, eds., Encyclopedia of Chemi-
cal Processing and Design, Marcel Dekker, New York, NY (1976)
18. Mark, H.F., M. Grayson, D. Eckroth, eds, Kirk-Othmer Encyclopedia
. 'Technology, 4th ed., John Wiley and Sons, New York, NY
(1991)
19. Canadian Renewable Fuels Association, Emissons Impact of Ethanol,
viewed 7/11/03 5

Winter 2004

classroom

TOP TEN WAYS

TO IMPROVE

TECHNICAL WRITING

JOHN C. FRIENDLY
Massachusetts Institute of Technology * Cambridge, MA 02139

While engineers often claim that they spend more
time writing than they do on any other single task,
providing constructive criticism of students' re-
ports is the most difficult and thankless task a faculty mem-
ber may face. Most schools do not have the luxury of having
a writing specialist who can help engineering students with
their reports, and even if students take a writing course, they
need feedback on their technical reports.
What rules of grammar, usage, and writing style should
students and faculty focus on? English usage changes with
time, and experts do not always agree, but in spite of numer-
ous excellent (and voluminous) style guides,E'-61 editing for
correct usage need not be a daunting task. There is a rela-
tively small list of topics that are particularly troublesome,
even for well-educated chemical engineering students.
In this paper, ten general suggestions are offered to help
improve one's technical writing style. They have been gleaned
during the past six years from several hundred drafts of in-
dustry reports submitted by over a hundred students at the
David H. Koch School of Chemical Engineering Practice at
MIT. Practice School students are candidates for the Masters
degree, and all have been well educated in some of the best
chemical engineering programs, both here and abroad. Re-
ports are submitted by two or three students working as a
group on real industrial projects at a company site. All re-
ports are written with an impending deadline, with two re-
ports expected during the typical one-month project dura-
tion.
The engineering education literature contains many ex-
amples of technical writing as part of the curriculum7-12] and
of writing pedagogy. [1314] In contrast, this top-ten list is in-
tended to supplement standard usage and style manuals that
have more depth. Strunk and White"151 remains a classic for
its brevity and good advice, and the ACS Manual ofS'tyle"'

is a comprehensive book that is useful to chemical engineers.
There are two useful manuals written by chemical engi-
neers.[1-19 No writer should suffer from a lack of reference
material. Spell- and grammar-check software should be used
as a minimum level of guidance, and style guides are avail-
able on the World Wide Web.[20,21l
This paper is intended to focus attention of both instructors
and students on the most prevalent writing problems. With
apologies to David Letterman, I will present and discuss the
top-ten list in reverse order. Each will be illustrated with ac-
tual examples of sentences from report drafts.

-10-
Select Words with Care
Misuse or overuse of some words occurs frequently enough
in technical writing to deserve special mention and ranks tenth
on my list of admonitions. There is such a diverse range of
examples that it almost defies categorization, but several of
the more common ones will be used to illustrate the problem.
It is well known that a spell or grammar checker cannot be
relied on as the sole source of misused words. Writing must
be proofread with care to make sure you have said what you
think you said. Sometimes an inadvertent slip seems so ap-

JohnC. Friendly has been Senior Lecturer and
Station Directorof the David H. Koch School of
Chemical Engineering Practice at MIT since
1996. In this capacity he has had assignments
at about a dozen different companies, at a va-
riety of sites both in the United States and
abroad. Before joining MIT he taught in the
Chemical Engineering Department at the Uni-
versity of Rochester

@ Copyright ChE Division ofASEE 2004

Chemical Engineering Education

propriate that it cannot be distinguished from a deliberate put-
on, as in
Original: This would lead to extra liquor sipping cost, which
is given in row 4.
Better: This would lead to extra liquor shipping cost,
which is given in row 4.

Chemical engineering students frequently use
the words setup/set up, scaleup/scale up, and
shutdown/shut down in their reports and mis-
use is not uncommon. The following example
shows that set up should be used when a verb
phrase is needed:

Original: The apparatus is setup so that any
overflow would be collected in the
trap.
Better: The apparatus is set up so that any
overflow would be collected in the
trap.

If the objective of a technical report is to get
across a message to the reader, pretentious
words have no place.[22] Perhaps no word gets
overused as much as utilize. It has a well-de-
served reputation of pretentiousness and should
probably never be used, since use is a simpler
synonym. Beware of trendy big words (such
as -ize verbs made from nouns, or nouns made

wording or words follow a pattern. This pattern can be in
verbs, nouns, adjectives, phrases, clauses, and sentences. It
can be extended to the organization of paragraphs, or even to
sections of a report. It improves the style and can make the
reader better understand that the ideas are parallel.

. .. editing for
correct usage
need not be
a daunting
task.
There is a
relatively small
list of topics that
are particularly
troublesome,
even for
well-educated
chemical
engineering
students.

from verbs) that sounds like bureaucratese (another example!)
at its worst. Do not try to make your prose impressive-make
it understandable.
For the most part, students have a good sense of the proper
use of words. Occasional lapses occur, however, on common
word pairs. Look out for there/their, J.. ../,. , between/
among, it's/its, continuously/continually, varying/various,
and ,/'. .'h.. 1' // ;. . th''.., It is easy to slip up and use
the wrong one.
Finally, technical writing is necessarily replete with acro-
nyms. Some are so common (such as CSTR), that they may
not need definition, but it is best to be cautious and consider
the reader. If a chance exists that your report will be read by
someone without your same perspective (and that includes
virtually everyone), define your acronyms the first time they
are used, and even more frequently if necessary. Never use
so many different acronyms that your reader is forced to di-
vert attention away from what you are saying to mentally
decode the terminology.

-9-
Use Parallel Construction
Writing is more effective when parallel ideas are presented
in parallel fashion. The reader's burden is lessened when the

Two obvious situations that call for parallel
construction are in enumerated lists and com-
pound expressions joined by correlative con-
junctions. Each one of the enumerated section
headings of this paper is an imperative admo-
nition starting with a verb and followed by its
object. Parallel construction may not always
be possible to maintain, but deviations from it
can be unnecessarily jarring to the reader. On
the other hand, correlative constructions using
the conjunction pairs both...and, either...or,
neither..nor and notonly.. but also can be mis-
leading or even incorrect if the words follow-
ing the correlative conjunctions are not paral-
lel to each other. Consider the example below.
In the original form, a verb form follows ei-
ther, but a noun phrase follows or. The natural
correction would be to move either so that
based on applies to either noun phrase. Both
noun phrases following the correlative con-
junctions are parallel, and it is clear that the
values will be assigned in either case.

Original: Values that are either based on engineering terms
or financial terms will be assigned to each piece of
equipment.
Better: Values that are based on either engineering terms
or financial terms will be assigned to each piece of
equipment.

-8-
Avoid Passive Voice and First Person
Good prose is direct and forceful. This is no less true in
technical writing. It is better to say that the subject did some-
thing than to say that something was done by the subject.
Technical writing tends to overuse the passive voice, some-
times with good reason. It is not wrong to use the passive
voice, but is should be avoided when possible.
Most technical writing also tends to avoid using the first
person. The message conveyed should focus on the technical
content without putting undue focus on the authors. Unfortu-
nately, the choice is often between using the first person (or
its close equivalent "the author") and using the passive voice.
It is not wrong to use the first person, but it should be avoided
when possible.
In the following example, the active voice makes the sen-
tence simpler and more direct. In this case, the Microsoft Word

Winter 2004

Most important, always consider
those who will be reading what you have
written and try to make it easier for
them to grasp your message.

grammar checker not only identified the passive sentence but
also suggested an improvement. Consider whether rewriting
each passive sentence would improve the flow of the sen-
tence and still convey the same information. If your sentence
is too complicated for the grammar checker to offer an im-
provement, maybe the sentence should be simplified.

Original: Two methods are being examined by the company
for possible implementation.
Better: The company is examining two methods for
possible implementation.

Technical writing should usually emphasize your accom-
plishments, not you yourself. This is the reason for avoiding
the first person, as illustrated in the example below. Using
other words, such as the authors, the group, and the project
team, may avoid the first person, but they do not avoid plac-
ing the emphasis in the wrong place. Use them advisedly,
even if it means using the passive voice.
Original: We followed established protocols to carry out the
measurements.
Better: Measurements were made following established
protocols.

- 7-
Use Proper Punctuation
The wide variety of possible punctuation problems justi-
fies its ranking of seventh on the top-ten list of things to watch
for. Most writers have a good sense of how to punctuate prop-
erly, so a comprehensive summary of the rules seems un-
necessary. Only two of the more common rules will be
mentioned here.
Technical writing too often uses long and complicated sen-
tence structures. If this is really necessary, good writing prac-
tice guides your reader through long sentences by using a
comma whenever it is appropriate to pause slightly. The fol-
lowing is a good example of where a comma prevents the
words from running together:
Original: The tin-catalyzed racemization rate also decreases
resulting in higher quality product.
Better: The tin-catalyzed racemization rate also decreases,
resulting in higher quality product.
The single comma should never be used to separate the
subject from the predicate of the sentence or the verb from its
predicate complement, however. The reader should proceed
directly from one to the other with no pause.

A related situation with the use of a colon arises frequently
in technical writing. The colon has only one proper use in
sentences: it separates a definition, a list, or other explana-
tory material from the rest of a complete sentence. It should
never be used to separate a verb from the rest of the predicate
or any other part of speech from its required complement.
The original version of the example below uses the list as the
direct object of the preposition into. The colon should not be
used there. If you want to use the colon, add the following or
some other object before the colon. The same rules apply if
the explanatory material is set off on the following line, as in
an enumerated list or an equation.
Original: These mechanisms can be classified into: solid-
solid interactions, liquid necking, adhesive and
cohesive forces, and chemical reactions.
Better: These mechanisms can be classified into solid-
solid interactions, liquid necking, adhesive and
cohesive forces, and chemical reactions.
Or: These mechanisms can be classified into the
following: solid-solid interactions, liquid necking,
adhesive and cohesive forces, and chemical
reaction.

-6-
Ensure Agreement in Number
Subjects and verbs must agree in both number and person.
Similarly, pronouns must agree with their noun antecedents.
Since most technical writing is done in the third person, per-
son agreement is not usually a problem. Number agreement,
however, can sometimes be a problem, especially in two com-
mon instances: recognizing the number of certain nouns and
recalling the true subject of a more complicated sentence.
The latter problem appears frequently enough in student re-
ports to justify this admonition as sixth most important.
A common mistake is to give the verb the number of the
closest noun rather than the true subject of the sentence. The
subject in the example below is measurements, not extrac-
tion, and the verb should thus be plural. Intervening phrases
or clauses, especially when they end with a noun, can draw
the writer's attention away from the true subject.
Original: The temperature measurements for the lab-scale
extraction was compared with the simulation
described above for validation.
Better: The temperature measurements for the lab-scale
extraction were compared with the simulation
described above for validation.
It is well known that words such as kinetics, economics,
and physics are singular in spite of the final s. Data can be
more troublesome. Classically plural, as the counterpart of
the currently unused datum, data has acquired a collective
use as well, requiring a singular verb. A good key to the dif-
ference is whether data points are or data set is can be sub-
stituted. If you can substitute either one, your sentence is prob-

Chemical Engineering Education

ably too vague to be useful. My suggestion is to be as helpful
to the reader as possible and avoid ambiguity. Think first that
the word data is plural and use data set if you really want it
used in the collective sense.

- 5 -
Place Modifiers with Care
Modifiers should always be placed as close to what they
modify as possible. No ambiguity about what word the modi-
fier belongs to should exist. The classic examples of inad-
vertent absurdities introduced by misplaced modifiers are easy
to catch, and the more subtle ones are fodder for technical
editors. Technical writing spawns more modifying words and
phrases than is consistent with clarity. The more modifiers
introduced into a sentence, the more likely that some ambi-
guity will arise. Grammar-check software can be used to alert
you to too many modifiers in your sentences. If the sentence
cannot be recast to avoid some of them, at least check to make
sure they are modifying what you wanted them to modify so
the reader will face no ambiguity.
The next example illustrates that the simple placement of a
modifier can drastically alter the sense of a sentence. In the
original wording, one might picture Erickson submerged in a
caustic solution making the diffusion measurement, instead
of the reaction occurring in the caustic tank. Place the modi-
fying phrase after the word reaction rather than as an intro-
ductory phrase.
Original: In the caustic retention tank, Erickson (1995) has
already confirmed that the neutralization reaction
is diffusion controlled.
Better: Erickson (1995) has already confirmed that the
neutralization reaction in the caustic retention tank
is diffusion controlled.

When a phrase has no word that it can logically modify, it
is called "dangling." The following is a good example. The
opening participial phrase should modify the person doing
the comparison. Placement of the phrase suggests that the
subject of the sentence would be the agent, but neither it nor
the cooking system could possibly be what the phrase modi-
fies. By the time the long modifying phrase was completed,
the writer had forgotten that the agent should be the subject
of the sentence.
Original: Comparing the characteristics of the steam tunnel
and those of the RotaTherm. as claimed by Gold
Peg and its distributors, it appears that the
RotaTherm steam fusion continuous cooking
system would be more advantageous.
Better: Comparing the characteristics of the steam tunnel
and those of the RotaTherm, as claimed by Gold
Peg and its distributors, we concluded that the
RotaTherm steam fusion continuous cooking
system would be more advantageous.

-4-
Use a Hyphen Only When Needed
Technical writing is plagued with jargon, and authors need
to learn how to use it consistently. Too often words are coined
ad hoc, using standard prefixes in combination with techni-
cal words to form a new word with a precise meaning un-
derstood by the reader. When to hyphenate such a prefix
is clearly not well defined, if one is to judge by the num-
ber of times that non-linear appears in respected publica-
tions. A good dictionary should always be the accepted
arbiter, but even the best ones will not cover all the tech-
nical terms clever students choose to use. This problem
frequently puzzles students.
The general rule is that particles such as bi, by, co, de, non,
pre, re, un, etc., that are not words by themselves should not
be hyphenated when added as a prefix to a word. (Modem
usage is different from that in older literature when new com-
pound words were hyphenated until they became accepted in
the vocabulary.) Also, no hyphen is called for when a num-
ber of longer prefixes are used, and the ACS Manual of Style
gives a long list of them, including anti, poly, post, counter,
super over under, infra, pseudo, etc. Consider the following
example.
Original: Agitate the device for a pre-determined period.
Better: Agitate the device for a predetermined period.
Two exceptions to the above rule should be noted. First,
use a hyphen when omitting it might cause confusion to the
reader. Any time ambiguity in meaning or pronunciation might
result, a hyphen should be used. Think, for example, of the
interpretation of "post-aging" if a hyphen is not used. Also,
always use a hyphen when the modified word requires a capi-
tal letter (for example, non-Newtonian). Second, consider
using a hyphen whenever the prefix introduces a double vowel
into the word. A hyphen is not needed in well-known words,
such as cooperative, however. For example, I would con-
sider preexponential a common enough term in chemical
engineering to permit dropping the hyphen, but others
would still require it.
Compound modifiers (words used together to modify a
noun) should be hyphenated. Application of this rule is
straightforward in many cases, but in others it is not. In the
example below, small-scale is a modifier of batch vessel. Note,
however, that batch is also a modifier of vessel. It is not hy-
phenated with small-scale. In this case, batch vessel seems
more natural as the noun expression being modified.
Original: Experiments were performed in a small scale batch
vessel, with samples taken periodically for
rheology measurements.
Better: Experiments were performed in a small-scale batch
vessel, with samples taken periodically for
rheology measurements.
Common technical terms that have a meaning together

Winter 2004

should not be hyphenated, however, even when used as a
modifier or descriptor. The hyphen tends to take away from
the common meaning of the expression mass transfer in the
example that follows.
Original: The capping experiments so far have been useful
for obtaining estimates of mass-transfer param-
eters.
Better: The capping experiments so far have been useful
for obtaining estimates of mass transfer param-
eters.

-3-
Go "Which" Hunting
This is a classic admonition from Strunk and WhiteE151 that
White apparently added to the original version.E"23 How often
it is ignored is perhaps surprising and is what makes it the
third most frequent writing problem I've encountered. Too
frequently it appears that the rules of usage are not known
rather than being consciously subverted.
That is a relative pronoun used to introduce a restrictive
clause, one that is necessary for the definition of the anteced-
ent that it should immediately follow. If the clause is removed,
the sentence will not convey its full meaning or the same
meaning. Such a restrictive clause should not be set off from
the antecedent by commas.
Which is a pronoun used to introduce a nonrestrictive clause,
one that is incidental to the definition of the antecedent that it
should immediately follow. Such a nonrestrictive clause can
be omitted without destroying the sense of the rest of the
sentence, and it should be set off from the rest of the sentence
by commas. In the example that follows, the sentence ending
at "parameters" would be incomplete-the following clause
is restrictive to the nature of parameters being described. The
clause should be introduced by that rather than which. The
grammar check in Microsoft Word will catch the incorrect
use in the original sentence.
Original: a, and b, are parameters which can be determined
by flux measurement.
Better: a, and b, are parameters that can be determined by
flux measurement.
Unfortunately, some good writers will use which in place
of that to introduce a restrictive clause. It has had an accepted
literary use for effect,241 although the advantage is more of-
ten than not difficult to see. Whether such use was purpose-
ful or inadvertent is impossible to determine. For modern tech-
nical writing, it is probably best to avoid such use and to go
which hunting as White advises.
Which clauses may also be used to modify the sense of the
entire main clause of the sentence. This use is hardly neces-
sary, however, and a simple rewording can avoid it. The reader
is spared the possible ambiguity of trying to discover the noun
that the which clause modifies. In technical writing this use
should probably also be avoided. The following example,

although not incorrect as originally written, shows that chang-
ing the which clause to a participial phrase avoids possible
confusion about whether the which clause actually modified
the natural antecedent solution.
Original: CO, was observed bubbling out of solution, which
would result in a higher pH.
Better: CO, was observed bubbling out of solution,
resulting in a higher pH.

-2-
Use Direct and Concise Statements
The second most common problem with writing styles is
verbosity. Writing concisely is an art that needs to be prac-
ticed. If there is a direct way to say something, use it. If there
is a shorter way to say something, use it. Of the many ways
verbosity appears in student reports, two have been selected
here for illustration.
An introductory phrase or clause can be useful in making a
transition from, or connection to, previous sentences and to
orient the reader to the main clause that follows. A common
writing problem is the use of such a phrase to indirectly say
what the sentence is about when a more direct and concise
approach would suffice. Consider the following example in
which the introductory prepositional phrase was meant to help
the reader know what was being compared. The shorter sen-
tence is more direct and less awkward, however, and con-
veys the sense just as well.
Original: Between water content and temperature, the latter
had the stronger effect on the viscosity.
Better: The temperature had a stronger effect on the
viscosity than water did.
A common example of verbosity is to use a phrase in place
of a single word. Many phrases have become cliches and
should not be used at all. Others should be used with discre-
tion. In the following example, due to the fact that is used
when a simple because would be appropriate. Other phrases
you should look out for include the reason is because, it is
because, considered to be, by means of in order to, and for
the case where. Other phrases, such as in terms of as is un-
derstood, result of is that of kind of the fact that, and type of
might best be eliminated entirely.
Original: This was due to the fact that more water condensa-
tion from the vapor was required to vaporize the
additional hexane.
Better: This occurred because more water condensation
from the vapor was required to vaporize the
additional hexane.

-1-
Use Specific and Precise Language
By far the most common weakness I have found is a fail-

Chemical Engineering Education

ure to be specific enough. This may arise because of uncer-
tain knowledge of new material or because of the material's
relevance, but it shows in a number of ways. In many cases,
specific information is easy to include; in others it may not
be, but the wording should not be vague or imprecise.
Of the many different types of nonspecific writing, three
have been singled out for illustration here. The first type is
related to weak words that include such as, like, including,
for example, various, diverse, certain, and some. They do
have a definite place in writing, but too frequently they ap-
pear to weaken the strength of an otherwise specific state-
ment. In the next example, no other property was of interest
in the study, and the use of such as added an element of vague-
ness that was totally unnecessary. Look for examples in your
own writing and ask yourself if the specific cases would not
serve your purpose better. Reserve the use of such as for places
where you truly need to give illustrative examples from a
much larger set.

Original: A fundamental study was conducted to obtain
fundamental data such as isosteric heat of adsorp-
tion.
Better: A fundamental study was conducted to obtain the
isosteric heat of adsorption.

The second type of shortcoming is a failure to use specific
numbers when possible. When conveying technical results
in a report, specific numerical values should be used when-
ever possible. The next example shows that amounts with
nonspecific adjectives of degree should be replaced by spe-
cific values when possible. Although the original statement
may not be wrong, the more specific the reporting, the better
the result usually is. Watch out for similar modifiers, such as
majority, most, high, low, large, small, and even some, and
other expressions such as around about, approximately, and
the order of magnitude, to see if they can be removed by
using specific numerical values. Reserve the use of such words
for situations in which the numerical values are not precise,
but in which you want to convey some sense of magnitude.
Original: A representative crude oil composition containing
high amounts of tocopherol was used as the feed
for these processes.
Better: A representative crude oil composition containing
2% tocopherol was used as the feed for these
processes.

The third type of nonspecific writing deals with the pre-
sentation of results. Too frequently, students feel that it is
sufficient to present their results in a table or graph without
explanation. Although this is sometimes enough, more often
it is not. Only in rare cases will the readers be able to pick out
the gist of the results and draw the same conclusion that the
author did. It is the responsibility of the writer to point out
what the results showed and how conclusions were drawn
from them. Do not force the readers to interrupt their train of
thought in the report to study the details of the results. Chances

are, their focus will be different from your own.

CONCLUSION

Writing technical reports or assessing someone else's writ-
ing should not be an overwhelming task. The top ten sugges-
tions made here can be used to good advantage in focusing
on the most common problems in technical writing. Practice
in recognizing when and how writing can be improved will
go a long way toward making you a better technical writer.
Most important, always consider those who will be reading
what you have written and try to make it easier for them to
grasp your message.

REFERENCES

1. Burchfield, R.W., ed., The New Fowler's Modern English Usage, Ox-
ford University Press, New York, NY (2000)
2. Siegal, A.M., and W.G. Connolly, TheNewYork7 I
and Usage, Time Books, New York, NY (1999)
3. Grossman, J., ed., The Chicago Manual oj .... 14th ed., University
of Chicago Press, Chicago, IL (1993)
4. Wilson, K.G., The Columbia Guide to Standard American English,
Columbia University Press, New York, NY (1993)
5. Goldstein, N., ed., The AssociatedPress . . elManual,
Associated Press, New York, NY (1993)
6. Rubens, P., ed., Science and Technical Writing: A Manual oJ ..
2nd ed., Routledge, New York, NY (2001)
7. Newell, J.A., D.K. Ludlow, and S.P.K. Sternberg, "Development of
Oral and Written Communication Skills Across an Integrated Labora-
tory Sequence," Chem. Eng. Ed., 31, 116 (1997)
8. Schulz, K.H., and D.K. Ludlow, "Group Writing Assignments in En-
gineering Education," J. Eng. Ed., 85, 227 (1996)
9. Hirt, D.E., "Student Journals: Are They Beneficial in Lecture Courses?"
Chem. Eng. Ed., 29, 62 (1995)
10. Ludlow, D.K., and K.H. Schulz, "Writing Across the Chemical Engi-
neering Curriculum at the University of North Dakota," J. Eng. Ed.,
83, 161 (1994)
11. Ybarra, R.M., "Safety and Writing: Do They Mix?" Chem. Eng. Ed.,
27, 204 (1993)
12. Pettit, K.R., and R.C. Alkire, "Integrating Communications Training
into Laboratory and Design Courses," Chem. Eng. Ed., 27, 188 (1993)
13. Sharp, J.E., B.M. Olds, R.L. Miller, and M.A. Dyrud, "Four Effective
Writing Strategies for Engineering Classes,"J. Eng. Ed., 88, 53 (1999)
14. Dorman, W.W., "Engineering Better Writers: Why and How Engineers
Can Teach Writing," Eng. Ed., 75, 656 (1985)
15. Strunk, W., and E.B. White, The Elements oj . .. 4th ed., Allyn and
Bacon, Boston, MA (1972)
16. Dodd, J.S., ed., The ACS . Guide, 2nd ed., American Chemical
Society, Washington, DC (1997)
17. Blake, G., and R.W. Bly, The . . .ri .. . ting, Longman,
New York, NY (1993)
18. Haile, J.M., .. . .... Macatea Productions, Central, SC (2002)
19. Bly, R.W., "Avoid These Technical Writing Mistakes," Chem. Eng.
Prog., p. 107, June (1998)
20. Bartleby, , accessed September 2003 (2003)
21. Lexico LLC, , accessed September
2003 (2003)
22. Phatak, A., and R.R. Hudgins, "Grand Words But So Hard to Read,"
Chem. Eng. Ed., 27, 200 (1993)
23. Strunk, Jr., W., The Elements o. ... W.P. Humphrey Press, Ithaca,
NY (1918)
24. White, E.B., "The Family Which Dwelt Apart," from Quo Vadimus,
Part I, Harper & Brothers, New York, NY (1939) O

Winter 2004

[n9, learning in industry

UOP-CHULALONGKORN UNIVERSITY

INDUSTRIAL-UNIVERSITY

JOINT PROGRAM

SANTI KULPRATHIPANJA, ANN KULPRATHIPANJA
UOP LLC * Des Plaines, IL

Since recovery of natural gas began in the Gulf of Thai-
land in the late 1970s, the need for petrochemical tech-
nology in that area has continually increased due to
the rapid development of value-added processes for natural
gas and LPG. Examples of such processes are dehydrogena-
tion of ethane to ethylene and of propane to propylene. In
addition to natural gas conversion, other areas of petroleum
and petrochemical processing for converting petroleum to
higher value-added products are of increasing interest in
Thailand. One example is the conversion of naphtha to aro-
matics, followed by the separation of individual aromatics
from each other. The individual pure aromatics can then be
converted to even higher value products. For example,
para-xylene can be converted to terephthalic acid, and sub-
sequently to polyester.
Because of the high demand for petrochemical technology
in Thailand, an international graduate program in "Petro-
chemical Technology and Polymer Science" was inaugurated
in 1992 at Chulalongkorn University, one of Thailand's
prominent universities. Through this international graduate
program, select students who are enrolled in the Petroleum
and Petrochemical College (PPC) at Chulalongkorn Univer-
sity have an opportunity to perform research for their thesis
at one of three participating universities located in the United
States. The participating U.S. universities and departments
include the Department of Macromolecular Science and En-

gineering at Case Western Reserve University, the Depart-
ment of Chemical Engineering at the University of Michi-
gan, and the School of Chemical Engineering and Materials
Science at the University of Oklahoma. When the Petroleum
Technology Program was launched in 2002, the international
graduate program was also extended to include an institute
located in France, the Institut Francais du Petrole.
Through these international graduate programs, U.S. and
French faculty members teach at PPC each year, and in addi-
Santi Kulprathipanja has worked for UOP
LLC since 1978. He is currently an R&D Fel-
low and has been recognized as a distin-
guished UOP inventor for being named on
more than 90 U.S. patents. His works have
resulted in many of UOP's commercial sepa-
ration processes. He has edited a book en-
titled Reactive Separation Processes, co-
authored a chapter on "Liquid Separation",
and published more than 30 technical papers.

Ann Kulprathipanja is a patent attorney at
Kinney andLange, a boutique Intellectual prop-
erty law firm in Minneapolis, MN. She was a
previous internee at UOP and interacts with the
UOP-PPC student research program in the
area of intellectual propertyA
Copyright ChE Division of ASEE 2004

Chemical Engineering Education

This column provides examples of cases in which students have gained knowledge, insight, and experi-
ence in the practice of chemical engineering while in an industrial setting. Summer internships and co-op
assignments typify such experiences; however, reports of more unusual cases are also welcome. Description
of the analytical tools used and the skills developed during the project should be emphasized. These ex-
amples should stimulate innovative approaches to bring real-world tools and experiences back to campus for
integration into the curriculum. Please submit manuscripts to Professor W. J. Koros, Chemical Engineering
Department, Georgia Institute of Technology, Atlanta, GA 30332-0100

tionto teaching, some of the U.S. faculty members work with
a Thai counterpart in supervising graduate students. Because
they are jointly supervised by U.S. and Thai faculty mem-
bers, some of the Thai students at Chulalongkorn University
are given the opportunity to carry out part of their thesis work
at one of the three U.S. universities.
After initial implementation of the international program,
PPC recognized the importance of expos-
ing its graduate students to practical ex-
perience. Thus, the international graduate
program subsequently expanded its col-
laboration to an industrial setting. The
UOP-PPC program is a first endeavor at
providing Thai students with an opportu- a(
nity to carry out research in an interna- gaii
tional industrial environment.
engineers

Polymer Science Program is currently supported by UOP
Housing expenses, along with a limited stipend for living
expenses while the students are conducting experiments at
UOP, are also provided by UOP each year. Travel expenses
from Thailand to the United States are paid by the stu-
dents while expenses incurred by attendance at the tech-
nical conference are provided by the university. UOP's

Exposure to industrial practices
provides the students with a more
comprehensive background than a solely
cademic-based education. The experience
led then acts as a model for scientists and
s in the refining and petrochemical fields.

INDUSTRIAL INVOLVEMENT

The program was begun with the purpose of producing
graduates of high international standards and developing
world-class research and development (R&D) in the petro-
leum and petrochemical fields. As part of the program, in-
dustrial scientists are invited to give lectures and to super-
vise graduate students in their research at PPC.
In conjunction with this purpose, in 1997 Dr. Santi
Kulprathipanja of UOP LLC, a graduate of Chulalongkorn
University with over 25 years of industrial experience, was
invited to give special experience- and industrial-application
based lectures. In addition to his technical expertise, Dr.
Kulprathipanja's knowledge of both the Thai and American
cultures functions as a useful bridge by providing insight as
to how to most effectively assist the students in adapting to
their new environment.
UOP is a company known for process innovation, technol-
ogy delivery, and catalyst/adsorbent supply to the petroleum
refining, petrochemical, and gas processing industries. In
1998, Dr. Kulprathipanja supervised his first graduate stu-
dent at PPC, and she later presented her research at a Cana-
dian chemical engineering conference.
Observing that the program would be beneficial to Thai
students, Dr. Kulprathipanja agreed to supervise two of them
in 1999, allowing one to perform research at UOP for two
weeks. From this beginning, future students supervised by
Dr. Kulprathipanja were permitted to conduct basic research
at UOP. Prior to returning to Thailand to complete their gradu-
ate work, the students are given an opportunity to present
their research at a meeting of the American Institute of Chemi-
cal Engineers (AIChE), the American Chemical Society
(ACS), or the North American Membrane Society (NAMS).

INVOLVEMENT/CONTRIBUTIONS OF UOP
The industrial aspect of the Petrochemical Technology and

participation caters to the mutual interests of the com-
pany and the students.
Through the program, UOP has an opportunity to help con-
tribute to the establishment of petroleum and petrochemical
R&D in Thailand by educating the students. The students learn
industrial techniques while obtaining valuable research ex-
perience. With the guidance of other knowledgeable research
scientists and technicians at UOP, the Thai students are ex-
posed to proper experimentation procedures and safety guide-
lines, which are more stringent in the U.S. In return, through
the students' research, UOP gains useful data and basic ana-
lytical information that it might otherwise not have the time
or resources to explore.

CASE STUDIES
While at UOP, the students focused on four major research
areas: adsorption, mixed matrix membranes, reactive sepa-
ration, and catalysis. The following case studies will demon-
strate the students' capabilities as they researched areas of
adsorption and mixed matrix membranes at UOP LLC.

* Case 1 - Adsorption: The ParexTM process, which uses
UOP's well-known SorbexTM "simulated moving bed" ad-
sorptive separation technology to separate p-xylene from other
C-8 aromatics, generates more than half of the p-xylene in
the world. Because of UOP's expertise in C-8 aromatics ad-
sorptive separation, three students were encouraged to carry
out adsorption research in September and October of the years
2000 through 2002. The purpose of the adsorption study was
to understand the interaction mechanism between the
adsorbents and adsorbates. The adsorbents were zeolites X
and Y exchanged with Li, Na, K, Rb, Mg, Ca, Sr, and Ba.
The adsorbates were C-8 aromatics: p-xylene, m-xylene, o-
xylene, and ethylbenzene. The adsorbents were characterized

Winter 2004

using x-ray, TGA, ammonia-TPD, and chemical analysis. The
students were initially trained to prepare adsorbents and C-8
aromatic feed stock. They subsequently studied the interac-
tion using a myriad of techniques, including: the multicom-
ponent dynamic pulse test to determine adsorbent selectivity
to each C-8 aromatic, the multicomponent dynamic break-
through to measure adsorbent selectivity, mass transfer rate
and capacity for each C-8 aromatic, and single and multi-
component equilibrium adsorption isotherm to measure ad-
sorbent selectivity and capacity for each C-8 aromatic. The
results were then analyzed by a model simulation. In brief,
the study indicated that the interaction mechanism between
the adsorbents and C-8 aromatics is influenced by various
factors, including: the acid-base interaction between zeolite
and C-8 aromatics, exchanged cation size, C-8 aromatics feed
composition, and zeolite Si/Al ratio. The results were used to
fulfill the students' MS theses1- 31 and were presented at the
AIChE meetings. UOP benefited from the results by gaining
a basic understanding that will assist in further C-8 aromat-
ics separation improvement development.

* Case 2 - Mixed Matrix Membranes: There were two
types of mixed matrix membranes (MMM) developed at UOP
LLC in the early 1980s. The first MMM has zeolite embed-
ded in the cellulose acetate (CA) polymer phase.i45 The sec-
ond MMM is produced by casting an emulsion of polyethyl-
ene glycol (PEG) and silicone rubber (SIL) on a porous
polysulfone (PS) support.[6-9] It was found that both types of
MMMs offered many interesting features in enhancing se-
lectivity and permeability if the MMM was composed of a
comparable pair of polymer and zeolite or PEG. Based on
this finding, four students were invited to the UOP Research
Center during September and October of 1999 to 2002 to
study/explore/discover new MMMs for interesting applica-
tions. Their objectives were to develop new types of MMMs
for olefin/paraffin separation and carbon dioxide separation
from natural gas. During the program, the students were
trained to formulate MMMs, carry out permeation studies,
and analyze data. Many encouraging MMMs were developed
by the students for olefin/paraffin separation.110-11] For ex-
ample, the students found that ethylene/ethane and propy-
lene/propane selectivity were enhanced by PEG/SIL/PS
MMM.1101 Their selectivity was reversed with NaX/CA and
AgX/CA MMMs, however.[11] In the case of carbon dioxide
separation, a novel type of MMM was developed to enhance
both CO2/N2 selectivity and CO2 permeability. The MMM
was composed of PEG, activated carbon, and silicone rubber
on polysulfone .12,13" Through this novel MMM, it was found
that activated carbon can stabilize PEG and further enhance
CO2 permeability and selectivity. In addition to the basic un-
derstanding that UOP obtained from the students' work on
activated carbon and PEG, UOP also filed a patent applica-
tion due to the novel nature of the silicone rubber on
polysulfone composite MMM. The data and analyses obtained

from the research were used to fulfill the students' MS the-
ses1io-13] and were presented at the AIChE meetings.

CONCLUSION
The Petrochemical Technology and Polymer Science Pro-
gram stresses the reality that most graduate students will even-
tually work in industry. Exposure to industrial practices pro-
vides the students with a more comprehensive background
than a solely academic-based education. The experience
gained then acts as a model for scientists and engineers in the
refining and petrochemical fields. In addition to the experi-
ence obtained by the students, UOP also benefited from the
students' work. UOP has gained basic research information
and has continued to use the information to further commer-
cial process development. Overall, with the collaboration of
UOP management, scientists, technicians, and others, the stu-
dents in the program gained practical experience, presenta-
tion experience, and a more established reputation. The par-
ticipating universities also benefited by gaining recognition
on an international level.
The primary accomplishment of the program is to offer the
opportunity for students in developing countries to obtain a
solid foundation of knowledge by learning about other cul-
tures and working in a professional environment. The fol-
lowing paragraphs demonstrate the impact the program has
had on former participants.

TESTIMONIALS
By Ms. Warangkana Sukapintha and Mr. Thera
Ngamkitidachkul* (1999) * Learning under real working con-
ditions has broadened my vision and has enabled me to pre-
pare for practical work. For two weeks, UOP allowed me to
train in the R&D department, tour a UOP pilot plant, and
visit the engineering and patent departments. These opportu-
nities gave me the invaluable experience of seeing real work
in a real company. I learned that one of the most important
factors of doing work efficiently is being able to work well
as part of a team. Additionally, as an unknown graduate stu-
dent, it is almost impossible to be invited to an international
meeting. Therefore, the opportunity to present a paper and
attend the AIChE 2000 Spring National Meeting was one of
the greatest experiences of my life. Now, in addition to the
fundamental knowledge that I gained from my studies at PPC,
I have also expanded my vision through industrial training.
Overall, the opportunities to work under Dr. Kulprathipanja,
to visit UOP, and to attend an AIChE meeting helped poten-
tial employers realize my capabilities.
By Mr. Varoon Varanyanond, Ms. Worrarat Rattanawong,
and Ms. Passawadee Vijitjunya' (2000) * We obtained ben-
efits from our stay at UOP that could not be obtained solely
*Thera Ngamkitidachkul, Passawadee Vijitjunya, Prueng
Mahasaowapakkul, Kathavut Visedchaisri did not intern at UOP. They
carried out their research work at PPC.

Chemical Engineering Education

from the University. The strongest advantage of working in a
company was the availability of technical knowledge. Under
the guidance of an expert, we acquired wider and deeper
points-of-views. The state-of-the-art equipment and facilities
also enabled us to effectively work on our research. We felt
that anything was possible. The picture of how to apply the
knowledge that we obtained from the classroom became clear.
One of the most important educational tools we gained was
the safety indoctrination provided by UOP. We also had the
honor of presenting our work at an international conference
where we developed communication skills and a result-fo-
cused style of thinking. These skills are some of our stron-
gest points in getting a job. We believe the program will cer-
tainly give students a chance to develop themselves, as well
as profit industry. Last, but not least, we would like to ex-
press our gratitude to Dr. Kulprathipanja, who worked so hard
to give us this precious opportunity.

By Ms. Rattiya Suntornpun, Ms. Jutima Charoenphol,
Mr. Visava Lertrodjanapanya, and Mr. Prueng
Mahasaowapakkul' (2001) * For two months we were able
to carry out our research at UOP under the close supervision
of Dr. Kulprathipanja. This was a great opportunity for us to
learn from a person with a strong industrial background.
Meeting people from different backgrounds allowed us to
learn more than just technical know-how. For example, they
stimulated diverse ideas, increasing the likelihood that we
would find the best solution to any problem. Moreover, we
became more open-minded to other people's thoughts. We
also learned that there were no exceptions when it came to
safety matters. A large advantage of researching at UOP was
the access to information. While we sometimes have to wait
for a publication to be sent from abroad at PPC, this was
never a problem at the UOP library. At the end of the pro-
gram, our research was presented to an international audi-
ence at the AIChE 2001 Annual Meeting. We were able to
practice our oral presentation skills and learn from the ques-
tions people asked about our research. Overall, this experi-
ence gave us more confidence in ourselves, making us more
attractive to employers.

By Ms. Raweewan Klaewkla, Ms. Saowalak Kalapanulak,
Ms. Parichart Santiworawut, Ms. Suwanna Limsamutch-
aiku, and Mr. Kathavut Visedchaisri' (2002) * We received
a great opportunity from UOP to perform some of our re-
search at UOP. We learned various techniques such as: pre-
paring catalysts, casting membranes, setting up adsorption
experimental lines, and using modem analysis instruments.
An important observation that we made regarding UOP's
working style was that while they directed most of their at-
tention to their work, they were also prompt to provide each
other with assistance. This general rule-of-practice influenced
us to effectively work on our research. We were able to ob-
tain both high quality and high quantity work in a limited
amount of time. Before we left the United States, we also

had a chance to present our research at the 2002 AIChE An-
nual Meeting. This trip opened our minds to the international
world that we would not have been able to experience if we
stayed only in our country and our college. Moreover, we
learned a lot from the different cultures, languages, foods,
living styles, and beautiful places. These impressive things
could not have happened without Dr. Kulprathipanja and the
UOP LLC staff. We would like to express our thanks and let
them know that we are all very appreciative.

ACKNOWLEDGEMENTS
Integral in making this program successful are the indi-
vidual efforts of certain UOP R&D staff: Dr. Laszlo Nemeth,
Dr. James Rekoske, Dr. Linda Cheng, Dr. Joe Kocal, Dr. Greg
Lewis, Mr. Greg Maher, Mr. Jaime Moscoso, Mr. Darryl
Johnson, Mr. James Priegnitz, Mr. Vasken Abrahamian, Mr.
Dave Mackowiak, Mr. Sathit Kulprathipanja and Mrs. Wanda
Crocker, and faculty members of the PPC, Chulalongkorn
University: Professor Somchai Osuwan, Assistant Professor
Pramoch Rangsunvigit, Associate Professor Thirasak
Rirksomboon, Assistant Professor Pomthong Malakul and Dr.
BoonyarachKitiyanan. Special acknowledgements are also
due to Dr. Robert Jensen, Dr. Jeff Bricker, Dr. Stan
Gembicki, Dr. Jennifer Holmgren, Associate Professor
Kunchana Bunyakiat, and Mrs. Apinya Kulprathipanja for
their hospitality, and to UOP TCO for its financial sup-
port of the program.

REFERENCES
1. Ngamkitidachakul, T., MS Thesis, 'Fundamentals of Xylene Adsorp-
tion Separation" Chulalongkorn University, Bangkok, Thailand (2000)
2. Varanyanond, V., MS Thesis, "Competitive Adsorption of C.-aromat-
ics and Toluene on KY and KBaX Zeolites" Chulalongkom Univer-
sity, Bangkok, Thailand (2001)
3. Suntornpun, R., MS Thesis, "Acid-Base Interaction between C.-aro-
matics and X and Y Zeolites" Chulalongkorn University, Bangkok,
Thailand (2002)
4. Kulprathipanja, S., R.W. Neuzil, and N.N. Li, "Separation of Fluids
by Means of Mixed Matrix Membranes" U.S. Pat. 4,740,219 (1988)
5. Kulprathipanja, S., R.W. Neuzil, and N.N. Li, "Separation of Gases
by Means of Mixed matrix Membranes" U.S. Pat. 5,127,925 (1992)
6. Kulprathipanja, S., "Separation of Gases From Nonpolar Gases" U.S.
Pat. 4,606,740 (1986)
7. Kulprathipanja, S,, and S.S. Kulkami, "Separation of Gases From Non-
polar Gases" U.S. Pat. 4,606,060 (1986)
8. Kulprathipanja, S., S.S. Kulkami, and E.W. Funk, "Multicomponent
Membranes" U.S. Pat. 4,737,161 (1988)
9. Kulprathipanja, S., S.S. Kulkami, and E.W. Funk, "Separation of Gas
Selective Membranes" U.S. Pat. 4,751,104 (1988)
10. Sukapintha, W., MS Thesis "Mixed Matrix Membrane for Olefin/Par-
affin Separation" Chulalongkorn University, Bangkok, Thailand (2000)
11. Rattanawong, W., MS Thesis "Zeolite/Cellulose Acetate Mixed Ma-
trix Membranes for Olefin/Paraffin Separations," Chulalongkorn Uni-
versity, Bangkok, Thailand (2001)
12. Serivalsatit, V., MS Thesis "Mechanism of the Mixed Matrix Mem-
brane (Polyethylene Glycon/Silicone Rubber) Separation for Polar
Gases", Chulalongkorn University, Bangkok, Thailand (1999)
13. Charoenphol, J., MS Thesis "Mixed Matrix Membranes for CO2/N2
Separation", Chulalongkorn University, Bangkok, Thailand (2002) O

Winter 2004

classroom

USE OF CONCEPTESTS

AND INSTANT FEEDBACK

IN THERMODYNAMICS

JOHN L. FALCONER
University of Colorado * Boulder CO 80309-0424
any studies have emphasized the fact that coop-
erative learning can improve engineering educa-
tion.11,2] One form of cooperative learning in phys-
ics and chemistry departments is in-class ConcepTests3'4]-
multiple-choice conceptual questions posed to the class. Af-
ter all the students respond with an answer, they are asked to
discuss the answers among themselves (peer instruction) and
are given the opportunity to change their answer.
Mazur[3] showed a lack of correlation between students'
conceptual understanding of physics and their ability to do
quantitative problems. They could do quantitative problems
better than conceptual problems that used the same concept.
He found that students memorized algorithms for solving the
problems without understanding the concept, and thus had
difficulty when a problem they had to solve was different
from ones they have previously solved. He reported a gain in
student performance with the use of ConcepTests. The stu-
dents' conceptual understanding increased because they
were better able to explain concepts to one another than
their teachers could. The percentage of students with the
correct answer always increased after they discussed the
question with their peers.
This effectiveness of ConcepTests can be further improved
if students are graded on their answers because it increases

John L. Falconer is Professor of Chemical
and Biological Engineering and a Presidents
Teaching Scholar at the University of Colo-
rado at Boulder He received his BS from the
Johns Hopkins University and his PhD from
Stanford University. He teaches courses in
thermodynamics, reactor design, research
methods and ethics, and catalysis. His cur-
rent research interests are in heterogeneous
photocatalysis and the preparation and char-
I acterization of zeolite membranes.

both their participation and their motivation. The grading is
done with IR transmitters and receivers, as described below.
My experience in a thermodynamics course showed the fol-
lowing advantages:
* Students liked using ConcepTests and .. i,." instant
feedback on how well they understood material as it
was presented to them.
* The instructor obtained instant feedback on how well
the class understood a concept.
* Students were more motivated to be prepared and thus
learned more in class.
* Attendance in class was higher than in previous
semesters when ConcepTests were not used . 1 Iih,,1. hi
statistics were not obtained for the previous semes-
ters, attendance was over 90% when ConcepTests
were used and graded.)
* Everyone participated in class.
* The discussions among students were quite lively.
Students interacted, teaching and learning from their
fellow students. This creates a more engaged class
and students hear more than one explanation. This
increases learning.
Although ConcepTests were a small part of the course grade,
grading them motivated the students. For the thermodynam-
ics course, the lowest five days of grades were dropped to
allow for sickness, outside activities, etc. The ConcepTest
grades then counted either 5% or 10% of the final course
grade. The higher of the two grading methods was used for
each student. Since the average on the ConcepTests was 88%,
almost all students counted the ConcepTests as 10% of their
grade. An important aspect was the use of an absolute grad-
ing scale for the course. This encouraged students to cooper-
ate; they were also required to do homework in groups.
This brief article describes ConcepTests and the relatively

@ Copyright ChE Division ofASEE 2004

Chemical Engineering Education

inexpensive technology available that significantly improves
their application. Both the technology and ConcepTests have
been in use for some time in physics and chemistry depart-
ments. The purpose of this article is to indicate that they are
also effective in chemical engineering courses, particularly
those courses that require significant conceptual understand-
ing, and that inexpensive technology exists for implement-
ing the test and getting instant feedback.
Examples used during the Fall 2002 semester for a junior-
level chemical engineering thermodynamics course will be
presented here. Grading and instant feedback were accom-
plished by installing IR detectors in the classroom and re-
quiring students to purchase IR transmitters clickerss) manu-
factured by H-ITT.[5 There were fifty students in the class.

EXPLANATION OF CONCEPTESTS
The ConcepTests with transmitters clickerss) works as fol-
lows:
1. The instructor poses a conceptual question and
presents possible answers (multiple choice).
2. Each student picks an answer by selecting A,B,C,D, or
E on a clicker.
3. The instructor displays a histogram of answers for the
class to see. If most answers are correct, a short
explanation is given and the next topic is started.
4. If many of the answers are incorrect, students are told
to discuss the question with their neighbors. This peer
instruction is a critical aspect of ConcepTests and
learning. It fosters student involvement and engage-
ment.
5. Students are allowed to change their answers after the
discussion. As a result, most of the students end up
with the correct answer and a better understanding.
6. If most students have the correct answer, a brief
explanation is given. If not, the question is discussed
further, and the instructor provides additional ideas to
help the students learn the concept.
Three receivers were mounted high on the walls in the room
for a class of fifty students. The receivers are small (3.5 x 2.5
x 1.5 cm) and are daisy-chained together by cables. The
cost of 3 receivers and cables was around $600. The re-
ceivers collect the signals and send them to a PC running
acquisition software, which can be downloaded free from
the H-ITT web site.E51
Each student has their own hand-held transmitter (clicker),
purchased from the bookstore for $30. The H-ITT hand-held
IR transmitter, similar to a TV remote control, has a unique
ID number. It is slightly larger than a pen and is battery oper-
ated. Each student responds to the multiple-choice questions
by aiming the clicker at a wall-mounted receiver and press-
ing A, B, C, D, or E. The H-ITT acquisition program display

is also projected onto a screen for the entire class to see. The
ID number (or the student initials) of each clicker is displayed,
indicating that the student response has been successfully
collected, but it does not show the student answer. The H-
ITT acquisition program summarizes the data and displays
the class responses in histogram form.
After class, a separate program associates student names
with the remote ID numbers and grades the responses in-
stantly. It allows the instructor to assign point values to each
answer for each question (e.g., 3 points for a correct answer
and 1 point for an incorrect answer). The software also al-
lows a list of the student names and point totals to be quickly
exported into a spreadsheet.

EXAMPLES FROM THERMODYNAMICS
Several examples from the thermodynamics course are pre-
sented here. Many students initially had problems answering
these types of questions since some of them require higher
levels of Bloom's taxonomy. The examples are presented to
give the reader an idea of how ConcepTests are applied in
class. Similar problems were then used on the course exams,
but without the multiple-choice options and with the require-
ment that the students explain the reason for their answers.

1. Components (A and B)
are in vapor-liquid
equilibrium. One mole
of liquid (xA = 0.4) and
0.1 mol of vapor (Y =
0.7) are present (see
Figure 1). When 0.5 mol
of A is added and the
system goes to equilib-
rium at the same T and
P, what happens?
A. The amount of liquid
increases.
B. The amount of liquid
decreases.
C. The concentration of
A in the gas phase
increases.
D. The concentration of
A in the liquid phase
increases.

2. Is the fugacity of water
at 1500C and 100 atm
closer to
A. 1 atm
B. 5 atm

Vapor

y1=0.7

Liquid

x1=0.4

Figure 1. Two-component
vapor-liquid phase equi-
librium in a piston/cylin-
der at constant pressure
equilibrium.

Winter 2004

C. 50 atm
D. 100 atm

3. For the H-x, diagram at 80 �C in Figure 2, what is the
maximum value of the partial molar enthalpy in cal/
mol of component A?

100

80

0 60

, 40
-I-

20

0

0 0.2 0.4 0.6 0.8 1
XA
Figure 2. Enthalpy of a binary mixture versus mole
fraction of component A.

A. 50
B. 22
C. 85
D. 100
E. 0

4. Two identical flasks at 45 �C are connected by a tube.
One flask (A) contains water and the other (B)
contains the same amount of a 95/5 mixture of water
and salt. After five hours
A. Beaker A has more water.
B. Beaker B has more water.
C. The amounts of water do not change since they are
at the same temperature.
D. All the salt moves to beaker A.

5. Consider the reversible reaction and the indicated
number of moles present at equilibrium:
CaCO3(s) <- CaO(s) + C02(g)
10 mol 0.2 mol 10 mol
If we push down on the piston (see Figure 3) to
decrease the volume to half and keep the temperature
constant, what happens at equilibrium?
A. The CO2 pressure almost doubles.
B. CaO and CO2 react, so the CO2 pressure does not
change.
C. The system is at equilibrium, so nothing changes.
D. All the CO2 reacts.

6. 6 mol A and 4 mol B
are in equilibrium at
100 �C and 3.0 atm.
A and B are com-
pletely immiscible in
the liquid phase.
Their vapor pressures
at 100 �C are
P Asat = 2.0 atm
P at= 0.5 atm.
What phases are
present?
A. Liquid B and
vapor of A + B
B. Two liquids
C. Two liquids in
equilibrium with
vapor
D. All vapor
E. Liquid A and
vapor of A + B

C02

CaCO3(sf ,

Figure 3. Gas-solid
chemical equilibrium
in a
piston/cylinder.

7. Water alone is present
and is in VLE at 1.2 atm in a piston/cylinder. You
inject 5 cm3 of air into the system, but keep P and T
constant. What happens?
A. All the water vaporized.
B. All the water condenses.
C. Some water vaporizes.
D. Some water condenses.

FEEDBACK FROM THE
FALL 2002 THERMODYNAMICS CLASS
At the end of the Fall semester, students turned in an anony-
mous typed course evaluation to the TA. These evaluations
were given to the instructor after course grades were posted.
One area that the students were asked to address was the use
of clickers and ConcepTests. Partial comments from fifteen
of those evaluations follow. Almost everyone in the class liked
the clickers and ConcepTests.
* The greatest part about it was that you made
thermodynamics a fun class to attend. The IR
transmitters did not follow a straight lecture and I
found they are a good idea, and I found them to be
quite useful in understanding the ConcepTests.
* There was one thing in particular that I really
enjoyed, and that was the clicker questions.
* As for the instant response clicker system, it was
generally a big help. I think it is essential to

Chemical Engineering Education

teaching such technically difficult material
as we study in thermodynamics. Being Th,
able to immediately apply what we were [Co
learning to a problem and receive instanta-
neous feedback on our understanding, as a
class, was fantastic.
* Although I was a bit skeptical of the
transmitters at first, I found that I actually
liked them a lot. It kept the class interest-
ing to be able to participate every day.
* The transmitters were very effective in adding to
the class as a learning experience. They gave
support to myself in times when I felt unwilling to
ask a question for fear I was the only one who
didn't understand.
* The ConcepTests were extremely helpful in
getting a grasp on what is happening. I also liked
the use of the transmitters.
* I thought the clickers worked well in class. These
questions were very useful at helping me grasp the
conceptual part of the course.
* I thought the overhead ConcepTests were a great
idea, and a good usage of the clickers.
* I felt the use of the transmitters greatly enhanced
my understanding of the topics we discussed.
* The IR transmitters receive two thumbs up. I was
skeptical of them at first, but they really help in
making sure that not only I but the majority of the
class understands what is being taught.
* I also liked the concept questions.... I thought the
IR transmitters worked very well and were used
well. The IR transmitter is good because there is
no peer pressure factor when you're answering the
question for the first time, and you can get a good
idea of the class understanding of the concept.
My favorite parts to this course were the supple-
ments in the notes and the IR transmitter...I felt the
IR transmitter and the ConcepTests were a
valuable tool to this class.
* I thought the best aspects of the course were the
transmitters, the reviews, and the homework help
sessions. The transmitters were definitely a good
way to get people to participate.
* I felt the IR transmitter and ConcepTests were a
valuable tool in this class.
* Ultimately I found that the clicker really helped
my learning. It also keeps you involved with the
lecture, rather than just mindlessly copying down
notes.
The concerns expressed by the students were small. The
biggest concern was that they had to spend $30 to purchase a

e purpose of this article is to indicate that
rncepTests] are also effective in chemical
engineering courses, particularly those
courses that require significant
conceptual understanding ...

transmitter they could use only in one course. Since they
should be able to sell their transmitters to students in next
year's class, that should become less of a problem. Some stu-
dents were concerned that the grading in every class forced
them to come to class more often. Two students did not like
the transmitters or the ConcepTests.

SUMMARY

Even though students could work numerical problems,
many did not have a good grasp of the thermodynamic con-
cept involved. For example, they could calculate the vapor
pressure at a given temperature with Antoine's equation, but
a large fraction of them did not understand the concept of va-
por pressure well enough to answer questions such as #7 above.
For many of the ConcepTests used, more than half the class
initially answered incorrectly, but the percentage of correct
answers increased, usually dramatically, after discussions with
other students.
The H-ITT software was easy to use in class, and the stu-
dents could readily see their clicker ID number on the pro-
jected display. Since their ID number always appeared in the
same location on the screen, it was easy to find. We have
since installed the detectors in a second room in the engi-
neering building, and two other faculty members have indi-
cated they will use the clickers in their classes in the future.

ACKNOWLEDGMENTS
I could not have incorporated this method into my class
without the help and advice of Dr. Michael A. Dubson in the
Physics Department at the University of Colorado. I would
also like to acknowledge the funds from the President's Teach-
ing Scholar program and from the Dean's Office to purchase
the equipment.

REFERENCES
1. Felder, Richard M., at EducationPapers.html>
2. Wankat, P.C., and F.S. Oreovicz, TeachingEngineering, McGraw-Hill,
New York, NY (1993)
3. Mazur, E., Peer Instruction: A User's Manual, Prentice Hall, Upper
Saddle River, NJ (1997)
4. Landis, C.L., A.R. Ellis, G.C. Lisensky, J.K. Lorenz, K. Meeker, and
C.C. Wamser,( . - . , ConcepTests: A Pathway to Interactive Class-
rooms, Prentice Hall, Upper Saddle River, NJ (2001)
5. 7

Winter 2004

classroom

RUBRIC DEVELOPMENT FOR

ASSESSMENT OF

UNDERGRADUATE RESEARCH

Evaluating Multidisciplinary Team Projects

JAMES A. NEWELL, HEIDI L. NEWELL, KEVIN D. DAHM
Rowan University * Glassboro, NJ 08028

Experts agree on the importance of involving under-
graduates in research-based learning11-3] and team-
work.[4-6] The Boyer Commission suggested that re-
search-based learning should become the standard for under-
graduate education.F] Many universities are responding to this
challenge by introducing multidisciplinary laboratory or de-
sign courses.[8,9] At Rowan University, we developed a method
of addressing these diverse challenges while also implement-
ing valuable pedagogical hands-on learning experiencesi1o111]
and technical communications.[12-14]
At Rowan University, all engineering students participate
in an eight-semester course sequence known as the engineer-
ing clinics.[151 In the junior and senior years, these clinic
courses involve multidisciplinary student teams working on
semester-long or year-long research projects led by an engi-
neering professor. Most of the projects have been sponsored
by regional industries. Student teams under the supervision
of chemical engineering faculty have worked on emerging
topics that included enhancing the compressive properties of
Kevlar, examining the performance of polymer fiber-wrapped
concrete systems, advanced vegetable processing technology,
metals purification, combustion, membrane separation pro-
cesses, and many other areas of interest. Every engineering
student participates in these projects and benefits from hands-
on learning, exposure to emerging technologies, industrial
contact, teamwork experience, and technical communications.
Difficulties arise in trying to assess student learning and
performance in project-based team settings, however. Angelo
and Cross[16] provided significant suggestions for assessing
the attitude of students toward group work, but provided little
insight into distinguishing individual and team performances.

One difficulty is that evaluating the semester-long perfor-
mance of teams working on projects involves a substantial
number of variables. Clearly, successful completion of the
project's technical aspects is an essential component for dem-
onstrating student understanding, but Seat and LordE171 ob-
served that while industry seldom complains about the tech-
nical skills of engineering graduates, industrial employers and
educators are concerned with performance skills (i.e., inter-
personal, communication, and teaming). Lewis, et al., "18 cor-
rectly observed that if students are to develop effective team-
ing skills, teaming must be an explicit focus of the project.
It is unreasonable to expect students to achieve specific
learning objectives from a series of courses when the faculty
members themselves are unclear about what the learning ob-
jectives are and how to measure them. Young, et al.,[19] dis-

James Newell is Associate Professorof Chemical Engineering at Rowan
University. He currently serves as Secretary- Treasurer of the Chemical
Engineering Division ofASEE and has won both ASEE's Ray Fahien
Award for his contributions to engineering education and a Dow Out-
standing New Faculty Award. His research interests include high-per-
formance polymers, outcomes assessment, and integrating communi-
cation skills through the curriculum.
Heidi Newell is currently the assessment coordinator for the College of
Engineering at Rowan University. She previously served as the assess-
ment consultant for the University of North Dakota. She holds a PhD in
Educational Leadership from the University of North Dakota, an MS in
Industrial and Organizational Psychology from Clemson University, and
a BA in Sociology from Bloomsburg University of Pennsylvania
Kevin Dahm is Assistant Professor of Chemical Engineering at Rowan
University. He received his BS from Worcester Polytechnic Institute in
1992 and his PhD from the Massachusetts Institute of Technology in
1998. His primary technical area is in chemical kinetics and mecha-
nisms. His current primary teaching interest is integrating process simu-
lation throughout the chemical engineering curriculum, and he received
the 2003 Joseph J. Martin Award for work in that area.

� Copyright ChE Division ofASEE 2004

Chemical Engineering Education

cussed development of a criterion-based grading system to clarify
expectations to students and to reduce inter-rater variability in grad-
ing, based on the ideas developed by Walvoord and Anderson.E"20
This effort represented a significant step forward in course assess-
ment; however, for graded assignments to capture the program-
matic objectives, a daunting set of conditions would have to be
met. Specifically,

* Proper course objectives that arise exclusively from the educational
objectives and fully encompass all of these objectives must be set
* Tests and other graded assignments must completely capture these
objectives
* Student performance on exams or assignments must be a direct
i, rl,, r. .,, of their abilities and not be influenced by test anxiety,
poor test-taking skills, etc.

There should be a direct correlation between student performance
in courses and the overall learning of the students only if all of
these conditions are met every time. Moreover, much of the peda-
gogical research warns of numerous pitfalls associated with using
evaluative instruments (e.g., grades on exams, papers, etc.) within
courses as the primary basis for program assessment. 211
Obviously, a more comprehensive assessment method for a team-
oriented, research-project based course mustbe developed. Woods[221
listed the following five fundamental principles for assessment of
teams:
1. Assessment is based on performance

TABLE 1
Summary of Specific Indicators for Areas of Importance

Area of Importance Specific Indicators
Technical * Defined objectives
* Demonstrated technical awareness
* Obtained and interpreted appropriate results
* Formulated supportable conclusions
* Properly considered error
* Provided recommendations for future work

Logistical * Organized project
* Met deadlines
* Executed project plan
* Kept detailed records

Laboratory Operation * Maintained safe practices
* Developed hazardous operations (HAZOP) report
* Dressed appropriately
* Proper use/maintenance of equipment
* Performed end-of-semester shut down

Teaming * Division of labor
* Professional conduct
* Learning experiences for all team members

Part of the purpose of this
pilot program was to clarify
for the students the expectations in
junior/senior clinic by providing
specific information about
their learning goals.

2. Assessment is a judgment based on evidence rather
than on feelings
3. Assessment must have a purpose and have clearly
defined performance goals
4. Assessment is done in the context of published goals
and measurable criteria
5. Assessment should be based on multidimensional
evidence
Rowan's Chemical Engineering Department is imple-
menting the following strategy for improved assessment
of student team projects: decide on the desired learning
outcomes for the clinic, develop indicators that demon-
strate whether or not the teams (and each member of the
team) have achieved each of the outcomes, develop ru-
brics to evaluate student performance in each of the ar-
eas, and present all of this information to the students at
the start of the project.

PILOT PROGRAM
In the junior/senior engineering clinic, each student
team submits a final written report and gives an oral pre-
sentation, which allows the communication aspects of
the project to be evaluated directly, but the remaining
elements of a successful project experience had to be
identified and measured. As a first effort to address the
assessment of team performance in project-based re-
search experiences, the faculty developed the following
list of four learning objectives of primary importance
that were common to all projects:
* Technical performance
* Project planning and logistics
* Laboratory operation
* Teaming
Once these objectives were identified, specific indica-
tors were developed for each so the students would have
clearly defined behaviors. Table 1 summarizes these in-
dicators.
With the specific indicators determined, the next step
involved developing descriptive phrases that would as-
sist both students and faculty members in evaluating stu-
dent performance. It became clear that specific descrip-
tions of the level of performance in each area would be

Winter 2004

required. The goal of our rubrics was to map student work
directly to the individual learning outcomes. As Banta1231
stated, "The challenge for assessment specialists, faculty, and
administrators is not collecting data but connecting them."
The assessment rubric also followed the format developed
by Olds and Miller[24] for evaluating unit operations labora-
tory reports at the Colorado School of Mines.
The decision to frame the rubrics based on only three lev-
els was significant and requires explanation. At one time,
many of the other program-assessment instruments used by

Rowan's Chemical Engineering Department used a 5-point
Likert scale with qualitative labels (5=excellent, 4=very good,
3=good, 2=marginal, l=poor), but the qualitative natures of
the descriptive labels led to confusion in scoring. Some pro-
fessors have different distinctions between "excellent" and
"very good" and tended to use them more than the descrip-
tive phrases that define the difference between levels for each
indicator. More important, if the rubrics are well designed,
the descriptive phrases should stand alone, without the need
for subjective clarifiers such as "excellent" and "good." Ulti-

TABLE 2
Behaviors Corresponding to Technical Performance

Indicator An "A" Team A "B" Team A "C"-or-Lower Team
Defined objectives Is actively involved in defining aggressive Aids in defining objectives. Some may be Takes little initiative in defining the project.
and achievable objectives that thoroughly too simplistic or unrealistic.
address fundamental project needs.

Demonstrated technical Clearly demonstrates awareness of the work Shows understanding of the work in the Fails to demonstrate an awareness of the
awareness of others and establishes a context for field, but has limited depth and breadth. work of others and the significance of
their project. Shows an understanding of Knowledge is limited to faculty-provided of their project.
information from multiple literature sources. materials.

Obtained appropriate Obtained meaningful results with minimal Produced some results but not enough Generated few meaningful results.
results wasted effort. (or too many).

Interpreted data Provided thorough and correct analysis of Provided analysis but partially incorrect or Little meaningful analysis of data or
appropriately data. insufficiently thorough. blatantly incorrect.

Formulated supportable Formulated and adequately supported Needed significant help in formulating Conclusions are absent, wrong, trivial, or
conclusions meaningful conclusions, meaningful conclusions or lacked unsubstantiated.
sufficient support for their conclusions.

Properly considered error Used appropriate mathematical and technical Error analysis is largely qualitative or Sources of error and reproducibility issues
skills to quantitatively express limitations of incomplete, are ignored or misinterpreted.
of the data.

Provided Makes insightful recommendations about Makes broad or obvious suggestions Makes no plausible suggestions for future .
recommendations for future work. for future work. work.
future work

TABLE 3
Behaviors Corresponding to Project Planning and Logistics

Indicator An "A" Team A "B" Team A "C"-or-Lower Team

Organized project Effectively organizes project tasks to Identifies relevant tasks but may struggle Has difficulty converting broad objectives to
minimize wasted time and effort. with setting priorities and planning. specific tasks.

Met deadlines Consistently meets deadlines. Misses some deadlines despite reasonable Routinely ignores deadlines.
effort.

Executed project plan Effectively and safely executes the project Executes the project plan but has difficulty Works haphazardly with little chance of
plan. Makes significant progress. overcoming setbacks. achieving project objectives.
Modifies the plan as necessary.

Kept detailed records Keeps detailed records easily followed by Keeps a lab notebook but records lack Keeps poor, sketchy, or no records.
others. These records include a laboratory organization or contain omissions.
notebook, computer files, purchase records,
and others.

Chemical Engineering Education

mately, we decided to eliminate such descriptors and divide
rubric elements by listing behaviors that demonstrated the
level (1, 2, or 3) at which the student had obtained the de-
sired learning outcomes.[251
These previously developed rubrics, however, were pro-
grammatic assessment tools that were seen and used only by
the faculty. Part of the purpose of this pilot program was to
clarify for the students the expectations in junior/senior clinic
by providing specific information about their learning goals.
Students tend to be more focused on grades than on learning
outcomes, so characterizations such as "level 1 vs. level 2"
would be meaningless to them, and subjective phrases such
as "excellent" and "good" would be subject to the same short-
comings described above. Further, if grading truly represents

the measure of achievement of learning outcomes, it is not
unreasonable to present the behaviors that demonstrate suc-
cessful attainment of a learning outcome in terms of grades.
Consequently, the rubrics were written for presentation to the
students in terms of behaviors that an A-Team would demon-
strate, a B-Team would demonstrate, etc., Tables 2 through 5
provide the rubrics.
Both the chemical engineering faculty at Rowan and the
reviewers of this paper questioned if the "C-or-Lower" range
was too broad. Some items were barely acceptable, while
others could be dangerous. There was even a question about
whether or not laboratory safety could be scaled at all. We
decided to stay with three levels for several reasons. First,
we did not want students bargaining about the lower-level

TABLE 4
Behaviors Corresponding to Laboratory Operations

Indicator An "A" Team A "B" Team A "C"-or-Lower Team

Maintained safe practices Develops and follows procedures that account Develops and follows procedures consistent Fails to develop and follow safe procedures
for safety and clean-up. Lab is clean and neat. with safe practices but sometimes misses and/or clean up.
minor safety issues or fails to clean up.

Developed Hazardous Conducts a thorough Haz-Op. Performs a Haz-Op but focuses on obvious Fails to perform a Haz-Op or performs one
Operations (HAZOP) issues without depth (e.g., does not check inadequately.
report MSDS sheets).

Proper use/maintenance Treats equipment with care and performs
of equipment necessary maintenance.

Usually handles equipment properly but has
an occasional lapse.

Uses equipment carelessly or fails to maintain

Performed Lab area is neat and clean. Lab notebook Must be pushed by the faculty member for Fails to accomplish some of the listed items.
end-of-semester and electronic copies of all data and reports the behaviors described previously.
shut down are provided to the faculty member. Samples
and materials are labeled appropriately and
are either stored or disposed of properly.

TABLE 5
Behaviors Associated with Teaming

Indicator An "A" Team A "B" Team A "C"-or-Lower Team
Division of labor Has all members making significant Progresses satisfactorily but some members Internal conflicts result in team failing to
contributions to a project that progresses . feel that workload distribution was . achieve project goals.
satisfactorily. disproportionate.

Professional conduct Consistently behaves in a professional Usually behaves in a professional manner Frequently fails to behave in a professional
manner (shows up for meetings prepared and (shows up for meetings prepared and on manner (shows up for meetings prepared and
on time; treats vendors, technicians, team time; treats vendors, technicians, team on time; treats vendors, technicians, team
members and staff with courtesy and respect; members, and staff with courtesy and members and staff with courtesy and respect;
external communications are formal and. respect; external communications are formal external communications are formal and
businesslike). Always dresses appropriately and businesslike). Usually dresses businesslike). Frequently fails to dress
(long pants and safety glasses in labs; appropriately (long pants and safety glasses appropriately (long pants and safety glasses in
business attire for industrial meetings and in labs; business attire for industrial labs, business attire for industrial meetings and
presentations, etc.). meetings and presentations, etc.).Does not presentations, etc.).
repeat errors.

Learning experiences for Has all team members demonstrate a Has all technical issues understood by Has team members with significant gaps in
all team members thorough understanding of the technical someone on the team, but is segmented. their understanding of technical issues.
issues of the project. Some members do not have the whole
picture.

Winter 2004

Faculty distributed the tables to the students at the beginning of the semester,
referred to them throughout the semester in giving feedback on student performance,
and used them to aid in assigning and justifying a final grade.

behaviors (e.g., "I can be late
for three meetings and still get TA
a 'C,' but the fourth one gets me Faculty Assessmi
a 'D'."). The lowest-level be- (1 strongly disa
haviors were to be avoided en-
tirely, so we chose not to put a
distinction between "bad" and * The grading rubrics helped m
"really bad." The other impor- of my project.
tant point to keep in mind is that
* The grading rubrics helped m
the rubric items do not repre- would be graded.
sent individual grades, but
rather a holistic approach to * The grading rubrics helped m
evaluating all of the factors on that I otherwise might not h:
a team. If the team has mostly I referred to the grading rubri
A-level performances but also
has some "C-or-Lowers," it * Ithink that clinic is more fair
would likely lower their project
d to a * I would like to use the rubrics
grade to a "B."

RESULTS
AND DISCUSSION

The rubrics have two uses, each of which was piloted within
the Chemical Engineering Department during the 2002-03
academic year. The first is that it will facilitate grading that is
uniform, fair, and clearly understood by the students. Faculty
distributed the tables to the students at the beginning of the
semester, referred to them throughout the semester in giving
feedback on student performance, and used them to aid in
assigning and justifying a final grade.

The second use of the rubrics is assessment of the junior/
senior clinic program as a whole. As mentioned above, sim-
ply using course grades as a primary assessment tool (even
when the grades are fair and based on well-constructed crite-
ria) has pitfalls. In the junior/senior clinic, for example, there
is a danger that students will perform well overall but have
widespread deficiencies in one or two areas. In such a case,
the fact that most teams earned A's and B's for the semester
would imply that students in the junior/senior clinic are meet-
ing the desired learning outcomes, when in reality there is a
need for specific improvement. As part of the pilot assess-
ment program, faculty went through the eighteen indicators,
one by one, and examined the level of performance dem-
onstrated by each team with respect to each indicator.
Through this process, specific problem areas were uncov-
ered even when the overall student performance was ob-
jectively very good.

BL
ent of
gree..

e expl.

e deter

e cons
ave co

cs duri

using

again

Chemical Engineering Education

Faculty members were
E 6 asked to assess the effective-
f Grading Rubrics ness of the rubrics. Table 6
.4 strongly agree) indicates that the faculty
clearly felt the rubrics were
Mean Response useful in improving fairness
ain the expectations 3.80 and linking the grading to the
learning objective. In our an-
nual assessment review, how-
rmine how my team 3.70
ever, the faculty decided that
it would be more valuable to
ider project issues 3.30 have the students do a mid-
nsidered. semester assessment of

ng the semester. 3.40 progress based on the rubrics.
Ideally, this should help both
grading rubrics. 3.70 the team and the professor
Snext semester. 3.80 identify areas that need im-
provement while there is still
time to adjust. Specific faculty
comments about the rubrics
included, "I felt much more confidant that my grade meant
something," and "I was able to use items from the rubrics to
drive my teams and help keep them on track."
Student comments about the rubrics were more mixed. They
were discussed with a focus group of seniors who had par-
ticipated in the clinic the previous year without the rubrics.
Their consensus was that the rubrics were useful and prob-
ably the correct way to do things, but one student asked,
"Couldn't you have waited until I graduated to implement
these?" The students also expressed concern that the rubrics
could be used as a basis for artificially lowering grades.
Ironically, part of the impetus for developing the rubrics
was a concern that grading that seemed arbitrary might lead
to grade inflation. In fact, more "A"s were given using the
rubrics than had been given the previous year when no ru-
brics were used. The faculty attributed the change to improve-
ment by the students. When we told the students what we
expected them to do, more of them did it.

FUTURE WORK

Although development of the above rubrics represents a
significant step forward, the results presented here describe a
pilot study. Substantial work remains to be addressed. Mean-
ingful assessment instruments must be developed to gauge
student and faculty perceptions of these criteria. Are the criti-
cal learning objectives addressed in these rubrics and are the

measurements accurate? Appropriate and meaningful
weightings must be developed for each of the behaviors. While
appropriate dress has been listed as an important part of the
project, one would be unlikely to argue that it is as signifi-
cant a learning objective as "drew meaningful and support-
able conclusions."

Once the rubrics have been optimized, the next major task
to be addressed is differentiating the performance of indi-
viduals from the performance of the team. It is possible that a
team could have one (or more) member who fully attains the
desired learning outcomes, but whose teammates fall sub-
stantially short of achieving those outcomes. Currently,
the Chemical Engineering Department at Rowan Univer-
sity uses a peer-assessment technique modeled after a pro-
cess described by Felder.126

Although this is a useful tool, it is somewhat over-reliant
on student evaluation of peers. Our experience indicates that
reasonably successful teams generally recommend an equal
distribution of points, while the recommendation of less suc-
cessful teams often are clouded with personal issues and re-
sentments. Because students tend to focus on grades rather
than on learning outcomes, their responses tend to be ho-
listic (person X should get 50% of the points) and more
about evaluation and grading, but less about achieving
specified learning outcomes.

A major thrust of this effort is to develop evidence-based
tools to complement the Felder survey, such that students
could more meaningfully assess the performance of their
teammates without defaulting to meaningless (e.g, "every-
one contributed equally"), hierarchial (e.g., "person X was
terrible," but no reasons provided), or personal assessments.
Moreover, the students will be required to cite specific evi-
dence linking their evaluations to the specific desired learn-
ing outcomes. Ideally, in addition to aiding the faculty mem-
ber in attempting to discern individual achievement from
a group experience, forcing an evidence-based approach
may help the students recognize the importance of the
learning outcomes.

REFERENCES

1. Gates, A.Q., P.J. Teller, A. Bernat, N. Delgado, and C.K. Della-
Pinna, "Expanding Participation in Undergraduate Research Us-
ing the Affinity Group Model," J. Eng. Ed., 88(4), 409 (1999)
2. Kardash, C.M., "Evaluation of an Undergraduate Research Expe-
rience: Perceptions of Undergraduate Interns and Their Faculty
Mentors," J. Ed.Psychology, 92, 191 (2000)
3. Zydney, A., J.S. Bennett, A. Shahid, and K. Bauer, "Impact of Un-
dergraduate Research Experience in Engineering," J. Eng. Ed.,
91(2), 151 (2002)
4. Guzzo, R.A., and M.W. Dickson, "Teams in Organizations: Re-
cent Research on Performance and Effectiveness," Ann. Rev. of
Psychology, 47, 307 (1996)
5. Katzenbach, J.R., and D.K. Smith, The Wisdom of Teams: Creat-
ing the High Performance Organization, Harvard Business School

Press, Boston, MA (1993)
6. Byrd, J.S., and J.L. Hudgkins, "Teaming in the Design Labora-
tory," J. Eng. Ed., 84(4), 335 (1995)
7. Boyer Commission on Education of Undergraduates in the Re-
search University, Reinventing Undergraduate Education: A Blue-
print for America's Research Universities, New York, NY (1998)
8. King, R.H., T.E. Parker, T.P. Grover, J.P.Gosink, and N.T.
Middleton, "A Multidisciplinary Engineering Laboratory Course,"
J. Eng. Ed., 88(3), 311 (1999)
9. Barr, R.E., P.S. Schmidt, T.J. Krueger, and C.Y. Twu, "An Intro-
duction to Engineering Through an Integrated Reverse Engineer-
ing and Design Graphics Project," J. Eng. Ed., 89(4), 413 (2000)
10. Heshmat, A.A., and A. Firasat, "Hands-On Experience: An Inte-
grated Part of Engineering Curriculum Reform," J. Eng. Ed., 85(4),
327(1996)
11. Schmalzel, J., A.J. Marchese, and R. Hesketh, "What's Brewing
in the Engineering Clinic?" Hewlett Packard Eng. Ed., 2(1), 6
(1998)
12. Newell, J.A., D.K. Ludlow, and S.P.K. Sternberg, "Progressive
Development of Oral and Written Communication Skills Across
an Integrated Laboratory Sequence," Chem. Eng. Ed., 31(2), 116
(1997)
13. Van Orden, N., "Is Writing an Effective Way to Learn Chemical
Concepts?" J. Chem. Ed., 67(7), 583 (1990)
14. Fricke, A.C., "From the Classroom to the Workplace: Motivating
Students to Learn in Industry," Chem. Eng. Ed., 33(1), 84 (1999)
15. Newell, J.A., A.J. Marchese, R.P. Ramachandran, B. Sukumaran,
and R. Harvey, "Multidisciplinary Design and Communication: A
Pedagogical Vision," Int. J. Eng. Ed., 15(5), 376 (1999)
16. Angelo, T.A., and K.P. Cross, Classroom Assessment Techniques:
A Handbook for College Teachers, 2nd ed., Jossey-Bass, Inc., San
Francisco, CA (1993)
17. Seat, E., and S.M. Lord, "Enabling Effective Engineering Teams:
A Program for Teaching Interaction Skills," J. Eng. Ed., 88(4),
385 (1999)
18. Lewis, P., D. Aldridge, and P. Swamidass, "Assessing Teaming
Skills Acquisition on Undergraduate Project Teams," J. Eng. Ed.,
87(2), 149 (1998)
19. Young, V.L., D. Ridgway, M.E. Prudich, D.J. Goetz, and B.J.
Stuart, "Criterion-Based Grading for Learning and Assessment in
the Unit Operations Laboratory," Proc. 2001 ASEE Nat Meet.,
Albuquerque (2001)
20. Walvoord, B.E., and V.J. Anderso'n ... . Grading: A Tool for
Learning and Assessment, Jossey-Bass, Inc., San Francisco, CA
(1998)
21. Terenzinis, PT., andE.T. Pascarella, Howl ... C .... Students:
Findings and Insights from Twenty Years ofResearch, Jossey-Bass,
Inc., San Francisco, CA (1991)
22. Woods, D.R., "Team Building: How to Develop and Evaluate In-
dividual Effectiveness in Teams," Workshop at 2000 American
Institute of Chemical Engineering (AIChE) National Meeting, Los
Angeles, CA (2000)
23. Banta, T.W., J.P. Lund, K.E. Black, and F.W. Oblander, Assess-
ment in Practice, Jossey-Bass, Inc., San Francisco, CA (1996)
24. Olds, B.M., and R.L. Miller, "Using Portfolios to Assess a ChE
Program," Chem. Eng. Ed., 33(2), 110 (1999)
25. Newell, J.A., K.D. Dahm, and H.L. Newell, "Rubric Development
and Inter-Rater Reliability Issues in Assessing Learning Out-
comes," Chem. Eg. Ed., 36(3), 212 (2002)
26. Kaufman, D.B., R.M. Felder, and H. Fuller, "Accounting for Indi-
vidual Effort in Cooperative Learning Teams," J. Eng. Ed., 89, 2
(2000) 5

Winter 2004

classroom

TEACHING ENGINEERING COURSES

WITH WORKBOOKS

YASAR DEMIREL
V,,,,i, Polytechnic Institute and State University * Blacksburg, VA 24060

S society expects that a modem college education will
turn out students who are analytical, intellectually cu-
rious, culturally aware, employable, and capable of
leadership.'ll Some important skills needed for all degree pro-
grams are problem solving, communication (written and oral),
team or group work, learning, and information processing
and technology. Instructors feel rewarded and satisfied
when they sense that they have made a difference in the
life of a student.[1]
All institutions of higher education emphasize that teach-
ing is important and give high priority to developing learn-
ing and teaching strategies that focus on promoting students'
subject-specific skills, knowledge, understanding, critical per-
spective, and intellectual curiosity.[2-14] Some of the strate-
gies are active and cooperative learning,[3,111 problem- or case-
based learning,112,131 and teaching through inquiry.[14] Active
and cooperative learning is one of the most frequently used
teaching methodologies.1-171] Development of new learning
and teaching methodologies should not be interpreted as
an obstacle to the research activity of a faculty member
and should be fully consistent with the university's re-
search strategy.[1s]
As Kennedy['1 suggests, new faculty members soon dis-
cover that effective lectures are hard to develop and deliver
and take much longer to prepare than they anticipated. Effec-
tive teaching incorporates forms of creativity that are not usu-
ally thought of as research but which actually analyze, syn-
thesize, and present knowledge in new and effective ways.1,17]
Traditional methods of learning and teaching embrace lec-
tures, seminars, workshops, and classes, as well as various as-
signments that require the use of books, handouts, handbooks,
and periodicals. As the student advances, incorporation of com-
puters and information technology such as "BLACKBOARD"
are developed. Currently, laptop computers are becoming com-

pulsory, and some courses are delivered entirely through the
use of computers and information technology with supporting
assignments. Some believe that the Internet has the potential of
replacing face-to-face teaching, but most courses still use the
chalkboard and verbal communication, and teaching and learn-
ing methods remain the responsibility of instructor and students.
It is widely recognized that students don't learn as much as
we try to teach them. Their native ability, their background, and
the match between their learning styles and the instructors' teach-
ing styles determines the level of learning.1171 To maximize the
level of their learning, we have to improve the effectiveness of
our teaching since, as instructors, we cannot do much about
their ability or background.117,19-211
Ineffective teaching can cause some students to drop courses,
lose self-confidence after getting bad grades, change majors, or
in the worst case, change to another institution or give up col-
lege altogether. Negative feedback of this nature can also nega-
tively impact future enrollment in engineering degree programs.
To address this problem, two trial workbook projects have
been introduced in two sophomore engineering courses at Vir-
ginia Tech: 1) introduction to chemical engineering thermody-
namics, and 2) chemical engineering simulations. This study
presents a first-hand experience with the preparation, use,
and assessment of workbook projects that are integrated

YasarDemirel is a visiting professorin the De-
partment of Chemical Engineering at Virginia
Tech. He received his PhD from the University
of Birmingham, England. He teaches senior
design, thermodynamics, transport phenomena,
and simulation. His long-term research focus is
coupled physical and biological systems and
stability analysis. He is the author of
Nonequilibrium Thermodynamics: Transport
and Rate Processes in Physical and Biological
Systems, published by Elsevier.

� Copyright ChE Division ofASEE 2004

Chemical Engineering Education

with class group work and the Internet teaching/learning
platform BLACKBOARD.

LEARNING AND TEACHING STYLES
In addition to theory, equations, and words, engineering
students are encouraged to work with course material that
includes real-world applications, pictures, diagrams, and dem-
onstrations.[191 An effective teaching technique should engage
students actively, stimulate a sense of enquiry, and encour-
age them to teach one another.[6-8,14] For example, group work,
which is widely used in science and engineering educa-
tion,[11,17,20,21] promotes problem-based learning and active par-
ticipation, which can lead to a deep learning that is more likely
to be retained. In group-work activity, two or three students
can apply a newly learned concept to solve a short problem
or to prepare a short essay.
Learning styles involve verbal or visual input modality,
sensing or intuitive perception, active or reflective process-
ing, and sequential or global understanding of course mate-
rial.[171 On the other hand, teaching styles involve an
instructor's emphasis on factual or theoretical information,
visual or verbal presentation, active or reflective student par-
ticipation, and sequential or global perspective. Learning and
teaching st. ls-' - :: :-1 are summarized in Table 1. Felder and
Si I cTi nui-1 emphasize, however, that these dimensions of
learning and teaching styles are neither unique nor compre-
hensive. Balances in various learning styles vary among stu-

TABLE 1
Learning and Teaching Styles[17,22,231

Learning Styles
Input Modality * Visual learners: Prefer to see
graphs, diagrams, flow charts,
plots, schematics
* VerbalLearners: Prefer
explanations (oral or written)

Teaching Stles
Presentation * Visual: Gr

* Verbal: Le
discussion

Perception * Sensing Learners: Focus on Content * Concrete:
sensory input, practical, observant
* Intuitive Learners: Focus on * Abstract:
imaginative and conceptual work, theoretical
theory, and models
Processing * Active Learners: Process actively Student * Active: Stl
think out loud, and like working Participation discuss
in groups
S * Learners: Process * Passive: S
introspectively, work quietly, like and listen
thinking and working alone or in
pairs

Understanding * SequentialLearners: Function in
continual steps and steady progress,
like analysis
* Global Learners: Need whole
picture to function, initially slow,
like synthesis

Perspective *Sequentia
progression

* Global: C
and relevan

A properly prepared workbook
makes the content of a textbook more visible,
extractable, and relevant for an application
or process. The instructor prepares the
workbook with all the essential verbal
and visual learning elements by using
the designated textbook, reference
books, and the publishers' web sites.

dents and depend on the field or their background. For ex-
ample, a student may be equally sensing and intuitive or one
of these learning styles may be dominant.
A student will learn more when teaching is done in his or
her preferred style.[17,24,25] For example, if teaching targets both
the visual and verbal learners, there is a good possibility that
learning is enhanced for the whole group. Felder and Brent1171
have suggested that there is a mismatch between learning and
teaching styles since most students are visual and sensing
learners but 90-95% of the content for most courses is verbal
and most instructors are intuitive learners. Such a mismatch
must be addressed for teaching to be effective.[17,22-251

PREPARING AND
WORKING WITH WORKBOOKS
A properly prepared workbook makes the content of a text-
book more visible, extractable, and relevant
for an application or process. The instructor
prepares the workbook with all the essential
verbal and visual learning elements by using
the designated textbook, reference books, and
the publishers' web sites. The verbal elements
aphs, Diagrams include all theory and analysis, definitions,
synthesis, and related applications. Figure 1
cture, reading, (next page) shows a typical page from a
workbook prepared for the thermodynamics
Factual course. The visual elements have most of the
related graphs, diagrams, schemes, configu-
Conceptual, rations, symbols for process flow diagrams
and streams, algorithms, flowcharts, tables,
pictures, figures, schematics, plots, analogies,
dents talk and and data. All the predetermined homework
assignments come from the textbook and ap-
tudents watch pear with small spaces allocated to each ques-
tion. The example problems, homework
problems, and group work are prepared to
relate the verbal and visual elements to each
l: Step-by-step other in an effective way. Most verbal ele-
ments are presented with bullets and in cat-
ontext and egorized boxes. Some of the visual and ver-
ce bal elements are deliberately left incomplete
or missing so the instructor and students can

Winter 2004

complete them together in the classroom. The quality of a
workbook depends on the instructor's experience, the
textbook's organization, the level of the course, and feed-
back from the students.
The instructor delivers the lecture with an overhead pro-
jector and transparencies of the workbook pages, joining
the verbal and visual elements of teaching. Students are
exposed to the workbook pages on the screen while they
work on them. Problem solving practices are performed
in the blank spaces allocated within the workbook. Be-
fore assigning homework questions, they are briefly dis-
cussed (see Figure 3).
In the presentation, all the related verbal and visual ele-
ments support each other and hence stimulate active stu-
dent participation, easy understanding, and relating the con-
cepts to applications. Lecturing with the workbook incor-
porates group work on a newly introduced topic by solving
a short problem or preparing short essays. This stimulates
teamwork and results in the students teaching one an-
other.[20,21] In addition to the group work, the BLACK-
BOARD multi-user education platform is used with the
workbook to provide supplemental course material, assign-
ments, useful sites, text objectives, test solutions, announce-
ments, and communications.

THE WORKBOOK TRIALS
Two workbooks were prepared and distributed to the ChE
students at Virginia Tech during the first lecture meeting of
two fundamental engineering courses. Although it was not
applied in this trial, the Felder index of learning st kIc - I or
any similar assessment study would be helpful for assess-
ing learning styles of students and for preparing small study
groups. Most of the students were sophomores, with small
numbers of juniors and seniors in both the courses. The
first workbook had 97 pages and was prepared for the text-
book Introduction to Chemical Engineering Thermodynam-
ics27] for the thermodynamics course. Some typical pages
completed in the classroom from this workbook can be seen
in Figures 2 to 4.
In Figure 2, the names of four thermodynamic potentials
are given in separate boxes. In an attached box, the system
is also defined as a closed system. All the primary proper-
ties of pressure P, volume V, temperature T, internal energy
U, and entropy S are related to each other in the boxes.
After completion, the boxes serve as visual elements con-
taining the related expressions for a well-defined system.
In the textbook, this same information is spread out and
may necessitate more time and effort for the students to

Figure 2. A typical thermodynamic-workbook page with
completed boxes for explaining the relations for thermo-
dynamic properties and derivations of the Maxwell rela-
tions.

Thermodynamic properties of fluids
* Property relations
System Properties
Homogeneous Internal energy Enthalpy Helmholtz energy Gibbs energy
fluid
with constant
composition
(closed
system)

* Maxwell relations

Exact differential equation ofa function F(x,y): dF = - dx + jF dy
\8xjy \ady

dU=TdS-PdV dH= TdS+ VdP dA= -PdV-SdT dG=VdP-SdT

* Enthalpy and entropy as functions of T and P
H = H(T,P)
Enthalpy
S= S(T,P)
Entropy

Figure 1. A typical workbook page for the
thermodynamics course.

Thermodynamic properties of fluids ___
* Property relations
System PrimarY Properties P, VT U, S
Homogeneous Internal energy, U Enthalpy, Helmholtz energy, Gibbs energy, G
fluid A
with constant AU=dQ4dW dH=-rdS+VJPdA - P
composition J G - V -JP- dT
(closed JU=TJS-P \ H U= PV 04
system) A = -T GS G 14 -Ts

* Maxwell relations

Exact differential equation of a function F(x,y): dF - dx -- dy odlc I
;^ = \.Sx~y [dy)^ aUrenX Y
d-F= M ilx+ N dy cr'rtwn
Cross relahbc . (

* Enthalpy and entropy as functions of TandP Cp c T -j. (- P T)VdP

Enthalpy - )a, ;- T - , v- D

SE n tS d= (P
Entropy 7f S I(s\ Sp r. __ - 3TD

(a) zT f+o; -0 Cp c p
21-T 2r T

Chemical Engineering Education

fully understand it. On the same workbook page, one of the
applications from the property relations has been demon-
strated through derivations of the Maxwell relations. This
associates a new concept with an application. The property
relations for enthalpy and entropy are further demonstrated
in a categorized way in the boxes.
The first part of Figure 3 relates the key expressions on
generalized correlations for liquids to the figure for reduced
density taken from the textbook. A short period of time for
group work follows this introduction so the students can find
the molar volume of ammonia at 310 K. The workbook con-
tains the selected homework problems from the textbook.
Before assigning them, they are briefly discussed, with em-
phasis on the critical points in the allocated boxes for each
question. This enables students to start their homework as-
signments with little or no outside help. Also, they will be
able to access the problems in the right location in the
workbook when they wish to review the course material
and the related problems.
Figure 4 starts with background information on vapor-liq-
uid equilibrium calculations. In the following box, three col-

Generalized correlation for liquids
* Racket equation for estimating the molar volumes of saturated liquids

( V T 0-2857
All i-r
Vst~ c^

p, V P2

, * - 7 -_

310 . | -!

- I
-- --

HW#3: 3'8; 3.32; 3.35; 3 45

;�3.53$

3.8 ld I rA : o C) T

Figure 3. A Zworkbook page containing a figure and home-o
N]: 4.9 ^V= ; / Jz 22ST2 C/^Si

Figure 3. A workbook page containing a figure and home-
work problems to be assigned from the textbook for the ther-
modynamics course.

umns identify the type of calculations, the variables to calcu-
late, and the variables specified for bubble point calculations
using the gamma-phi method. The box is related to the block
diagram underneath, which indicates how to start, proceed
with, and finish the calculations by using Equations 14.8 and
14.10 from the textbook, supplied in the box above. The block
diagram and equations are taken from the textbook and pro-
vide the necessary connections between the text and the dia-
gram. Therefore students will not be distracted by searching
for these equations when learning the block diagram.
The other workbook has 84 pages and was prepared for the
textbook Numerical Methods for Engineers,[28] used in the
simulation course. Figures 5 and 6 (next page) show some
typical pages completed in the classroom from this work-
book. In Figure 5, matrix operations are introduced with an
emphasis on multiplication of matrices. This concept is ex-
plained with a figure using the indices of coefficients matrix
and the two vectors for unknowns and constants related to
each other with the arrows. Next to that box, the computer
code for multiplication is supplied.

VLE calculations V . - . P ; ,
* The gamma/phi formulation ofVLE calculations Vt (P-P)]

i e RTX

Calculations Calculate Given
BUBLP x�_ x prsat x T
y, = (14.8) P = X (14.10)

Read /, {v,}, constants.
Set allt ', = 1.0. Calc. (yj by Eq. (14.8i .
Evaluate (P;}, {(y,). Evaluate . {'",
Cal. P by Eq. 0410).
No\

Print,,!.,) Is 6p
igiurel 14.1: Block dingrain for the calculation BuBI. P.

* Study Example 14.2 P r T

S L P-P1) 2. & IZ- &rvf-e

IJ [g( C(p -) Pt ( /A) T

Figure 4. A typical thermodynamic workbook page on va-
por-liquid phase equilibrium calculations completed in the
classroom. From the flowchart shown above, the steps of the
algorithm of bubble point calculations are discussed in the
classroom.

Winter 2004

For applying the rule of multiplication, a short group work
is carried out first and then linear algebraic equations are rep-
resented in matrix form. This form is constructed in a set of
two linear algebraic equations, and a 2-by-2 coefficients
matrix is created. Following this, the concept of inverse ma-
trix in introduced.
Figure 6 demonstrates the introduction of optimization.
Here, the concept of extremum is related to minimum and
maximums of a continuous function with some visual ele-
ments of figures immediately following. Later, the golden-
section search is explained with the dimensions from an old
Greek temple.
Some of the anticipated benefits of the workbooks are
* A detailed syllabus is an integrated part of the
workbook and helps the students jointly and effec-
tively use the textbook and workbook.
* It provides students with objective and vision state-
ments, main definitions, graphs, diagrams, and data in
a more apparent and categorized way than the
textbook (see Figures 2 and 3). It presents the course
material as a package of verbal and visual elements
and helps reach the students with various learning

Matrix operating rules
* Addition of two matrices

* Multiplication ofmatrices [03 9 [A] " g a,.

SUBROUTINE Hmmlt (a, b, C, m, n. 1)
DO -,1, [Alnxm [B]mxt = [C]nxt
Inlsrlor dime ,�os
00D k = 1. m areequal
sum = sum + ( i,k) - b(kQj) rmulpmllcon
END 00 isp" END 00he dimensions olthe rsLt
END 00
Fig. PT3.4 pseudocode to multiply an n by m matrix [A], by and m by I matrix [B]

SWe c an represent LAE in matrix-4 13 2[A {X} = {B consisting

* We can represent LAE in matrix form: [A]{{X) = (B) consisting

* Matrix of coefficients
* Vector of constants
* Vector of unknowns
02 Xt-Z2 =-

I " , t 1
;> XI + -,2

InVr A ' i a X

LA -

tpt

styles. This leads to effective use of the textbook.
* It makes note-taking easy and provides more time for
the students' critical thinking and interactions with the
instructor. This enhances deep understanding of the
course material.
* It reduces the mismatches among the teaching/
learning styles of the instructor, textbook, and
students and increases the visual elements, hence
stimulating effective teaching and learning.

* Working on the workbook with the instructor stimu-
lates the students' interest as the instructor and
students unfold the missing visual and verbal ele-
ments in the right location and moment.
* It provides easy access to definitions, analyses,
applications, synthesis, graphs, diagrams, figures,
tables, data, and worked and tested examples leading
to an effective learning and review of the course
material.
* It provides the homework assignments with brief
descriptions in boxes to relate them to the concepts of
the chapter.

___________ I

Optimization (One-dimensional unconstrained optimization)
* Optimization involves finding a value of x that yields an
minimum of a functionf(x) ; ()I
SI

extremum, either a maximum or
0'( )--o�

* Golden-section search: general-purpose, single variable optimization technique
Cx- ' x x{t 4 + 2.

Maximum

t Second
67 R iteration '

0JS 2 , I),

Figure 6. A completed page for optimization in the work-
book for the simulation course.

Chemical Engineering Education

Figure 5. A completed page on the matrix operations from
the workbook for the simulation course.

ASSESSMENT OF THE WORKBOOKS
Proper assessment of the workbooks is essential for mea-
suring their true level of effectiveness and developing the
best procedure for a particular course. Therefore, a workbook
will gain a level of maturity only after it is tried with an as-
sessment study. It is the author's intention to seek, through a
research proposal, a true assessment study from professional
organizations such as the Center for Excellence in Under-
graduate Teaching and the Center for Survey Research at
Virginia Tech. Only after such an assessment study will the
true effectiveness of workbook methodology be known.
Table 2 displays a preliminary questionnaire prepared by
the author, along with responses in percentages for the ther-
modynamic and simulations courses carried out after twelve
weeks with the workbooks. All the questions are treated with
the same weight. For the thermodynamics course, 47 students

TABLE 2
Preliminary Questionnaire for Assessment of the Workbo
1-disagree; 2-tend to disagree; 3-tend to agree; 4-agree; 5-not app

Thermodynamics
1 2 3 4 5
1 You have used WB in previous courses 75 10 2 0 13
2 WB contains a detailed syllabus 0 0 17 81
3 WB contains subject schedule from the textbook 0 4 13 77
4 WB provides objective, mission, and vision statements 0 0 23 73
5 WB provides related chapter and section readings 0 13 36 49
6 WB provides subject-related examples and homework problems 0 2 0 96
7 WB provides concepts, definitions, and working equations 0 2 19 79
8 WB enhances problem-based learning 0 4 23 71
9 WB enhances subject-specific skills and deep understanding 0 4 43 51
10 WB enhances problem-solving skills 0 17 36 45
11 WB makes it easy to locate subjects, definitions, and applications 0 4 30 64
12 WB relates a subject to data, tables, diagrams and figures 0 0 13 85
13 WB facilitates easy course note-taking 0 2 11 85
14 WB facilitates effective review of subjects and related problems 0 0 30 68
15 WB reduces mismatches between learning and teaching styles 2 4 51 39
16 WB reduces mismatches between textbook and instructor styles 0 2 47 49
17 WB offers a balanced teaching for various learning styles 0 6 45 45
18 WB encourages regular attendance 6 9 36 45
19 WB stimulates active learning 4 6 45 43
20 WB stimulates group work 0 9 42 49
21 WB facilitates higher grades from the tests 0 13 34 49
22 WB facilitates higher grades from the assignments 0 0 19 77
23 WB does not replace the textbook 4 32 19 45
24 WB stimulates effective use of the textbook 4 11 40 43
25 With group work and blackboard, WB becomes more effective 2 11 47 36
26 Overall, WB is beneficial in effective learning 2 0 26 68

responded and for the simulations course, 31 students re-
sponded.
The following responses deserve reviewing:
* Around 94% of students agree or tend to agree that
the workbook enhances problem-based learning, subject-spe-
cific skills, and deep understanding
* Around 90% of them agree or tend to agree that the
workbook reduces mismatches between learning and teach-
ing styles and offers a balanced teaching for various learning
styles
* Around 85% of the students agree or tend to agree
that the workbook stimulates active learning and group work
* Around 95% of the students agree or tend to agree
that overall, the workbook is beneficial in effective learning
Only 36% from the thermodynamics and 20% from the simu-
lation class disagree or tend to disagree
that the workbook does not replace the
textbook.

oks (WB) Some examples of written comments
liable on the questionnaire are:
SI do nothave any w,.. 'ii-.oi butI
1 Sinmlations % think the workbook is an excellent idea.

It helps a great deal in t, ii i, ,i is and
3 58 13 3 10 16 ,i,,,oi all the information in each
2 0 3 20 74 3 chapter
6 0 6 2371 0 * One way I think the workbook may
4 0 6 19 75 0 be improved is to carry examples not
2 0 13 39 48 0 included in the book. This would pro-
2 0 0 6 94 0 vide examples in addition to other
0 0 0 23 77 0 problems given in the book. Many times
2 0 3 45 52 0 I have already done book examples by
2 0 6 52 42 0 the time we get to them in class.
2 0 6 35 59 0 * Sometimes space becomes too
2 0 0 42 58 0 small or notes become a little confus-
2 0 0 19 81 0 ing; attendance still seems the student
2 0 6 9 85 0 responsibility. Overall, I believe the
workbook is a great learning tool!
2 0 0 34 66 0
4 0 13 26 61 0 * I do not have i... r.,. %i- i because
2 0 6 32highly approve of the use of work-
book. It gives the students time to re-
4 0 6 32 62 0
flect on what is going on in the class
4 3 3 32 62 0 instead of just blindly copying down
2 3 13 42 42 0 notes. I encourage all teachers to adopt
0 0 9 35 56 0 the workbook, which causes positive
4 0 3 49 42 6 interactions between student and
4 0 0 35 65 0 teacher
0 0 20 33 47 0 * Workbook allows instructor to go
2 0 6 35 59 0 over topics very quickly because notes
4 0 3 32 65 0 are already in front of you. I think it
4 0 3 16 81 0 would be more useful to go over each
concept in detail and make sure every-

Winter 2004

one understands. The workbook also closely mirrors the book.
If you don't understand the book, you probably will not un-
derstand the workbook.
* I really like the workbook. It makes the information a lot
more clear and cuts out all the messy derivations and extra-
neous information, so we can understand the concepts then
go back to look at it.
* The workbook is a good idea and an excellent study tool.
* The workbook is amazing! It condenses textbook into more
meaningful and useful notes; makes more dittic ult concepts
easier to understand. You can tell instructor cares about the
student learning and appreciation of the subject matter Needs
no improvements, love the workbook!
*I really like the workbook. It helps me greatly in the course
and I wish more teachers would use it. I understand more
and have learned a lot.
* Workbook helps keep me organized, and allows me to pay
attention in class and actively interact with what is going on.
It motivates learning, reviewing and comprehension. I wish
workbook would be used in all of my classes.

CONCLUSIONS
Preparation of the workbook, using it along with the group-
work activity and BLACKBOARD, and a preliminary as-
sessment study have been presented here. The assessment
study indicates that the workbook methodology may be an
effective strategy in learning and teaching. Most of the engi-
neering students who took the courses in thermodynamics
and simulation have found the workbooks beneficial in un-
dergraduate engineering teaching. This is mainly because the
workbooks, integrated with group work and BLACKBOARD,
may help reduce the mismatches in teaching and learning
styles, and may increase interactions between students and
faculty, hence stimulating active and collaborative learning
and effective teaching. The workbook trials need a true and
coordinated assessment study, however, in order to measure
their level of effectiveness in reducing the mismatches be-
tween learning and teaching styles.

ACKNOWLEDGMENTS
The author thanks Professor Erdogan Kiran for reading the
manuscript and providing constructive comments, and the
students Samuel F. Ellis and Michele A. Seiler for their help
in preparing Table 2.
Note: Electronic sample copies ofworkbooks for the courses
on thermodynamics and simulations are available in PDF
format upon request to the author at ydemirel@vt. edu.

REFERENCES
1. Kennedy, D., Academic Duty, Harvard University Press, Cambridge,
MA (1999)

2. Streveler, R.A., B.M. Moskal, R.L. Miller, and M.J. Pavelich, "Center
for Engineering Education: Colorado School of Mines," J. Eng. Ed.,
90(3), 381 (2001)
3. McCowan, J.D., "An Integrated and Comprehensive Approach to En-
gineering Curricula. Part Two: Techniques," Int. J. Eng. Ed., 18(6),
638 (2002)
4. Raju, G.K., and C.L. Cooney, "Active Learning from Process Data,"
AIChEJ., 44(10), 2199 (1998)
5. Haller, C.R., V.J. Gallagher, T.L. Weldon, and R.M. Felder, "Dynam-
ics of Peer Education in Cooperative Learning Workgroups," J. Eng.
Ed., 89(3), 285 (2000)
6. Felder, R.M., and R. Brent, "Effective Strategies for Cooperative Learn-
ing," J. Coop. Collaborat. Coll. Teach., 10(2), 63 (2001)
7. Kaufman, D.B., R.M. Felder, and H. Fuller, "Accounting for Individual
Effort in Cooperative Learning Teams," J. Eng. Ed., 89(2), 133 (2000)
8. Gokhale, A.A., "Collaborative Learning Enhances Critical Thinking,"
J. Tech. Ed., 7(1), 23 (1995)
9. Johnson, D., Active Learning: Cooperation in the College Classroom,
Bugess Publishing Company, New York, NY (1991)
10. Gosser, D.G., and V. Roth, "The Workshop Chemistry Project: Peer-
Led Team Learning," J. Chem. Ed., 75(2), 185 (1998)
11. Bean, J.C., Engaging Idea. The Professor's Guide to Integrating Writ-
ing, Critical Thinking, andActive Learning in the Classroom, Jossey-
Bass Publishers, San Francisco, CA (2001)
12. Kulonda, D.J., "Case Learning Methodology in Operations Engineer-
ing," J. Eng. Ed., 90(3), 299 (2001)
13. Fogler, H.S., and S.E. Leblanc, Strategies for Creative Problem-Solv-
ing, Prentice Hall, Engelwoods Cliffs, NJ (1994)
14. Buch, N.J., and T.F. Wolf, "Classroom Teaching Through Inquiry," J.
Profess. Issues Eng. Ed. Practice, 126(3), 105 (2000)
15, Morrell, L., R. Buxeda, M. Orengo, and A. Sanchez, "After So Much
Effort: Is Faculty Using Cooperative Learning in the Classroom?" J.
Eng. Ed., 90(3), 357 (2001)
16. Stice, J.E., R.M. Felder, D.R. Woods, and A. Rugarcia, "The Future of
Engineering Education IV. Learning How to Teach," Chem. Eng. Ed.,
34(2), 118 (2000)
17. Felder, R.M., and R. Brent, "Effective Teaching: A Workshop," Winona,
MN, October 5-6 (2001)
18. Wankat, P., "Tenure for Teaching," Chem. Eng. Ed., 37(1), 1 (2003)
19. Felder, R.M., "How to Survive Engineering School?" Chem. Eng. Ed.,
36(3), 30 (2002)
20. Felder, R.M., "It Goes Without Saying," Chem. Eng. Ed., 25(3), 132
(1991)
21. Felder, R.M., "How About a Quick One?" Chem. Eng. Ed., 26(1), 18
(1992)
22. Felder, R.M., and L.K. Silverman, "Learning and Teaching Styles in
Engineering Education," Eng. Ed., 78, 674 (1988)
23. Felder, R.M., "Reachingthe Second Tier: Learning and Teaching Styles
in College Science Education," J. College Sci. Teach., 23, 286 (1993)
24. Fitch, B., and A. Kirby, "Students' Assumptions and Professors' Pre-
sumptions: Creating a Learning Community for the Whole Student,"
College Teach., 48(2), 47 (2000)
25. Felder, R.M., G.N. Felder, and E.J. Dietz, "The Effects of Personality
Type on Engineering Student Performance and Attitudes," J. Eng. Ed.,
91(1), 3 (2002)
26. Felder, R.M., and B.A. Soloman learning/felder.htm> (2003)
27. Smith, J.M., H.C. Van Ness, and M.M. Abbott, Introduction to Chemi-
calEngineering Thermodynamics, 6th ed., McGraw-Hill, Boston, MA
(2001)
28. Chapra, S.C., and R.P. Canale, Numerical Methods for Engineers, 4th
ed., McGraw-Hill, Boston, MA (2002) O

Chemical Engineering Education

	Table of Contents
	Chuck Eckert of the Georgia Institute...
	The University of Alabama
	A fluidized bed adsorption laboratory...
	Positions available
	On the application of Durbin-Watson...
	Teaching electrolyte thermodyn...
	Book review
	Changing times and paradigms
	Nanostructured materials: Synthesis...
	The fuel cell: An ideal ChE undergraduate...
	Incorporating green engineering...
	Top ten ways to improve technical...
	UOP-Chulalongkorn University industrial-university...
	Use of conceptests and instant...
	Rubric development for assessment...
	Teaching engineering courses with...

MILLISECOND	CLASS.METHOD	MESSAGE
0	sobekcm_page_globals.constructor
0	sobekcm_page_globals.constructor	Application State validated or built
0	sobekcm_database.verify_item_lookup_object
0	sobekcm_page_globals.constructor	Navigation Object created from URI query string
0	sobekcm_database.verify_item_lookup_object
0	sobekcm_page_globals.display_item	Retrieving item or group information
0	sobekcm_page_globals.get_entire_collection_hierarchy	Retrieving hierarchy information
0	sobekcm_assistant.get_entire_collection_hierarchy
0	cached_data_manager.retrieve_item_aggregation
0	cached_data_manager.retrieve_item_aggregation	Found item aggregation on local cache
0	item_aggregation_builder.get_item_aggregation	Found 'all' item aggregation in cache
0	system.web.ui.page.page_load (ufdc.page_load)
0	sobekcm_page_globals.constructor.on_page_load
0	html_echo_mainwriter.add_style_references	Adding style references to HTML
0	html_echo_mainwriter.add_text_to_page	Reading the text from the file and echoing back to the output stream
96	html_echo_mainwriter.add_text_to_page	Finished reading and writing the file