<%BANNER%>
  • TABLE OF CONTENTS
HIDE
 Front Cover
 Table of Contents
 Chemical Engineering at Bucknell...
 Cooperative Weblab: A Tool for...
 A Simple Explanation of Comple...
 A Blended Approach to Problem-Based...
 A Synchronous Distance-Education...
 A Survey of the Role of Thermodynamics...
 Integrating Academic and Mentoring...
 Development of Problem Sets for...
 Teaching Technical Writing in a...
 Random Thoughts: Hard Assessment...
 Two Undergraduate Process Modeling...
 Introducing DAE Models in Undergraduate...
 Microfluidics in the Undergraduate...
 Back Cover
































Chemical engineering education
http://cee.che.ufl.edu/ ( Journal Site )
ALL VOLUMES CITATION THUMBNAILS DOWNLOADS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/AA00000383/00182
 Material Information
Title: Chemical engineering education
Alternate Title: CEE
Abbreviated Title: Chem. eng. educ.
Physical Description: v. : ill. ; 22-28 cm.
Language: English
Creator: American Society for Engineering Education -- Chemical Engineering Division
Publisher: Chemical Engineering Division, American Society for Engineering Education
Publication Date: Winter 2010
Frequency: quarterly[1962-]
annual[ former 1960-1961]
quarterly
regular
 Subjects
Subjects / Keywords: Chemical engineering -- Study and teaching -- Periodicals   ( lcsh )
Genre: serial   ( sobekcm )
periodical   ( marcgt )
 Notes
Citation/Reference: Chemical abstracts
Additional Physical Form: Also issued online.
Dates or Sequential Designation: 1960-June 1964 ; v. 1, no. 1 (Oct. 1965)-
Numbering Peculiarities: Publication suspended briefly: issue designated v. 1, no. 4 (June 1966) published Nov. 1967.
General Note: Title from cover.
General Note: Place of publication varies: Rochester, N.Y., 1965-1967; Gainesville, Fla., 1968-
 Record Information
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 01151209
lccn - 70013732
issn - 0009-2479
sobekcm - AA00000383_00182
Classification: lcc - TP165 .C18
ddc - 660/.2/071
System ID: AA00000383:00182

Downloads

This item has the following downloads:

Chemical Engineering at Bucknell University, Csernica, Gross, Jablonski, Raymond, and Vigeant ( PDF )

Cooperative Weblab: A Tool for Cooperative Learning in Chemical Engineering in a Global Environment, Le Roux, Reis, de Jesus, Giordano, Cruz, Moreira, Nascimento, and Loureiro ( PDF )

A Simple Explanation of Complexation, Elliott ( PDF )

A Blended Approach to Problem-Based Learning in the Freshman Year, Rossiter, Petrulis, and Biggs ( PDF )

A Synchronous Distance-Education Course for Nonscientists Coordinated Among Three Universities, Smith, Baah, bradley, Sidler, Hall, Daughtrey, and Curits ( PDF )

A Survery of the Role of Thermodynamics and Transport Properties in Chemical Engineering University Education in Europe and the USA, Ahlstrom, Aim, Dohtn, Elliott, Jackson, Jaubert, Macedo, Pokki, Reczey, Victorov, Zilnik, and Economou ( PDF )

Integrating Academic and Mentoring Support for the Development of First-Year Chemical Engineering Students in Hong Kong, Ko and Chau ( PDF )

Development of Problem Sets for K-12 and Engineering on Pharmaceutical Particulate Systems, Savelski, Slater, Del Vecchio, Kosteleski, and Wilson ( PDF )

Teaching Technical Writing in a Lab Course in Chemical Engineering, Lombardo ( PDF )

Random Thoughts: Hard Assessment of Soft Skills, Felder and Brent ( PDF )

Two Undergraduate Process Modeling Courses Taught Using Inductive Learning Methods, Soroush and Weinberger ( PDF )

Introducing DAE Models in Undergraduate and Graduate Chemical Engineering Curriculum, Mandela, Sridhar, and Rengaswamy ( PDF )

Microfluids in the Undergraduate Laboratory: Device Fabrication and an Experiment to Mimic Intravascular Gas Embolism, Jablonski, Vogel, Cavanagh, and Beers ( PDF )


Table of Contents
    Front Cover
        Page i
    Table of Contents
        Page 1
    Chemical Engineering at Bucknell University
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
    Cooperative Weblab: A Tool for Cooperative Learning in Chemical Engineering in a Global Environment
        Page 9
        Page 10
        Page 11
        Page 12
    A Simple Explanation of Complexation
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
    A Blended Approach to Problem-Based Learning in the Freshman Year
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
    A Synchronous Distance-Education Course for Nonscientists Coordinated among Three Universities
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
    A Survey of the Role of Thermodynamics and Transport Properties in Chemical Engineering University Education in Europe and the USA
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
    Integrating Academic and Mentoring Support for the Development of First-Year Chemical Engineering Students in Hong Kong
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
    Development of Problem Sets for K-12 and Engineering on Pharmaceutical Particulate Systems
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
    Teaching Technical Writing in a Lab Course in Chemical Engineering
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
    Random Thoughts: Hard Assessment of Soft Skills
        Page 63
        Page 64
    Two Undergraduate Process Modeling Courses Taught Using Inductive Learning Methods
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
    Introducing DAE Models in Undergraduate and Graduate Chemical Engineering Curriculum
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
    Microfluidics in the Undergraduate Laboratory: Device Fabrication and an Experiment to Mimic Intravascular Gas Embolism
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
    Back Cover
        Page 89
Full Text












Chemical engineering education









4p SPECIAL ISSUE:


Cooperative Weblab: A Tool for Cooperative Learning in ChE in a Global Environment (p. 9)
Le Roux. Reis. de Jesus. Giordano. Cruz, Moreira, Nascimento. Loureiro
A Simple Explanation of Complexation (p. 13)
a Elliott
.'
A Blended Approach to Problem-Based Learning in the Freshman Year (p. 23)
SRossiter, Petrulis. Biggs
A Synchronous Distance-Education Course for Nonscientists Coordinated Among Three Universities (p. 30)
c Smith, Baah, Bradley, Sidler, Hall, Daughtrey, Curtis
SA Survey of the Role of Thermodynamics and Transport Properties in ChE University Education
in Europe and the USA (p. 35)
c Ahlstrom, Aim, Dohrn, Elliott, Jackson, Jaubert, Macedo, Pokki, Reczey, Victorov, Zilnik, Economou

'4--
U-'

0 Development of Problem Sets for K-12 and Engineering on Pharmaceutical Particulate Systems (p. 50)
ul Savelski, Slater, Del Vecchio, Kosteleski, Wilson
3 Integrating Academic and Mentoring Support for the Development of First-Year Chemical Engineering
a Students in Hong Kong (p. 44)
E Q Ko, Chau
< a)
E Two Undergraduate Process Modeling Courses Taught Using Inductive Learning Methods (p. 65)
. C Soroush, Weinberger
.> " Random Thoughts: Hard Assessment of Soft Skills (p. 63)
U E Felder, Brent

Teaching Technical Writing in a Lab Course in Chemical Engineering (p. 58)
a U Lombardo
C O0
SMicrofluidics in the Undergraduate Laboratory: Device Fabrication and an Experiment to Mimic
c 3 Intravascular Gas Embolism ip. 81)
1- Jablonski, Vogel, Cavanagh. Beers
._ -c Introducing DAE Models in Undergraduate and Graduate Chemical Engineering Curriculum (p. 73)
E c
w (a Mandela, Sridhar. Rengaswamy
U Editorial:Tough Decisions in Tough Times (Inside front cover)
E Davis




Bucknell University













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

EDITOR
Tim Anderson

ASSOCIATE EDITOR
Phillip C. Wankat

MANAGING EDITOR
Lynn Heasley

PROBLEM EDITOR
Daina Bri di,. 1 .. . ...

LEARNING IN INDUSTRY EDITOR
William J. Koros, Georgia Institute -I ii. ,....'..

- PUBLICATIONS BOARD -

* CHAIRMAN *
John P. O'Connell
University of Virginia
* VICE CHAIRMAN *
C. Stewart Slater
Rowan University
* MEMBERS *
Lisa Bullard
North Carolina State
Jennifer Curtis
University of Florida
Rob Davis
University of Colorado
Pablo Debenedetti
Princeton University
Dianne Dorland
Rowan
Stephanie Farrell
Rowan University
Jim Henry
University of Tennessee, ( h.....,,... ,,
Jason Keith
Michigan Technological University
Suzanne Kresta
University of Alberta
Steve LeBlanc
University of Toledo
Ron Miller
Colorado School of Mines
Lorenzo Saliceti
University of Puerto Rico
Stan Sandler
University of Delaware
Margot Vigeant
Bucknell University


Vol. 44, No. 1, Winter 2010


Chemical Engineering Education
Volume 44 Number 1 Winter 2010


> DEPARTMENT
2 Chemical Engineering at Bucknell University
Csernica, Gross, Jablonksi, Raymond, and I �,1.,
> CLASS AND HOME PROBLEMS
50 Development of Problem Sets for K-12 and Engineering on Pharmaceutical
Particulate Systems
Savelski, Slater, Del Vecchio, Kosteleski, and Wilson
> CLASSROOM
44 Integrating Academic and Mentoring Support for the Development of
First-Year Chemical Engineering Students in Hong Kong
Ko and Chau
65 Two Undergraduate Process Modeling Courses Taught Using Inductive
Learning Methods
Soroush and Weinberger
> RANDOM THOUGHTS
63 Random Thoughts: Hard Assessment of Soft Skills
Felder and Brent
> LABORATORY
58 Teaching Technical Writing in a Lab Course in Chemical Engineering
Lombardo
81 Microfluidics in the Undergraduate Laboratory: Device Fabrication and
an Experiment to Mimic Intravascular Gas Embolism
Jablonski, Vogel, Cavanagh, and Beers
> SPECIAL SECTION: AICHE CENTENNIAL CELEBRATION
9 Cooperative Weblab: A Tool for Cooperative Learning in Chemical
Engineering in a Global Environment
Le Roux, Reis, de Jesus, Giordano, Cruz, Moreira, Nascimento, and
Loureiro
13 A Simple Explanation of Complexation
Elliott
23 A Blended Approach to Problem-Based Learning in the Freshman Year
Rossiter, Petrulis, and Biggs
30 A Synchronous Distance-Education Course for Nonscientists Coordinated
Among Three Universities
Smith, Baah, Bradley, Sidler, Hall, Daughtrey, and Curtis
35 A Survey of the Role of Thermodynamics and Transport Properties in
Chemical Engineering University Education in Europe and the USA
Ahlstrom, Aim, Dohrn, Elliott, Jackson, Jaubert, Macedo, Pokki,
Reczey, Victorov, Zilnik, and Economou
> CURRICULUM
73 Introducing DAE Models in Undergraduate and Graduate Chemical Engi-
neering Curriculum
Mandela, Sridhar, and Rengaswamy

Inside front cover Editorial: Tough Decisions in Tough Times, Davis

CHEMICAL ENGINEERING EDUCATION (ISSN 0009-2479) is published quarterly by the Chemical Engineering
Division, American Societyfor 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 0 2010 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 ChEDivision,ASEE, which body assumes no responsibility for them. Defective copies replaced ifnotified within
120 days of publication. Write for information on subscription costs and for back copy costs and availability. POSTMASTER:
Send address changes to Chemical Engineering Education, Chemical Engineering Department., University of Florida,
Gainesville, FL 32611-6005. Periodicals Postage Paid at Gainesville, Florida, and additional post offices (USPS 101900).

1










Ief f department


ChE at...


Bucknell University


= CELEBRATING 100 YEARS


JEFFREY CSERNICA
MICHAEL GROSS
ERIN JABLONKSI
TIMOTHY RAYMOND
MARGOT VIGEANT
Bucknell University began as "The University at
Lewisburg" in 1846, during a time when transpor-
tation advances in the region-a bridge crossing
the Susquehanna River, a "turnpike" to the west, and a
vibrant canal system providing an easy link to east coast
cities - were transforming the area from what Philadelphians
called the "wilds of Pennsylvania" into a goods trading and
transport hub. The original charter describes a small liberal
arts university, offering a curriculum including ancient lan-
guages, logic, and rhetoric, as well as geology, trigonometry,
and astronomy.
In response to the industrial revolution and a rapidly
growing and increasingly urbanized U.S. population, trustee
William Bucknell and others sought to expand the scope of
the university in the late 1800s. The free-standing Chemical
Laboratory dedicated in 1890 ushered in a period of increased
attention to instruction in the physical sciences, and paved
the way for the B.S. in chemical engineering, first offered in
1909. Four graduates received the degree four years later, and
the currently enrolled class of 2012, our 100th, contains 29


students. In 1923, chemical engineering graduate Katherine
Owens became the first woman to receive an engineering
degree at Bucknell. Women have maintained a strong pres-
ence in the program, and make up 35% of the current student
population.
John Harris, university president during the early 20th
century, was a supporter of the new engineering programs
but actively discouraged them from seeking formal college
or even department status. He instead argued that engineer-
ing courses were "part of a liberal education, which rests
upon the substrata of mathematics, physics, drawing, and
chemistry." While department and college standing eventu-
ally came, the concepts of integration and synergism between
engineering and the liberal arts and sciences that Bucknell
embraces today (and which are now publicly recognized at
many other institutions) clearly had their roots in those early
days. The university remains primarily undergraduate (a
small master's program exists within the department), and
advantages of Bucknell's liberal arts atmosphere have been
stressed in engineering recruiting brochures since the 1920s.
Currently such materials state that Bucknell's "liberal arts
environment offers diverse learning opportunities, encourages
critical thinking, and supports engineering problem solving
in a societal context."
In 1922 the first wing of the Charles Dana Engineering
Building was constructed, although most chemical engineer-


� Copyright ChE Division of ASEE 2010


Chemical Engineering Education




























































Building Montage
Evolution of the engineering facilities on campus. Upper left (1925): first wing of the Dana building marks foray
from primary campus into adjacent farmland; upper right (1945): a second wing and connector complete Dana's
horseshoe geometry; bottom (2005): the current complex includes a renovated Dana Building (now seen with
a complete third story) and the attached Breakiron Building.


ing laboratory activities remained in the Chemical Laboratory
until 1939, when the second wing and connecting central sec-
tion of the Dana building were completed. Many elements of
the new chemical engineering laboratories were based on the
recommendations of the Engineers' Council on Professional
Development, the engineering accreditation body at that time,


and the chemical engineering program was formally accred-
ited two years later. A major renovation and further expansion
of this building was completed in 1984. In Fall 2004, space for
engineering studies was further increased by 45% through the
construction of the Lauren Breakiron Engineering Building,
which is connected to the original Dana building.


Vol. 44, No. 1, Winter 2010










CURRICULUM
Consistent with former President Harris's "substrata," the
current chemical engineering curriculum contains a broad
required science base of mathematics, chemistry, physics, and
biological and materials sciences. The program's general educa-
tion component is based on that of the arts and sciences college,
and contains a "global and societal perspectives" requirement.
Wihilik. ngin ic iing course tides and sequences would be largely
familiar to those in the discipline, particular emphases through-
out the curriculum are placed on realistic open-ended problems,
project management, experimental practice, professional and
interpersonal development, and independent learning, as il-
lustrated in the elements discussed below.
Laboratories Almost all core chemical engineering classes
contain a concurrent laboratory component, and students are
thereby continually relating theory and practice throughout
their four years in the program. These multiple laboratories
create many added opportunities for practice and formal
training in teamwork, communication, and the solution of
open-ended problems.
Dedicated instructional laboratories currently exist with the
following topical specializations:
* Analytical
* Bioprocess
* Fluid Flow
* Heat Transfer
* Kinetics
* Materials Science
* Polymer Science
* Process Control
* Unit Operations
Exploring Engineering Bucknell requires all incoming engi-
neering students to take the introductory Exploring Engineering
course in their first semester. The course is taught by a team of
nine faculty, two of whom are chemical engineers, and has an
enrollment of approximately 200. Since 2002, the course has
been taught in a modular fashion based on a seminar format.
Students take three discipline-specific, three-week seminars in
groups of approximately 25. The seminars are student-selected
and use a specific topic to introduce engineering concepts and
provide windows into the specific disciplines. Each seminar
contains a laboratory component in which students complete
an open-ended design project. In 2009 there were two seminars
representing chemical engineering-one based on producing
a modified ChemE Car, and the other on designing and mass
producing a superior material for sneaker soles.
Introductory ChE Course In the first-year Chemical En-
gineering Principles course, students immediately get their
hands dirty in the unit operations laboratory and begin their
exposure to pilot-scale equipment. The course is centered on
a team-based cooperative-learning environment. Five two-
week projects are proposed to teams of students, where they
4


are expected to address a problem with hand calculations,
simulation software, and a complex laboratory experiment.
In addition to formulation and presentation of a solution to
the open-ended problems, course objectives include formally
developing teamwork and professional skills.[1]
Senior Design In the past decade, the department has moved
from a traditional two-semester design sequence (in which both
semesters focused on a single simulation-based design) to one
in which the second semester is spent in a practical design ex-
perience, working on real problems posed by external clients. 2]
Building on the earlier senior design work, this second-semester
course, "Project Engineering," requires students to solve a
problem with a more narrow focus and produce a tangible
result for an actual client. Students work with problem defini-
tion, project management, genuine budget and time constraints,
and real deadlines and expectations that are meaningful beyond
the classroom. Project work can include product or process
development at the bench- or pilot-scale, prototyping, or
implementation in an industrial setting. For example, students
have designed novel coatings for a producer of athletic mats,
specified equipment for a pharmaceutical manufacturer to im-
prove process flexibility, designed portable heating equipment
for U.S. Army MREs ("Meals Ready-to-Eat"), and optimized
fermentation performance for a start-up microbrewery.
Seminar The mid-1990s saw the initial development of
the department's spring seminar course-weekly visits and
presentations by practicing professionals -into which each
student in the department is automatically enrolled. Beyond
coverage of various technical topics, the exposure to careers,
assignments, and problems of practicing chemical engineers
enhances our students' understanding of professional growth
and development beyond their college years. Many of these
benefits are realized through the seminars themselves, but
are supplemented through networking that occurs at associ-
ated speaker lunches with student sub-groups, and informal
speaker visits to classes and laboratories. In conjunction with
newly articulated program outcomes associated with EC2000,
we later targeted and arranged speakers over time to explore
specific "perspectives," such as those associated with ethical,
environmental, and societal issues.
Specialization Beyond the core curriculum, students can
tailor their experience to their interests, which may include
study abroad (see below), a five-year dual degree (second
degree in liberal arts or management, or a combined B.S./M.S.
in chemical engineering), or pursuit of a concentration within
the major (biological, environmental, materials, process).
Research Undergraduate research opportunities are con-
sidered a vital element of the program, and about 2/3 of
graduates will have spent at least one semester conducting an
"independent study" research project, working directly with a
faculty mentor. Many such students are encouraged to serve as
co-authors on archival publications, and to present their work
professionally at AIChE meetings and other venues.
Chemical Engineering Education










































Department faculty: (left to right, front row) Tim Raymond, Ryan Snyder, Jeff Csernica, Bill King, Kat Wakabayashi;
(back row) Mike Gross, Erin Jablonski, William Snyder, Mike Hanyak, Brandon Vogel, James Maneval, Margot Vigeant,
James Pommersheim; on sabbatical and not pictured: Mike Prince.


FACULTY
Currently the department carries 12 full-time faculty lines.
Consistent with expectations of a predominantly undergradu-
ate liberal arts institution and with responsibilities as described
in Bucknell's faculty handbook, our faculty are committed
to teaching excellence, and provide close personal attention
to students in the classroom, laboratory, and beyond. Many
faculty are both close followers of and contributors to devel-
opments in engineering education and pedagogy.
At the same time, however, Bucknell espouses a teacher-
scholar ideal, and faculty are expected to remain on the cutting
edge of their discipline through active research programs. In the
past four years, major outside funding has been obtained for: de-
velopment of a nanofabricaton laboratory; study of atmospheric
aerosols (NSF CAREER); enhancing engineering education;
and special instrumentation (atomic force microscope, polymer
extruder and pulverizer). Research activities are conducted
primarily with undergraduate students, and this feature brings
its own set of special challenges as well as rewards.
Jeff Csernica joined the faculty in 1989 after receiving
his Ph.D. at the Massachusetts Institute of Technology. His
work on transport in polymers with Colgate-Palmolive led to


a cosmetic composition patent. He has acted as coordinator
of the college's first-year Exploring Engineering course, and
is currently serving as department chair.
Michael Gross joined the faculty in 2007 after receiving
his Ph.D. at the University of Pennsylvania. His research
currently focuses on the development of solid oxide fuel cell
electrodes. He is a member of the CAChE task force develop-
ing modules that bring fuel cell technology into the traditional
chemical engineering undergraduate curriculum.
Michael E. Hanyak, Jr., holds a Ph.D. from the University
of Pennsylvania and began teaching at Bucknell in 1974.
Computer-aided engineering and courseware development
have been the focal points of his professional career, and ac-
complishments include publications and federal grants in the
area of systemic engineering education reform.
Bill King joined the department in 1983 and served as de-
partment chair from 1986 to 1998. His research interests are
in biotransport related to cancer treatment, and he currently
holds a dual appointment with the new and recently accredited
Biomedical Engineering Department at Bucknell.
Erin L. Jablonski came to the department in 2004, fol-
lowing a Ph.D. from Iowa State and a post-doctoral position


Vol. 44, No. 1, Winter 2010































Above, Chemical Engineering's "fluids wall," circa 1950,
which was dismantled during the 1984 renovation of the
Dana Engineering Building. Right, students working in
1939's new unit operations laboratory.

at NIST. Recently, she has used microfluidic techniques to
study diffusion in hydrogels, and has reported on novel ways
of integrating classroom and laboratory instruction through
project-based design activities.
Jim Maneval joined the department in 1991 after com-
pleting his Ph.D. at the University of California, Davis. His
research focuses on the use of NMR methods in systems
of engineering interest and the development of models for
transport processes in complex and multiphase materials. His
teaching interests include design and applied mathematics.
James Pommersheim retired in 2007 after teaching in
the department for more than 40 years. Early in his career he
introduced the applied math and transport theory courses into
the curriculum. Research leaves include time at NIST, NASA,
Penn State, ExxonMobil, and Occidental. In 2005 and 2006
he taught the senior design course at Syracuse University.
Mike Prince came to Bucknell in 1989 after receiving his
Ph.D. from U.C. Berkeley. His research examines the con-
nection between instructional practices and student learning
outcomes in engineering programs. He is co-director of the
National Effective Teaching Institute and active in a number
of initiatives to improve engineering education.
Tim Raymond began his time at Bucknell as an undergrad-
uate in 1993 and returned in 2002 as a member of the faculty
after completing his Ph.D. at Carnegie Mellon University.
He is active in both research and teaching of the physics and
chemistry of atmospheric aerosols, and is heavily involved
in AIChE local and student sections.
Ryan Snyder joined the faculty in 2009, following indus-
trial experience at Air Products, a Ph.D. from U.C. Santa
Barbara, and a post-doctoral position at Eli Lilly. His research
6


focuses on product and process design of structured (often
crystalline) products, such as those commonly found in phar-
maceuticals, foods, and nanomaterials.
William J. Snyder came to Bucknell in 1968 after complet-
ing a Ph.D. at Penn State and a post-doctoral assignment at
Lehigh University. His research focuses on thermodynamics,
polymer solutions, and pedagogy. His primary teaching areas
include chemical reaction engineering, fluid flow, thermody-
namics, and design.
Margot Vigeant is in her 11th year on the Bucknell faculty,
and has an active research program in chemical engineering
pedagogy, focusing on misconceptions in thermodynamics.
She was honored with the 2009 Fahien Award from the Chemi-
cal Engineering Division of ASEE, and is spending this year
as a part-time associate dean.
Brandon M. Vogel received a Ph.D. from Iowa State Uni-
versity and was an NRC postdoctoral fellow at NIST before
coming to Bucknell in 2007. His teaching interests are in
biomaterials, bioprocess engineering, and applied statistics,
and his research focuses on the synthesis of new materials to
detect, target, and treat disease.
Katsuyuki Wakabayashi became a member of the faculty
in 2007 after his Ph.D. work at Princeton and post-doctoral
position at Northwestern. His research focus in polymer hy-
brid processing provides for hands-on student work in both
class and independent-study settings. He is currently the
coordinator for the department's graduate program.
Chemical Engineering Education












































The 29 students of Bucknell's 100th chemical engineering class (2012), with insets of the four inaugural class members
(left to right, Joseph McKeague, Alexis Keen, Herman Zehner, and Hartley Powell).


STUDENTS AND SPECIAL ACTIVITIES
Bucknell students and faculty are engaged in many program
activities beyond those associated with formal curricular
and scholarly pursuits. The sampling below highlights the
program's ideals regarding broad student experience and
professional involvement, and its commitment to engineer-
ing education.
In 2007, Bucknell's AIChE student chapter organized and
hosted its second Mid-Atlantic Regional Student Conference.
One year later, they co-hosted the National Student Confer-
ence associated with AIChE's Centennial Annual Meeting
in Philadelphia. Bucknell's then-student chapter president
Danielle Woodhead was chosen to speak on behalf of the
entire national student body at the black-tie Centennial Gala
Banquet. AIChE selected Bucknell as an Outstanding Student
Chapter for 2008-2009.
Also regarding AIChE, Bucknell's ChemE Car teams have
distinguished themselves by including students from electrical
and mechanical engineering and chemistry, and they recently
placed first at the Mid-Atlantic regional competition in 2009.
Also of note, students Damon Vinciguerra and Ben Aldrich
both became heavily involved with national planning activi-
Vol. 44, No. 1, Winter 2010


ties, and as juniors now serve as regional liaisons for AIChE
representing the Mid-Atlantic and the Western regions,
respectively.
During February, the engineering college at Bucknell
celebrates National Engineers Week through a spirited yet
lighthearted competition among its six degree programs.
Many events highlight creativity outside of the technical
norm, and include creation of departmental banners, poetry,
and videos that often poke fun at engineering majors and their
stereotypes. The chemical engineering department has won
11 times in the last 20 years of competition.
Students in the department are strongly encouraged to
have a study-abroad experience while completing the degree
program. In the mid 1980s, Professor (and long-serving de-
partment chair) Robert Slonaker was instrumental in estab-
lishing the study-abroad programs in the engineering college.
Year-long and semester programs are now well established
for chemical engineers, all of which allow students to finish
their degree in the standard four-year time frame. Currently
about 25% of our students participate in these programs dur-
ing the academic year, and locations include England, Ireland,
Australia, New Zealand, and Spain.










Near right,
AIChE Student
Chapter President
Danielle Woodhead
addresses the
audience at the
2008 AIChE
Centennial Gala
Banquet in Phila-
delphia. Far right,
ChemE Car team
members at the
2009 Mid-Atlantic
regional com-
petition. Bottom
right, Professor
Mike Prince at
Bucknell's "How
to Engineer Engineering Education" summer work-
shop, which draws faculty from around the country.


Students interested in studying abroad but who choose
not to pursue a semester-long program can elect a summer
course called "Engineering in a Global and Societal Con-
text." This three-week course, conducted abroad and led by
a team of Bucknell engineering faculty, was modeled after a
similar successful program in Bucknell's humanities division.
While the course location, topical specialization, and instruc-
tors change yearly, the course's common denominator is a
focus on foreign local infrastructure, economies, and culture,
and how these affect engineering practice and technology
policy. Typical are visits to universities, manufacturing sites,
and government project locations, as well as talks from lead-
ers of industry, government, and academia. Recent sites have
included countries in both Europe and South America.
Under the direction of Erin Jablonski and funded in part
through her recent grant from the National Science Foundation,
Bucknell in 2008 began hosting a summer Engineering Camp
for pre-college students interested in engineering. In July 2009
the camp hosted 52 students (8th- 11th grade) from eight states
and the District of Columbia, and instructors included faculty
from each engineering department at Bucknell. The week-long
residential camp gives students a unique on-campus experience,
and activities include presentations, hands-on experiments, and
mini-design projects on topics ranging from nanotechnology
to urban planning to biomechanics.
Bucknell has recently completed its 8th consecutive summer
offering of the teaching workshop, "How to Engineer Engi-
neering Education." The event is directed by Mike Prince with
participation by several Bucknell engineering faculty including
Bill Snyder, Mike Hanyak, Margot Vigeant, and Tim Raymond.
It draws approximately 30 engineering faculty each year from
around the country and overseas. This hands-on workshop in-
troduces faculty to cutting-edge issues in instructional design,
active learning techniques, and best assessment practices.
8


SUMMARY
Bucknell's chemical engineering program was born and
continues to thrive at the unique intersection of technology
and the ideals of a liberal education. We look forward to con-
tinuing to build on those principles of the program founders,
and the subsequent 100 years of dedication and refinement by
many, as we embark upon our second century of commitment
and contributions to chemical engineering education.

REFERENCES
1. Hanyak, M., and T. Raymond, teaching g Material and Energy Bal-
ances to First-Year Students Using Cooperative Team-Based Projects
and Labs," Proceedings of the ASEE Annual Conference, American
Society for Engineering Education (2009)
2.. Vigeant, M., J. Maneval, W. Snyder, M. Hanyak, and M. Prince,
"Hands-On Chemical Engineering Senior Design: the Evolution from
Paper to Practice," Proceedings of the ASEE Annual Conference,
American Society for Engineering Education (2008) [
Chemical Engineering Education











r.I.1 AIChE special section


COOPERATIVE WEBLAB:

A Tool for Cooperative Learning in Chemical Engineering

in a Global Environment




G.A.C. LE Roux,1 G.B. REIS,2 C.D.F. DE JESUS,2 R.C. GIORDANO,2 A.J.G. CRuz,2
P.F. MOREIRA, JR.,1 C.A.O. NASCIMENTO,1 AND L.V. LOUREIRO1
SPolytechnic School, University of Sdo Paulo * 05508-900, Sdo Paulo, SP, Brazil
2 Federal University of Sdo Carlos * Sdo Carlos, SP, Brazil


T his paper describes the use of Weblabs -Web-based
experiments-for cooperative learning by students
working together from two different locations to
conduct experiments and write reports.
In 2004, the Sao Paulo State's Agency for Research De-
velopment (FAPESP) established a program to study and
develop the usage of technology and applications in advanced
Internet for research and educational purposes (KyaTera).
The platform is an optical high-speed packet network in-
terconnecting a number of laboratories, research institutes,
and universities in the state of Sao Paulo, Brazil. One of the
projects of KyaTera is the Cluster of Weblabs for Chemical
and Biochemical Process Engineering, which form a collab-
orative research between University of Sao Paulo (USP); the
Federal University of Sao Carlos (UFSCar); and University
of Campinas (UNICAMP). This project aims to develop a
set of real experiments, available through the Internet, for
chemical engineering students at the undergraduate level.
Other KyaTera projects and the annual report can be found
at .[1]

REVIEW AND PREVIOUS WORKS
The application of Weblab experiments in chemical engi-
neering education was first reported in 1998.[2] After this paper,
the same author published a series of articles on the subject
in the Proceedings of the American Society for Engineering
Education and in subsequent papers.[39] These articles show
the evolution of Weblabs' usage for academic purposes either
locally or internationally.


Unit operations and process control Weblabs for chemical
engineering education purposes were also developed.", 11] Only
more recently were collaborative Web-based experiments in
local environments implemented and presented.[12 13]

G.A.C. Le Roux is an associate professor of chemical engineering at
The University of Sao Paulo. He has research interests in mathematical
modeling, simulation, and process synthesis and control, with emphasis
on parameter estimation methods and system identification.
G.B. Reisis a Ph.D. candidate in the Department of Electrical Engineering
at The University of Sao Paulo in Sao Carlos. He has a M.Sc. in chemical
engineering from The Federal University of Sao Carlos.
C.D.F. de Jesus is a process engineer at the Bioethanol Science and Tech-
nology Center. Dr. de Jesus received his Ph.D. in chemical engineering
from The Federal University of Sao Carlos. His main activities are process
modeling, software development, data acquisition of biotechnological
experiments, and biodiesel plants commissioning.
R.C. Giordano is a full professor of chemical engineering, and is head of
the Chemical Engineering Department and of the Laboratory for Develop-
ment and Automation of Bioprocesses (,www.ladabio.deq.ufscar.br.) at
The Federal University of Sao Carlos.
A.J.G. Cruz is an associate professor of chemical engineering at the Fed-
eral University of Sao Carlos. His research focuses on biotechnological
applications of computer-aided process engineering.
P.F. Moreira, Jr., is a research fellow at The University of Sao Paulo. His
roles include software and hardware development and system integra-
tion specialist.
C.A.O. Nascimento is a full professor of chemical engineering, and head
of the Center for Environmental Research and Training at The University
of Sao Paulo. Dr. Nascimento is also a member of the Environmental Com-
mittee of the Industrial Federation of State of Sao Paulo.
L.V. Loureiro is a professor of chemical engineering at The University of
Sao Paulo. Dr. Loureiro is also executive director of the Fulbright Commis-
sion in Brazil and has an extensive background in international educational
and graduate academic exchanges.
� Copyright ChE Division of ASEE 2010


Vol. 44, No. 1, Winter 2010










Two international workshops -
iLabs Workshop @ MIT, Jan.
24-27, 2005, in Massachusetts,
and Weblabs in Chemical Engi- r-----
neering, July 8, 2005, in Cam- air
bridge, England-were organized
by the Massachusetts Institute of N2
Technology and University of
Cambridge, respectively, with Mas flow
,. .ller.
representatives from many insti-
tutions from various countries,
e.g. Technologico de Monterrey,
Mexico; ENSIACET, France;
University of Leipzig, Germany;
and University of SAo Paulo,
Brazil. The purpose of the work- Figure
shops was to discuss Weblabs as
an educational tool, along with the potential developments
and outcomes.
A comprehensive set of Weblabs for chemical engineer-
ing student training has been developed at University
Leipzig by Ralf Moros, under supervision of Prof. Helmut
Rapp. Their work in the field was recently presented at
the 2nd International Workshop on e-learning and Virtual
and Remote Laboratories, in February 2008, at Potsdam,
Germany, at an event organized by The University of
Potsdam.

WEBLAB AND COOPERATIVE WEBLAB IN
CHEMICAL ENGINEERING
The term Weblab was first employed to describe real experi-
ments-in contrast to virtual ones-remotely operated via
the Internet. This type of experiment has been proposed and
implemented worldwide by some major universities. In this
article it is introduced a unique approach for Weblabs: Coop-
erative Weblab (CW). This new format for Weblabs promotes
intercultural experiences to students while also allowing them
to develop communication skills. CW experiments must be
performed by two groups of undergraduate students, in two
different locations.
The collaboration among the students is achieved by
gathering participants of two different Weblabs into working
groups that are asked to simultaneously solve a technical
problem, for which an experiment setup is available. This
procedure emulates challenges that will frequently take
place in their future professional lives. The CW also fosters
the learning of essential chemical engineering concepts.
Students from different Weblabs must work cooperatively
to achieve these goals.

CASE STUDY: CW BETWEEN USP AND UFSCAR
The CW concept was tested with two Weblabs of chemical
engineering departments, one at USP and the other at UFS-
10


1. Diagram of mass transfer experimental setup.

Car, both supported by KyaTera. These Weblabs are 250 km
apart and they are at two of the top five chemical engineering
programs in Brazil.

Mass Transfer Weblab at UFSCar
During the aerobic cultivation of microorganisms or cells
in tank bioreactors, the level of dissolved oxygen must be
kept high enough for the organisms to thrive. Thus, it is im-
portant for the education of chemical engineers to learn the
fundamentals of mass transfer herein involved, and to get
familiar with techniques that assess rates of oxygen transfer
from the gas phase into the liquid culture medium as well.
A scheme of the Weblab for mass transfer experiments is
presented in Figure 1.
The dissolved oxygen is removed from the liquid phase
by bubbling nitrogen into the medium. After reaching zero
oxygen concentration, the nitrogen flow stops and air flow
starts. An electrode probe measures the dissolved oxygen
(DO) in the liquid phase. The mass transfer coefficient is
represented by the parameter kLa that is estimated by fitting a
model for the change of the DO to the experimental data. The
experiment aims to calculate kLa values at different operating
conditions of air flow rate and stirrer speed employing the
gassing-out method.14 15]
The bioreactor is an aerated and stirred tank reactor. Gas
(air and nitrogen) under pressure is supplied to the sparger (a
ring with holes) located inside the reactor and above of the
impeller. The system consists of a stirrer with two Rushton
impellers. This impeller is typically a disc with 6 to 8 blades
designed to pump fluid into a radial direction. The Weblab was
built employing National Instruments hardware for data acqui-
sition and LabVIEW software as the supervisory system.
Figure 2 shows the main screen of the Weblab for mass
transfer experiments. The users can choose the experiment
operating conditions (air flow rate and stirrer speed) through
this screen.


Chemical Engineering Education


c-FP 2020 Data acquisition
National instruments equipment
r " ; , -. . -;, -;. - ;, - , -77 ;


Comp ter












Reactor Temperature Control Weblab at USP

The Continuous Stirred Tank Reactor (CSTR) is one of
the most simple and, at the same time, powerful devices in
the chemical industry. It involves many important aspects
of chemical engineering process (e.g., heat transfer, mass
transfer, chemical reactions) leading, in this experiment, to
the development of process control strategies ( pqi.ep.usp.br>).


-at-, kLa Experiment

LaDABio * 1 _1" I
Theory
The Bioreactor
Experiment
Download results






Data acquisition V

D- -- .Mi
S*:.......... ),
Comp- Fid PoiN t -. FP 20M-
t z, AO. D, ,,




Figure 2. Mass Transfer Weblab interface

Figure 2. Mass Transfer Weblab interface.


*st- p , .ui i PZ1. T, im Puaji . ,,Arrr i
r-v I-10"nr I ~ olF..rIrsF0h- re ts 'ri Plr- ol -1 1ea Cornlllllll[ PunIS I Load rm


The reactor is a jacketed glass vessel with 4.5 L capacity.
At the inner part of the reactor there is a stainless steel coil
connected to a 3,000 W thermal bath, a hot water source, a
temperature sensor, and a double helix impeller. Thejacket is
connected to a cold water source, at room temperature. The hot
and cold water flows can be controlled by electro pneumatic
valves. An electrical heater is plunged inside the reactor, to
simulate different reactions' thermodynamic behaviors. A
digital power module controls the electrical tension


supplied to the resistor.

The reactor temperature can be controlled by
varying the hot water flow to the coil or the cold
water flow to the jacket. Figure 3 shows a screen-
shot of the temperature control interface. This
screen also includes a Webcam to remind the users
that they are dealing with an actual setup and not
with a simulator.


PERFORMING THE CW EXPERIMENTS

To perform the CW experiments, two students at
UFSCar and two at USP form a group. An instructor
supervises the students at each Weblab. The group is
invited to define the experimental procedure and the
tasks of each member during the experiment. Students
communicate using videoconference software that
gives participants an improved sense of reality.


ermnr, l Br, \


P-OJLs PumL . F.l-(Junrl
J ma Ma ;qi rn4lr 11-
b MA ]..,L/ r. ,| '--"--
Ma . rn.Lip r.r.

If~, [ Pe lOt rLr.rn OlI l




V ne i ir. i
Stdit oo " "-



FLTD t.re CJnera LI d I �I
._.


3oten d COntrvl
..







5.4 59-


vwePl vessel
I ,"At ,JUl AD


el lat I sn,
VL hll
Ij.5( � ' iailL


Figure 3.
Reactor
Temperature-
Control
Weblab
interface.


IlhE Im. tma|


Vol. 44, No. 1, Winter 2010


I -r � I-I











During the experiments students exchange information and
opinions about the phenomena that take place. In the mass
transfer experiment, the main concerns are about the quality of
the mixing, the size of the bubbles, and the various problems
that arise during the experiment (e.g., bubbles blocking the tip
of the probe, high coalescence phenomena, and conditions of
inefficient mixing). In the reactor temperature control experi-
ment, a step change on the coil flow is performed to obtain
the response curve. The students must perform calculations
together during the experiment to apply the Cohen-Coon pro-
cedure to obtain the PID tuning parameters.[16] The set of PID
tuning parameters is implemented by closing the control loop
and is tested by different methods (e.g., set point changes and
disturbance rejection tests). Students are invited to test other
disturbances with the immersed resistor in the reactor.
A high level of interactivity takes place among the students
during the experiment. At the end of experiment, the students
must process the data and prepare the final group report. The
students are also invited to use the network to communicate
and prepare the report.

CONCLUSIONS

The CW experiments are a privileged learning environment
and are complementary to the experiments run in regular labs.
There are at least two positive aspects of the CW experiments
worth pointing out. The first is that the students can better
develop teamwork skills through quasi-professional life ex-
periences. In a CW experiment, they must work with formerly
unknown colleagues, with different engineering education,
to reach common goals. This is a big challenge, particularly
for undergraduates, accustomed only to interacting with
their engineering schoolmates and teachers. Furthermore,
the open-ended problems methodology adopted in the CW
experiments also invites the participants to have a stronger
interaction, fostering the teamwork skills. The other aspect
is the playfulness of remote control of real equipment and
interaction with colleagues in a distant location, in an envi-
ronment-the Internet-that is very familiar to the students.
It also stimulates them to perform the CW activities, e.g., the
experimental procedures and reports, with additional interest
and dedication for even well-known experiments.
The overall opinion of students was extremely positive
and encourages the increase of the number of participat-
ing institutions in the CW experiments. This will allow a
sound assessment of results in a shorter period of time. CW
experiments are valuable educational tools in a world with
an increasing interaction between diverse people in general
and, particularly, between professionals in the field. Special
effort is necessary to fit this kind of activity in the curricula


of chemical engineering programs, because it involves two or
more institutions. In the case of Weblabs in different countries,
some practical hurdles must also be faced and overcome,
such as the difference in time zones, calendars, and language.
Nonetheless, the impact of these issues is outweighed by the
potential improvement that this tool can bring to the chemical
engineering education.

ACKNOWLEDGMENTS
Authors would like to thank the SAo Paulo State's Agency for
Research Development (FAPESP) for its financial support.

REFERENCES
1. Kyatera ProgramAnnual Reports, available at
2. Henry, J., "Running Laboratory Experiments via the World Wide
Web," ASEE Annual Meeting, Seattle, WA, June 1998. Available at

3. Henry, J., "24 Hours, 7 Days Lab Experiments Access on the Web All
the Time,"ASEE Annual Meeting, St. Louis, MO, June 2000. Available
at
4. Henry, J., "Laboratory Remote Operation: Features and Opportuni-
ties," ASEE Annual Meeting, Charlotte, NC, June, 2001. Available at

5. Henry, J., "Using the Modern Chemical Engineering Laboratory
at a Distance," ASEEAnnual Meeting, Montreal, Quebec, Canada,
2002. Available at cfm?id=16833>
6. Henry, J., and C. Knight, Modern Engineering Laboratories at a Dis-
tance, Int. J. Eng. Ed.,19 (3), 403-408, 2003.
7. Henry, J., and H.M. Schaedel, "International Cooperation in Control
Engineering Education Using Online Experiments, European J. ofEng.
Ed., 30(2), 265 (2005)
8. Henry, J., and R. Zollars, "Introducing Reality Into Process Control
Classes," ASEE Annual Meeting, Portland, OR, 2005. Available at

9. Slater, C.S., R.P Hesketh, D. Daniel Fichana, J. Henry, A.M. Flynn,
and M. Abraham, "Expanding the Frontiers for Chemical Engineers
in Green Engineering Education," Int. J. Eng. Ed., 23(2), 309 (2007)
10. Shin, D., E.S. Yoon, K.Y. Lee, and E.S. Lee, "AWeb-Based, Interactive
Virtual Laboratory System for Unit Operations and Process System
Engineering Education: Issues, Design, and Implementation," Com-
puter and Chem. Eng., 26, 319-330 (2002)
11. Selmer, A., M. Goodson, M. Kraft, S. Sen, V. Faye McNeill, B.S.
Johnston, and C.K. Colton, "Performing Process Control Experiments
Across the Atlantic," Chem. Eng. Ed., 39(3) (2005)
12. Gillet, D., A.V. Ngoc, and Y. Rekik, "Collaborative Web-Based Ex-
perimentation in Flexible Engineering Education," IEEE Transactions
on Education 48(4), 696 November (2005)
13. Henry, J., and R. Zollars, "Learning-By-Doing and Communications
within a Process Control Class,"ASEE Annual Meeting, Chicago, IL,
2006. Available at cfm?id= 1888>
14. Shuler, M. L., and E Kargi, Bioprocess Engineering: Basic Concepts,
2nd Ed., Prentice Hall International Series, Upper Saddle River, NJ,
2002.
15. Blanch, H.W, and D.S. Clark, Biochemical Engineering, Marcel
Dekker, Inc., New York (1997)
16. Seborg, D.A., T.H. Edgar, and D.A. Mellichamp, Process Dynamics
and Control, John .11,. & Sons, New York (1989) 1


Chemical Engineering Education











r.I.1 AIChE special section


A SIMPLE EXPLANATION


OF COMPLEXATION











J. RICHARD ELLIOTT
The University ofAkron * Akron, OH44325-3906


It seems that the freshmen are onto me. As part of our
freshman "Chemical Engineering Computations" course,
the students can choose speakers for four half-lectures.
Last Spring, they chose me to give a presentation about hy-
drogen bonding. Rumor has it that they heard about a fanatical
thermodynamics professor with a soft spot for the topic of
hydrogen bonding. Their strategy was two-fold: first, to but-
ter this guy up by asking him to ramble on about his favorite
subject, and second, to scout this dangerous territory called
thermodynamics. In my turn, I saw this as a teachable moment.
If I made the presentation sufficiently accessible, they might
actually learn something about thermodynamics.
But what computational model can be accessible to fresh-
men in 25 minutes and explain hydrogen bonding and its role
in chemical and biomolecular engineering? The key thermo-
dynamic impact of hydrogen bonding is on the activity coef-
ficient, a dimensionless expression of the fugacity. Fugacity is
one of the most dreaded words in chemical engineering, even
among seniors. Therefore the presentation must very gently
focus first on introducing the activity coefficient, then on the
role of hydrogen bonding. Little scouting of my own revealed
that they were already performing flash computations. So they
knew about K = y /x,* but the only solution model they knew
was Raoult's Law (K1 = Pt//P). This, then, was my way in.
Most students are aware of the ethanol+water azeotrope and

� Copyright ChE Division ofASEE 2010
Vol. 44, No. 1, Winter 2010


that oil and water do not mix. Moreover, they are all aware
of the term "hydrogen bonding" from high school courses
in biology. From these familiar points of reference, I for-
mulated the following introduction, emphasizing qualitative
concepts and interactive computational exercises to appeal
to a broad range of learners at an early stage in their studies.
The approach emphasizes computations, since that is the
course's subject, but introduces the vocabulary of solution
thermodynamics.

A 5-MINUTE INTRODUCTION FOR FRESHMEN
TO THE LIMITATIONS OF RAOULT'S LAW
According to Raoult's law, the vapor mole fraction of
ethanol in water is greater than the liquid at all compositions.
Then distillation to gasohol should be no problem because it is
constantly enriching. But experiments show that yE < x when

Richard Elliott is a professor of chemi-
cal and biomolecular engineering at the
University of Akron, where he has taught
since 1986. He is a coauthor of the text
Introductory Chemical Engineering Ther-
modynamics, with Carl Lira of Michigan
State, published by Prentice-Hall. He is a
coauthor of more than 50 papers, princi-
pally focused on experimental measure-
ments, theory, and molecular simulation
of hydrogen bonding systems.

* Commonly used symbols are defined in the Nomenclature section.










xE > 0.9. Furthermore, Raoult's Law cannot explain liquid
phase separation, yet nobody would put a 10% water solution
into their gas tank. Hydrogen bonding helps to explain these
"non-idealities" and many more.
The reason for the azeotrope in the ethanol+water system
is that the water does not entirely "like" the ethanol. We
can characterize this disdain with "modified Raoult's Law"
(K1 = 1 P 1t/P), where m, is called the "activity coefficient."
If a little ethanol mixes with the water, the vast majority of
water molecules are surrounded by other water molecules so
they barely know the difference. In that case, Raoult's law
provides an adequate description for the water molecules.
Mathematically, this is represented by Yw, 1. But when water
is surrounded by 90% ethanol, it can get very uncomfortable.
We can relate to this kind of discomfort when we are in an
unfamiliar crowd. Measurements show that ', 2 at 90%
ethanol. From the perspective of ethanol at 90% concentration,
however, Raoult's Law is fine and K = PEat/P. Furthermore,
at 78 �C PEt = 1 bar, so KEZ 1 at ambient pressure. But Pwt
S0.55 at 78 �C, so Kw = wP~t/P z 1.1. This means that the
water becomes richer in the vapor than the ethanol, and distil-
lation fails. This causes the separation to make gasohol to be
less direct and more expensive both in terms of dollars and
in terms of energy efficiency. The students nod when I say
this, but five minutes have already elapsed.

DEVELOPING A COMPUTATION-BASED
INTRODUCTION TO HYDROGEN BONDING
In the remaining 20 minutes, I introduce a simplified ver-
sion of the "Modified Separation of Cohesive Energy Den-
sity" (MOSCED) model"l and apply it in an example. The
MOSCED model is a modification of Scatchard-Hildebrand
theory that separates the cohesive energy density into a
dispersion term, a polarity term, and two hydrogen bonding
terms (one for acidity and one for basicity).[1] Characterizing
acidity and basicity is the key to explaining hydrogen bond-
ing. Although less known than models like van Laar or Mar-


gules, MOSCED is better suited to an intuitive explanation
of hydrogen bonding, as detailed in the "rationale" section
below. Keep in mind that this presentation is for a course in
"Chemical Engineering Computations," so the students should
not be surprised to see a few equations. Then I assign three
homework problems. I refer to this model as the "simplified
separation of cohesive energy density" (SSCED) model. The
simplifications of SSCED are designed to convey key con-
cepts in a manner that is consistent with presentations through-
out thermodynamics and separations processes. Quantitative
precision is not necessary for this qualitative introduction, but
a computational model means that the students can "learn by
doing." In other words, the concepts, symbols, and vocabulary
become familiar as they practice their chemical engineering
computations. Even qualitatively, the interpretation of a term
like the binary interaction parameter of Scatchard-Hildebrand
theory (k ) is fundamental and intellectually challenging to
students. Learning its meaning and use at such an early stage
would be worthwhile.

ORGANIZATION OF THE MANUSCRIPT
With this background, the remainder of the manuscript can
be outlined. The section immediately after this one resumes
the presentation to students, closely following the notes for
the freshman lecture. The objective is that students should
understand hydrogen bonding sufficiently to anticipate the
sign of deviations from ideality and have some idea of its
magnitude. Students are incidentally exposed to the relevance
of hydrogen bonding in formulations, biofuels, distillation,
liquid phase separation, environmental science, and "political
intrigue." The interpretation that k < 0 indicates favorable
mixing can be illustrated graphically with the square-well
potential, and reinforced with Conceptining' 1 as demon-
strated in the assessment section below. Assessments also
show that students can quickly rank order solutions according
to their non-ideality as an outcome of this presentation. As
shown in the rationale section, the SSCED model reinforces
the interpretation of k and links it to the acidity and basicity


TABLE 1
Sample values of physical properties. 6, 6', a, and P all in I.I c'ni
T,(K) F i . I. o MW 298 6(J/cm3)1/2 a p 6'
Acetone 508.2 4.70 0.306 58 0.79 19.64 0.00 11.14 19.64
Benzene 562.2 4.90 0.211 78 0.87 18.73 0.63 4t 18.60
Chloroform 536.4 5.40 0.216 119.2 1.48 18.92 5.80 0.12 18.88
Ethanol 516.4 6.38 0.637 46 0.79 26.13 12.58 13.29 18.67
Iso-octane 544.0 2.57 0.303 114 0.70 14.11 0 0 14.11
Methanol 512.6 8.10 0.566 32 0.79 29.59 17.43 14.49 19.25
MTBE 497.1 3.43 0.266 88 0.74 15.17 0 7.40 15.17
Water 647.3 22.12 0.344 18 1.00 47.86 50.13 15.06 27.94
p-xylene 616.3 3.51 0.326 106 0.86 17.90 0.27 1.87 17.87
t This value was modified slightly from the value of Lazzaroni, et al. (2.24)


Chemical Engineering Education










that give rise to hydrogen bonding, while the Scatchard-Hil-
debrand model does not. Finer points about the advantages
and limitations of the SSCED model are also addressed
in the rationale section. Several of these finer points are
intended for enthusiasts of thermodynamics and hydrogen
bonding. I conclude with a brief review of the assessments
of student learning and a perspective on how students may
benefit from presentation of such a model at an early stage
in the curriculum. Altogether, the presentation illustrates the
current status of student preparedness, learning objectives,
interactive learning, and assessment in thermodynamics at
the 100th anniversary of AIChE.

RESUMING THE FRESHMEN PRESENTATION
WITH A SIMPLE COMPUTATIONAL MODEL
THAT ACCOUNTS FOR HYDROGEN BONDING
Resuming from the first five minutes of the lecture, the tech-
nical name for the factor, T,, is the "activity coefficient." When
, = 1 the situation of the i* component is "ideal." When , <
1, the i* component is unusually comfortable. Such a formula-
tion would make a great solvent if you had a nasty stain to
remove. When i > 1, the component is uncomfortable, like
the water in 90% ethanol. Finally, when m > 10, the compo-
nent "hates" its environment so much that it may separate, like
the water in gasoline. A very simple computational model can
describe all of these situations and help to design formulations
to achieve chemical engineering goals. It is,

RTlnm V, (1- 4)2 ( 6 - +2k1262 (1)

Where R = 8.314 J/mole-K, T is the temperature in Kelvins,
V =MW /pL is the liquid molar volume at 298K and I) =
x V /Xx V is the volume fraction, analogous to weight fraction.
The bracketed terms require some explanation. The term k12
is a correction factor that characterizes specific interactions,
principally hydrogen bonding. We discuss k12 later. The
other term addresses the modified solubility parameter, 8'.
If we assume for the moment that k12 = 0, then Ab' provides
a concise and quantitative measure of m. If Ab' = 0, then
the solution is ideal and Raoult's Law is fine. Otherwise, the
solution becomes non-ideal.
The solubility parameter is related to the energy density
of a compound. This energy can be quantified by the heat of
boiling. When you boil water, for example, molecules are
extracted from their congenial environment to a lonely vapor,
where they can share little energy with others. They prefer
to share energy. That is why you must add heat. More heat
must be added if they share more energy. If the same boiling
pot is used to characterize various compounds, then more
small molecules fit in it than large ones, and even more heat
is required. Therefore, it is the energy density that character-
izes how strongly a compound sticks to itself. This kind of
energy density is something quite different from the explosive


energy density of a compound like trinitrotoluene (TNT), so
we need a distinctive name for it. That name is the "cohesive
energy density," defined by,

62 = vU"/V J cm/ 3 (2

where Uvap is the energy of vaporization and the rationale for
squaring 8 is explained in the thermodynamics course. In the
absence of hydrogen bonding, 8 = 8'. Therefore, in terms of
8', discomfort of a component in solution is not caused by
dislike for the other components, but by a strong preference
for its own company. You may have heard that an extroverted
engineer is one who looks at your feet when he is talking to
you. Any introverted engineers in the room should relate to
this perspective on the definition of discomfort.
In the presence of hydrogen bonding, the hydrogen bond-
ing contribution must be separated from 8, hence the name
for this model as the SSCED model (simplified separation
of cohesive energy density model, pronounced "sked," like
sled). This separation is given by

62 2)2 + la (3)

Where a characterizes the compound's acidity and ( char-
acterizes the basicity. The acidity and basicity can be mea-
sured spectroscopically by probing how strongly compounds
interact with a standard reference base and a standard acid.
Sample values of 8', a, and ( are given in Table 1. The a and
contributionss distinctly characterize the hydrogen bonding
contributions. Counting them separately means that 8' is
smaller than 8, making the estimated y's closer to 1 when
k 2=0. Acidity and basicity characterize favorable interactions
when acids and bases combine (k12<0) and unfavorable inter-
actions when acids and bases cannot combine (k12>0). This
is the essential effect of hydrogen bonding. Note that water
stands out in Table 1 as a compound with remarkably high
energy density, both in terms of 8' and in terms of a(3. The
water molecule is very small, essentially the size of a single
oxygen atom, but it has a large dipole moment (reflected in
8') and strong hydrogen bonding (reflected in a(3).
We now return to the quantity k12. If k12 =0, then , > 1,
always, but there are situations when m, < 1. When mixing
acids like HC1 with water, for example, the compounds "like
each other" so much that you need to be careful. A more mod-
erate example is given by mixing acetone with chloroform, in
which case mixture boiling experiments show that , < 1. The
proton of the chloroform is made mildly acidic by the elec-
tronegative chlorine atoms pulling on its electrons. The high
density of electrons on the carbonyl oxygen of acetone makes
it mildly basic. Organic chemistry courses should reinforce
these concepts of electron distributions. These considerations
are represented by the guideline that
k1(2 --2 )(32 1)/(4.;) (4)


Vol. 44, No. 1, Winter 2010












In terms of 6',

discomfort of a

component in

solution is

not caused by

dislike for

the other

components, but

by a strong

preference for its

own company.

You may have

heard that an

extroverted

engineer is one

who looks at your

feet when he is

talking to you.

Any introverted

engineers in the

room should

relate to this

perspective on

the definition of

discomfort.


For the chloroform+acetone example, this formula gives
k12 =(5.8-0)(0.12-11.14)/(4*19.64*18.88)= -0.035 (5)

Note how the order of subtraction results in a negative value for k12 when one of the components is
acidic and the other is basic. If you switched the subscript assignments, then Aa would be negative
and A(3 would be positive, but k12 would still be negative. This negative value makes the value of ,
smaller, and that is basically what happens when hydrogen bonding is favorable.
Something else happens when one compound forms hydrogen bonds but the other is inert. Taking
iso-octane(1) as representative of gasoline and mixing it with water(2),
k12 =(0-50.13)(0-15.06)/(4*27.94*14.11)= 0.479 (6)

This large positive value will add to the large (Ab')2 such that 1 >>10, indicating the liquid phase
split that we anticipated. We can quantify the phase split by noting that
x~l/ wici\ >100 (7)

Knowing the saturation limit of water contaminants can be useful in environmental applications.
As a final example, note that we recover an ideal solution when both components hydrogen bond,
as in the case of methanol+ethanol.
k12 =(17.43-12.58)(14.49-13.29)/(4*19.25*18.67)=0.003 (8)

In this case, we see that hydrogen bonding itself is not the cause of solution non-ideality. A mismatch
of hydrogen bonding is required to cause non-ideality.
We can summarize our observations about hydrogen bonding as follows.
1) Ignoring hydrogen bonding entirely (i.e., assuming a=fi=Ofor all compounds) would lead to larger
estimates of solution non-ideality in all cases (6' = 6 then).
2) Ignoring hydrogen bonding would also undermine our ability to anticipate favorable interactions
.-1. ,,, , Eq. (4), as in the acetone+chloroform system.
3) Hydrogen bonding solutions can also be ideal solutions if both components have similar acidity and
basicity, as in the methanol+ethanol example.
4) Hydrogen bonding leads to very unfavorable interactions when one component associates strongly
and the other is inert, as in the iso-octane+water example. This is known as the hydrophobic effect.
Applications of these insights abound in chemical engineering. For example, what third compound
could you add to ethanol+water to make the solution more ideal so that pure ethanol could be ob-
tained? The extension of the SSCED model to multicomponent systems is simple, as discussed in
the thermodynamics course. How soluble is vitamin C in the bloodstream relative to its solubility
in body fat? What about vitamin E? Aspirin? Tylenol? You just need to know the activity coeffi-
cients of these compounds in water and n-octanol (a reasonable approximation of body fat). What
solvent should you use to safely remove an undesirable embellishment from a classical painting?
The embellishment probably used a different paint, so you need to find a solvent with , < 1 in the
embellished paint but , > 1 in the classical paint. Quantitative understanding of fields from art res-
toration to zoology to agribusiness would be impossible without unifying concepts like hydrogen
bonding. Students should retain these concepts and reinforce them as they take complementary
courses throughout their curriculum.
The following example and homework assignments illustrate a range of behaviors that can
be explained with the SSCED model. These behaviors include favorable interactions as in
acetone+chloroform as well as unfavorable interactions as in isooctane+water. Illustrating the entire
range of behaviors at the outset is intended to avoid misconceptions such as thinking that the activ-
ity coefficient is always greater than 1. To emphasize the qualitative nature of this model and the
high value of experimental data, the final homework illustrates how the preliminary guideline can
be refined using experimental data.


Chemical Engineering Education










Example 1.
Estimate the K-value for 10mol%chloroform in 90% ac-
etone at 350K and 0.1MPa. You may assume that loglO(Psa"
P)=7(1+co)(1-T/T)/3.
Solution:
The value of k12 = -0.035 is given by Eq. (5). Vc = 119.2/1.48=
80.5 and VA = 58/0.79 =73.4.
The volume fraction is: 0,c = 0.1*80.5/( 0.1*80.5+0.9*73.4)
=0.109
mc = exp{ 80.5*(1-0.109)2*( (19.64-18.88)2
2*0.035*18.88*19.64 )/(8.314*350) } = 0.573.
P't = 5.40*10( 7*1.216*(1-536.4/350)/3 ) = 0.166 MPa.
Kc = c *Pat/P = 0.573*0.166/0.1 = 0.951

Homework 1.
Gasohol is made by distilling a solution known as beer
(~5mol% ethanol). Compute the K-values of ethanol and
water at 5mol% ethanol and 358.5K and compare them to
the K-values at 95mol%ethanol and 350.6K. Assume that P=
0.1MPa. Explain the impact of activity coefficient on your re-
sults. You may assume that logl0(Ps"'/P)=7(1+0)(1-Tc/T)/3.
Solution:
k12 = (12.58-50.13)(13.29-15.06)/(4*18.67*27.94) = 0.032
At 358.5K and 5% ethanol, Pat = 0.1327 and P't = 0.0658
MPa according to the assumed vapor pressure equation. The
water is nearly pure and computation confirms that
w = 1.016.
Details for ethanol:
OE = 0.05*58.5/(0.05*58.5+0.95*18) = 0.146.
E = exp{ 58.5*(1-0.146)2 ( (18.67-27.94)2 +
2*0.032*18.67*27.94 )/(8.314*358.5) } = 5.517.
This gives KE = 7.319 and Kw = 0.669.
At350.6Kand95'. .lalI pao t = 0.0997 and Pat= 0.0491MPa
E
according to the assumed vapor pressure equation. The ethanol
is nearly pure and computation confirms that TE = 1.001.
For water,
1Y = exp{18*( (18.67-27.94)2 + 2*0.032*18.67*27.94 )
/(8.314*350.6) } =2.046
This gives KE = 0.998 and Kw = 1.002.
Overall, the activity coefficient makes distillation easier at
low concentrations of ethanol, but the large activity coeffi-
cient switches to the water at high ethanol concentrations and
makes water slightly more volatile than ethanol. This is why
distillation fails to completely purify this system.
Homework 2.
The American experience with methyl tert-butyl ether
(MTBE) in the '90s approaches qualification as a fiasco.


Rumor has it that congressmen from corn states thought that
a mandate for 10% "oxygenated fuel" would boost demand
for ethanol, but they did not specify ethanol as the oxygen-
ated fuel of choice. Within four years MTBE was the number
one synthetic chemical produced in the world. What nobody
anticipated was how MTBE might affect groundwater. It
imparts a bitter taste and nasty smell even at parts per bil-
lion. Gasoline is stored in underground tanks, and the tanks
leak. Estimate the solubility of MTBE in water at 298K and
compare it to that of iso-octane and benzene.
Solution: For iso-octane, Eq. (6) gives k12 = 0.479.
x,=l/ 1 = 1/exp{ 114/0.70*( (14.11-27.94)2 +
2*0.479* 14.11*27.94)/(8.314*298) } = 5.8E-17
For benzene, k12 = (0.63-50.13)(4 -15.06)/(4*27.94*18.6)
= 0.263
x,=l/exp{ 78/0.87*((18.6-27.94)2 + 2*0.263*18.6*27.94 )
/(8.314*298) } = 2.2E-6
For MTBE, k12 = (0-50.13)(7.4 -15.06)/(4*27.94*15.17)=
0.226
xM =1/exp{88/0.74*( (15.17-27.94)2 + 2*0.226*15.17*27.94)
/(8.314*298) } = 4.0E-8
So the solubility of MTBE is much higher than that of iso-
octane. This estimate would need to be checked with experi-
mental data, but the essential observation is that the basicity
of the ether suggests checking it out. Benzene is interesting
because its estimated solubility is similar to that of MTBE.
Benzene does not taste or smell like MTBE, but it is carci-
nogenic. Nevertheless, nobody seems to be talking about the
solubility of benzene in groundwater... at the moment.
Homework 3.
No theory should be mistaken as a substitute for experimen-
tal data, especially not such a simple theory as the SSCED
model, and especially on a subject as sensitive as groundwater.
Experimental measurements show that the mole fractions of
iso-octane, benzene, and MTBE in water at 25 �C are actu-
ally closer to 4E-7, 5E-4, and 1E-2, respectively. Use these
measurements to refine your estimates of k12 and predict the
solubility of benzene in water at 38 �C. [Hint: typing Eq. (1)
into a spreadsheet would make it easy to try various values
of k2.]
Solution: Iterating on k12 results in the following table.

Compound k, (predicted) k, (fit)
Iso-octane 0.479 0.04
Benzene 0.263 0.12
MTBE 0.226 -0.08
At 38 �C, x =l/exp{ 78/0.87*( (18.6-27.94)2 +
2*0.12*18.6*27.94)/(8.314*311) } = 0.00066. These results
illustrate limitations in the SSCED model for predicting water
solubility, especially for hydrocarbons. The SSCED model


Vol. 44, No. 1, Winter 2010










predicts a liquid phase split and small solubility, but cannot estimate aqueous solubility precisely. It turns out that understanding
the thermodynamics of water is very challenging, even for theories that are much more sophisticated than the SSCED model.
In fact, the MOSCED model itself applies an empirical value of 36 (compared to 18) for the molar volume of water at 25 C.[11
Empirical fitting in the SSCED model is constrained to adjusting k12, however. The prediction at 38 �C suggests that the solu-
bility of benzene increases by 70% when the temperature approaches body temperature. What does this suggest as a possible
next step in worrying about benzene solubility? (Hint: reread the first sentence of this problem statement.)

RATIONALE
The SSCED model provides a simplified and generalized introduction to the MOSCED model, but the MOSCED model is
designed for other purposes. Specifically, it is designed for infinite dilution activity coefficients instead of being a solution model
at all concentrations. The MOSCED model is given by,


2 21 )2
n V2 RT 2 2 X1) q1q2( - T1 (a2 -1) )
ln2 RT p1


aa ln [ 1
Vl V


Where X is the dispersion factor, T is the polarity factor, q is a factor ranging from 0.9 to 1. aa, p , and , are adjustable pa-
rameters characterizing solvent properties. At infinite dilution, they are specific values, but they must depend on composition
to change from one solvent to the next. That composition dependence is not addressed by the MOSCED model, but it poses no
problem for experts in thermodynamics. Parameters of an activity model like UNIQUAC could be determined from the infinite
dilution activity coefficients and activity coefficients at all compositions computed from UNIQUAC. But activity models like
UNIQUAC tend to be covered after models like MOSCED. The MOSCED model is based primarily on van der Waals mixing.
The terms involving (AX)2 and (V /V ) comprise the Scatchard-Hildebrand and Hory-Huggins contributions derived from the
van der Waals equation when constant packing fraction is assumed.I31 The other contributions are based on phenomenological
arguments. The UNIQUAC model is based on the concept of local compositions. Developing the nuances of fitting parameters
to one activity model then interpolating the free energy based on an entirely different activity model could overwhelm the at-
tention span of sophomores as well as freshmen. Another alternative would be to articulate composition dependencies for all
the parameters. This would detract from a simple explanation. On the other hand, eliminating the polarity factor (i.e., p = co),


Figure 1. Comparison of SSCED model to the conventional Scatchard-Hildebrand model. (a) methanol+benzene at
O.lOlMPa, data ref.8 SSCED: kl2=0.123; ScHil: k12=0. (b) Ethanol+water at 0.lOIMPa, data ref.9; SSCED: kl2=0.032;
ScHil: k12=0.


Chemical Engineering Education


Figure la
355
I - SSCED
350 - - - ScHil
x Expt
345 x

340 i- y

g335 -

330

325 -

320

315
0 0.2 0.4 0.6 0.8 1
xl-yl


Figure lb
380

360

340 -, * -

320



- 280 -'

260 -
240
240 ' - SSCED 4
' x Expt .
220 - - ScHil

200
0 0.2 0.4 0.6 0.8 1
xl-yl










setting aa = 0, and � = 2 for all compositions makes the model
much simpler and more broadly applicable while retaining
the separation that enables consideration of hydrogen bond-
ing influences.
The SSCED model was derived from the MOSCED model
by minimizing deviations in a somewhat crude manner. The
factor of 2 multiplying ap in Eq. (1) was determined by
minimizing the differences between the physical contribu-
tions of the two models. Analyzing the physical contributions
showed that 82 Z (X2 + P2/2) for non-associating compounds.
Note that 8 = 8' in the absence of association. For associating
compounds, Eq. (1) was rewritten as

62 () mp 10)

where m was an adjustable parameter. Minimizing the objec-
tive function X{(8')2 - (X2 + T2/2)}2 for 30 compounds in the
database of Lazzaroni, et al., 11 where 8' was computed from
Eq. (10), yielded an optimal value of m 2.2. The factor of 4
in the denominator of Eq. (4) was determined by minimizing
deviations in vapor-liquid equilibrium (VLE) data for the 10
binary systems of the first five compounds in Table 1, and
three VLE systems involving water with special emphasis
on ethanol+water. These particular mixtures were chosen to
illustrate the range of possibilities from strong solvation to
hydrophobicity. The values of a and P in Table 1 were taken
directly from the compilation of Lazzaroni, et al. Coinciden-
tally, this manner of separation retains the consistent interpre-
tation of cohesive energy density as a primary consideration
(e.g., w' > B'> 8 ', where W means water, B means benzene,
and 0 means iso-octane). This consistency is not immediately
apparent in MOSCED's X parameters.
Other simple alternatives include the Hansen solubility pa-
rameters[4, 5] and the original Scatchard-Hildebrand model.[5, 6]
The problem with the Scatchard-Hildebrand model is that it
overestimates the non-ideality of the solution. For example,
matching the experimental data for methanol+benzene re-
quires a negative value of k12= -0.035 when the Scatchard-
Hildebrand model is applied. Students then conclude that
methanol and benzene must "like" each other, because that
is what k12< 0 should mean. But in this case, the negative
k12 is cancelling the overestimation of the non-ideality from
the unseparatedd" cohesive energy density. Figure la com-
pares the Scatchard-Hildebrand model (with k12 = 0) to the
SSCED model (k12 = 0.123). The positive value k12 = 0.123
from Eq. (4) conveys that the primary role of benzene is to
disrupt methanol's hydrogen-bonding network, which is the
correct interpretation. A similar problem occurs with water
and nearly any other compound. Even ethanol+water is
predicted to be immiscible with the Scatchard-Hildebrand
model, as illustrated in Figure lb, and a large negative value
of k12 would be required to obtain reasonable agreement with
experiment. This kind of "two steps forward and one step


backward" makes the presentation unnecessarily confusing.
The Hansen solubility parameters, on the other hand, have
the advantage of being simpler than SSCED in some sense,
because there is a single hydrogen bonding parameter instead
of two. Nevertheless, Hansen's method cannot account for
activity coefficients less than one. This undermines the scope
of conceptual reasoning that should form the long-term
basis for students' thermodynamic insight. In deference to
Hansen's method, however, the separate contributions to the
solubility parameter are constrained to sum to the original
Scatchard-Hildebrand value. This adaptation from Hansen's
method is helpful in clarifying that SSCED provides separa-
tion, but no elimination.
An advantage of the MOSCED model is its direct ac-
counting of the specific molecular interactions involved in
hydrogen bonding. This accounting is based on spectroscopic
measurements that are independent of the desired activity
coefficients.'71 Kamlet-Taft parameters are dimensionless
measures of acidity and basicity, but the MOSCED model
recasts their values to provide dimensional consistency with
solubility parameters. This may open the door to more creative
ways for students to mesh analytical techniques with engi-
neering applications, as in catalysis for example. Interactions
of zeolite acid sites with molecular base sites may seem less
mysterious when the existence of molecular base sites has
been acknowledged at the outset. This improved chemical
insight can be pervasive throughout the curriculum.
Eq. (1) was deliberately expressed in terms of k12 instead
of simply substituting Eq. (4) directly, as in MOSCED. Note
that Eq. (4) is described as a "guideline." This means that it
is a starting point, but it leaves open the possibility of refining
the value as described in the third homework problem. The
value of experimental data is hinted at in the presentation,
and several assignments at the sophomore level lead students
through the process of finding relevant data and inferring re-
fined values of k12. A model like UNIFAC can predict activity
coefficients but sheds little light on the underlying chemical
interactions that lead to the behavior. Furthermore, the UNI-
FAC model makes it difficult to refine predictions in light of
experimental data for specific systems of interest.
Another alternative would be Wertheim's theory.[10l Wert-
heim's theory forms the basis of hydrogen-bonding equations
of state like the SAFT,11" PCSAFT,[121 and ESD113' models.
It is also based on rigorous statistical mechanics instead of
phenomenological arguments. Therefore a simplified integra-
tion of Wertheim's theory with Scatchard-Hildebrand theory
would have an advantage as a natural segue to the more so-
phisticated theories. In fact, this was the preferred alternative
initially. Wertheim's theory and its implementations have been
the driving forces in molecular thermodynamics for the past
15 years. These developments form the basis for appreciating
the qualitative behavior evident in the SSCED model. For


Vol. 44, No. 1, Winter 2010










example, the ESD model of cyclohexane+methanol shows
that the model indicates more stable solutions when hydrogen
bonding is recognized explicitly, consistent with experiment.J141
Unfortunately, a direct implementation of Wertheim's theory
would excessively complicate the model. Some indirect com-
promise may be feasible long-term, but the current form of the
SSCED model is satisfactory for present purposes.
Finally, it may be possible to relate the dispersion, acidity,
and basicity parameters to ab initio characterizations, as in
the COSMO-RS model.J151 This would reinforce the value of
quantum mechanical computations throughout the curriculum,
but freshmen (and sophomores) would be unlikely to appreci-
ate this level of sophistication. It would make more sense to
recast the ab initio results as reinforcing the SSCED concepts
after the fact, in the junior or senior year.
There is one substantial disadvantage of the SSCED model
that motivates the coverage of more advanced models like
UNIQUAC and SAFT. Since it has only one binary interac-
tion parameter, the magnitude of deviations from non-ideal-
ity can be adjusted, but not the skewness. The skewness of
the Gibbs energy is controlled by the volume ratio in the
SSCED model. This limitation, however, pertains to quan-
titative modeling, not to the conceptual and educational
device intended here.

ASSESSMENT
Assessments of student learning have not been performed
for the freshmen yet, but a very similar presentation pertains
to the sophomores and this has been assessed in class through
the ConcepTest methodology and through traditional exami-
nation questions.
ConcepTests pose simple questions to the class and allow
them to post their answers anonymously for quick compila-
tion. 21 Electronic devices typically facilitate this approach,
but it can be conducted with colored flash cards. In the strict-
est sense, ConcepTests should focus entirely on conceptual
questions, but a small adaptation permits engagement in
active learning for computational exercises as well. I refer
to these as "CompuTests." A few examples are given below.
The (%) quantities refer to the percentage of students who
answered correctly.
ConcepTest 1.
Referring to Figure 2, cases A and B correspond to charac-
terizations of the attractive energy between two molecules as
described by the square well potential. This attractive energy is
given by 12= (F1*8F2)2 (1- k12) where e12 gives the depth of the
square well. Note that the bottom of the well is -e 12 Provide
the response (A or B) corresponding to each situation.
a. Which (A or B) corresponds to k > 0? (72%)
b. Which corresponds to components ' i. " each other?
(100%)


c. Which corresponds to a higher "escaping tendency" for
component 1? (94%)
d. Which will give the !,.1. 1i bubble point pressure?
(50%)
ConcepTest 2. (94%)
Which of the images in Figure 3 properly depicts a maxi-
mum boiling azeotrope?
"CompuTest" 1. (43%)
Arrange the following mixtures from most compatible to
least compatible according the SSCED solubility parameter
criterion (k 12=0). 1) Pentane+hexane, 2) decane+decalin,
3) 1-hexene+ dodecanol, 4) pyridine+methanol, 5) diethyl
ether+n-heptane
A. 12345 B. 12534 C. 54123 D. 21543
"CompuTest" 2. (76%)
An azeotrope exists for n-butane(1)+ethyleneOxide(2) at
1.013 bars at -6.5 C and 78wt% butane. Estimate the activ-
ity coefficient of EtO ( 22) at the azeotropic composition and
temperature from the Scatchard-Hildebrand model assuming
k = 0. (See Table 2)
(a) 0.04 (b) 1.06 (c) 1.98 (d) 2.89
Examination Question 1. (89%)
Based on the Scatchard-Hildebrand solubility parameters
(k12 = 0), arrange the following mixtures from most ideal
to most non-ideal: a) 2-pentanone+l-pentene, b) 2-penta-
none+ naphthalene, c) ethanol +naphthalene, d) n-hexane+
ethanol.


20 ,,
- - - u22
10 - ------u12
--ull
0 ,I

3-10 -

,-20

-30 A
:,I ..............4r^
-40 - 4---- k 0=
40 - ........... k=0

-50 ---

-60
0.2 0.3 0.4 0.5 0.6 0.7
r (nm)
Figure 2.


Chemical Engineering Education











ILLUSTRATION FOR CONCEPTEST2:

355
V
350 (A)

I 345
E
E 340
335 3 LL
3 3 0 O...
0 0.2 0.4 0.6 0.8 1
x1-yl


370
365 L
360
5) 355
= 350
E 345
rL 340-
335 V
330-
0 0.2 0.4 0.6 0.8 1
x1-yl


350 (B)
I 345
E
E 340

335 - V

330 . . . .
0 0.2 0.4 0.6 0.8 1
x1-yl


370
365 V
360-
355
= 350 (D)
E 345
E 340
" 335 L
330 -
0 0.2 0.4 0.6 0.8 1
x1-yl


Figure 3.

TABLE 2
Compound Tc(K) Pc(MPa) w CpIg/R MW b(cal/cc)12 p298
n-BUTANE 425.2 3.80 0.193 11.89 58 13.50 0.60
EtOxide 469.0 7.10 0.200 5.80 44 21.72 0.89


Examination Question 2. (61%)
A common problem with recycling polyester is the impu-
rities from bottle caps and labels. The bottle caps typically
weigh 0.05 g and the bottles are 2g. The caps are polypropyl-
ene (PP) with molecular weight of 60,000 g/mol. The bottles
are polyethyleneterephthalate (PET), with molecular weight
of 10,000g/mol. The solubility parameter and density of PP
can be estimated as (bp= 14.11; 9=0.70), similar to those of
isooctane. The solubility parameter and density of PET can
be approximated as ( pp=17.90; 9=0.86), similar to those of
p-xylene. Estimate the infinite dilution activity coefficient for
PP in PET at 100 C assuming k =0.

CONCLUSIONS
Two issues pervade teaching in the chemical engineering
curriculum: time constraints and knowledge retention. If
you teach too much in too little time, little is retained. If you


teach too little, students cannot "connect the dots" from one
isolated fact to another. One approach is to articulate broadly
applicable concepts, like the SSCED model presented here.
This mindset leverages familiar chemical concepts like acidity
and basicity while dovetailing with the physical interactions
of the van der Waals model covered in physics coursework.
Leveraging the concepts presented in other coursework has
the two-fold advantage of saving time and rewarding students
for retaining what they learn from course to course.
The assessments show that this perspective is accessible
to sophomores at least. Assessments were not conducted for
freshmen at this early stage of adapting the presentation for
them, but we expect similar results to those for sophomores.
Students are able to quickly recognize solution non-ideality
and the impact this may have on solubility and volatility.
Students whose careers take them away from process design
may not remember how to compute an activity coefficient


Vol. 44, No. 1, Winter 2010












five years after graduating, but they can remember that acids
and bases interact strongly and that organic chemicals as well
as inorganic chemicals should be formulated to account for
those interactions.

NOMENCLATURE

Latin
K - vapor-liquid partition coefficient
k12 is the binary interaction parameter
MW - molecular weight (g/mol)
Pa - vapor pressure (MPa)
P - pressure (MPa)
R = 8.314 J/mole-K, the gas constant
T - absolute temperature (K)
Uvap internal energy of vaporization at 298K (J/mol)
V = MW/ pL is the liquid molar volume at 298K (cm'/mol)
x - mole fraction of ith component in the liquid phase.
y - mole fraction of ith component in the vapor phase.

Greek
a - characterizes the compound's hydrogen bonding acidity
(Table 1, MPa1')
- characterizes the hydrogen bonding basicity (Table 1,
MPa')
6 = (Uv /V)1 - total solubility parameter (MPa')
6' = (62 - 2ap)1 is the dispersion contribution to the solubil-
ity parameter. (MPa1')
4 = xV /Ex V is the volume fraction
, - activity coefficient

REFERENCES
1. Lazzaroni, M.J., D. Bush, C.A. Eckert, T.C. Frank, S. Gupta, and J.D.
Olson, "Revision of MOSCED Parameters and Extension to Solid
Solubility Calculations.]," Ind. Eng. Chem. Res., 44, 4075 (2005)


2. Falconer, J.L., "Use of ConcepTests and Instant Feedback in Thermo-
dynamics," Chem. Eng. Ed., 38, 64 (2004)
3. Elliott, J.R., "Fluid Structure for Sophomores," Chem. Eng. Ed., 27,
44(1993)
4. Hansen, C.M., Hansen Solubility Parameters: A User's Handbook,
CRC Press, Boca Raton, Fla. (2000)
5. Scatchard, G., "Equilibria in Non-Electrolyte Solutions in Relation
to the Vapor Pressure and Densities of the Components," Chem. Rev.,
321-333 (1931)
6. Hildebrand, J.H., J.M. Prausnitz, and R.L. Scott, Regular and Related
Solutions, Van Nostrand-Reinhold, New York (1970)
7. Kamlet, M.J., J.M. Abboud, M.H. Abraham, and R.W Taft, "Linear
Solvation Energy Relationships. 23. A Comprehensive Collection of the
Solvatochromic Parameters, pi*, alpha, and beta, and Some Methods
for Simplifying the Generalized Solvatochromic Equation," J. Org.
Chem., 48, 2877 (1983)
8. Kurihara, K., H. Hori, and K. Kojima, "Vapor-liquid equilibrium data
for acetone + methanol+ benzene, chloroform + methanol + benzene,
and constituent binary systems at 101.3 kPa.," J. Chem. Eng. Data,
43, 264(1998)
9. Bloom, C.H., C.W. Clump, and A.H. Koeckert, "Simultaneous mea-
surement of vapor liquid equilibria and latent heat of vaporization,"
Ind. Eng. Chem., 53, 829 (1961)
10. Wertheim, M.S., i li i.... 11. li .i I/ Directional Attractive Forces. I.
Statistical Thermodynamics," J. Stat. Phys., 35, 19 (1984)
11. Chapman, W.G., K.E. Gubbins, G. Jackson, and M. Radosz, "New
Reference Equation of State for Associating Liquids,"Ind. ( ..... i ,
Res., 29, 1709 (1990)
12. Gross, J., and G. Sadowski, "Perturbed-Chain SAFT: An Equation of
State Based on a Perturbation Theory for Chain Molecules," Ind. Eng.
Chem. Res., 40, 1244 (2001)
13. Elliott, J.R., S.J. Suresh, and M.D. Donohue, "A Simple Equation of
State for Nonspherical and Associating Molecules," Ind. Eng. Chem.
Res., 29, 1476 (1990)
14. Suresh, S.J., and J.R. Elliott Jr., "Multiphase Equilibrium Analysis
via a Generalized Equation of State," Ind. Eng. Chem. Res., 31, 2783
(1992)
15. Klamt, A., "Conductor-Like Screening Model For Real Solvents - A
New Approach To The Quantitative Calculation Of Solvation Phenom-
ena," J. Phys. Chem., 99, 2224 (1995) 1


Chemical Engineering Education











r.I.1 AIChE special section


A BLENDED APPROACH

TO PROBLEM-BASED LEARNING

In the Freshman Year







DIANE ROSSITER, ROBERT PETRULIS, AND CATHERINE A. BIGGS


The University of /.jj. p ./ * .i/..jji.. i./ S1 3JD, UK
Chemical Process Principles I (CPE 1002) is a core and
compulsory course taken by approximately 70 stu-
dents in the freshman year of a chemical engineering
undergraduate degree program at the University of Sheffield,
Sheffield, United Kingdom. The primary learning objective
is to develop the students' knowledge and understanding of
material balances, where the core learning content looks at
material balances for single unit operations, such as mixers,
splitters, separators, and reactors, and then various combina-
tions of these single units in series. This allows for purges
and recycles to also be considered.
This paper provides details of the chronological develop-
ments of the course, the issues emerging from student feed-
back, the actions taken based on this feedback, and the lessons
learned by staff. Finally, some conclusions are provided.

COURSE DEVELOPMENTS
Historically, CPE 1002 had a relatively high failure rate
(25% in 2003/04). Student feedback indicated that many of
the students were finding that learning the content was chal-
lenging, and they were struggling to make the connection
between the mathematical manipulations required and what
a practicing chemical engineer does. Further, the number of
students taking the course increased, from 28 in 2003/04, to
55 the following year, to about 70 in the past two years (see
Table 1, next page). To address the problems of unacceptably
high failure rates and increasing numbers of students, we
decided to adopt a different style of course delivery to try to
engage the students more effectively with the learning content
and its application.
Vol. 44, No. 1, Winter 2010


Despite the increased numbers of students in 2004/05, the
failure rate improved to 16% with the introduction of small-
group tutorials. This failure rate was still considered too high
for such core content, however, so we decided to implement
new pedagogical strategies that we hoped would improve
student learning even with increasing student numbers.
The new mode of course delivery was a type of Problem-
Based Learning (PBL). Our primary reason for this choice
was the evidence available in the literature on the effective-

Diane Rossiter is currently a senior univer-
sity teacher in the Department of Chemical
and Process Engineering and the assistant
director of Learning and Teaching for the
Faculty of Engineering at the University
of Sheffield.
Robert Petrulis
is now the exec-
utive director of
the Office of Pro-
gram Evaluation
at the University
2 of South Carolina. During the project, Bob
was a researcher in the Centre for Inquiry-
Based Learn-
ing in Arts and
Social Science
at the University
of Sheffield.
Catherine Biggs is currently a reader in
the Department of Chemical and Process
Engineering at the University of Sheffield.
She is currently an EPSRC Advanced Re-
search Fellow in the area of fundamental
bio-chemical engineering science.


� Copyright ChE Division of ASEE 2010


































ness of PBL in motivating students to learn.J11 We also had
good departmental links with the Department of Chemical
Engineering at the University of Queensland, Australia-and
in particular with Queensland's Professor Paul Lant. Lant's
department has had national success with their students adopt-
ing a type of PBL, referred to as project-centered learning.[2]
As part of his sabbatical at Sheffield, Professor Lant was
willing to assist with its introduction within our course. At
Queensland, this approach has been implemented program-
wide, whereas we were looking initially to introduce it for only
one course. Professor Lant visited Sheffield in the academic
year 2005/06 and brought all his learning resources relating to
his syllabus for Material and Energy Balances. This was the
start of a major transformation for our course, and a catalyst
for change within our department.
The initial transition was from traditional didactic delivery in
the academic year 2003/04 (see Table 1), when students num-
bers were relatively low, to a two-hour weekly problem-based
learning workshop with a one-hour-per-week supporting lecture
from 2005/06 onwards. The idea was to shift the emphasis from
the lecturer being the "sage on the stage" to "the guide on the
side,"[31 with the students engaging in problems that required
them to seek out information from the resources and supports
provided, and construct their own knowledge. We anticipated
that this would both improve content learning and, at the same
time, help students gain key intellectual skills needed for con-
tinued success in their chemical engineering course.
This curriculum design aligns with the definition of PBL
as "a conception of learning as an integrated process of
cognitive, metacognitive, and personal development" as
provided in Newman's review.[41 PBL has been introduced
successfully into many professional fields of study including
chemical engineering-see Woods,[11 who has carried out
many detailed studies.


"Traditional" PBL Approach (2005-2007)
Firstly, the CPE1002 course content was extended to cover
key personal and transferable skills such as group work, com-
munication skills, independent and self-directed learning,
and peer assessment, without losing any of the key technical
material. This was done by introducing weekly group work,
one industrial visit, and two major group assignments incor-
porating industrially relevant processes over the 12-week
teaching term. The group assignments were carried out over
a number of weeks.
The key to a successful PBL approach is the use of authentic
problems1'1 to engage the students while developing their core
technical skills. Professor Lant provided several case studies
relating to industrial processes, and the students worked in
small groups of four or five to carry out the weekly assigned
formative tasks and assignments. The industrial visit to a pulp
and paper mill1 was provided to enhance the students' under-
standing for one of the major assignments. Students were also
encouraged to do weekly homework problems to support their
learning and develop their problem-solving skills.
When the changes were first introduced in 2005/06, feed-
back from the students via the end-of-course questionnaire]51
was extremely positive, e.g.: "The tutorials were excellent,
really hard work, but incredibly useful in cementing the
course ideas." (student 1); and "The assignment group work
was a new challenge compared to the usual academic work.
This made the work more exciting and rewarding as well as
reinforce the key concepts and knowledge." (student 2). As
shown in Table 1, however, 23% of the students failed the
summative assessment compared with 16% the year before!
1 Unfortunately, the plant closed in January 2008 so for the academic
year 2008/09 the students have had a "virtual" site visit using video
resources from the Internet and a PowerPointpresentation provided
by the company.


Chemical Engineering Education


TABLE 1
Chronological Course-Delivery Developments
Academic year 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09
Student numbers 28 55 66 67 70 69
No. of staff 2 2 3 4 4 4
Lectures 2 hours 2 hours 1 hour 1 hour 1 hour 1 hour
(per week)
Tutorials 2 per term 3 per term weekly weekly weekly weekly
3 hours each 3 hours each 2 hours each 2 hours each 2 hours each 2 hours each
Large group Small group PBL-style PBL-style PBL-style PBL-style
Summative 80% exam 75% exam 50% exam 50% exam 50% exam 50% exam
assessment 20% course- 25% course- 50% assign- 50% assign- 40% assign- 40% assign-
work work ments and test ments and test ments ments
10% online test 10% online test
Failure rate* 7 (25%) 9 (16%) 15 (23%) 1 (2%) 4 (5.7%) 5 (7.2%)�
Notes for Table 1: * The student must obtain an overall weighted average mark of 40% or greater from the combined summative
assessment components. � Three of the five students who failed the summative assessment did not attend for the examination










A review carried out by Biggs[6] showed that those students
who did not participate in PBL group work had not gained
many of the associated summative assessment marks, and had
relied solely on achieving very high marks on the exam paper.
Since the exam paper was only worth 50% of the total course
marks, however, and at least 40% was required for a pass, this
was a high-risk strategy that usually failed. It is worth noting,
however, that most of the students who did not participate
in PBL group work also were not engaging with the rest of
their degree program. Thus, the problem wasn't necessarily
the PBL style of delivery, but these students' general lack of
motivation and engagement.
In the following year (2006/07), two additional changes
were made: 1) increased emphasis throughout the course by
academic staff on the need for students to engage with the
group PBL tasks and assignments to ensure success in the
course, given the 50:50 split between exam and coursework;
and 2) the weekly homework was marked so that the students
got regular, timely, formative feedback. This had the desired
effect of improving the overall summative assessment results:
Only 2% failed the course that year (see Table 1).
The introduction of weekly homework marking and the
size of the class, however, meant two members of academic
staff and two teaching assistants were now involved with the
course compared with the usual one member of academic staff
on other traditionally delivered courses within our depart-
ment. This had a significant impact on the running cost of the
course, and as such it would not be practical for all courses to
be delivered in this way without a major restructuring in the
departmental teaching allocation. This was unlikely to occur
in the short term. Hence, a different approach to providing
effective and regular feedback to the students was needed to
accommodate increasing class sizes and simultaneously to
support the weaker students.
Another significant challenge was the need to find a teaching
space to accommodate up to 70 students for aPBL-style tutorial.
(Due to timetabling constraints it was not possible to split the
cohort into, say, two smaller groups.) Recently, the University of
Sheffield has had a major investment in Inquiry Based Learning
(IBL), flexible learning, and teaching spaces (see CILASS'1 Web
site). The University's largest collaborator accommodated only
48 students, however. Hence, an alternative venue was sought
that had a flat floor with moveable tables. Such venues proved
to be very limited in number on our campus, although the con-
struction of new teaching spaces for larger student numbers is
currently in progress. As part of the PBL activities, it was also
essential that the students could see the projector screen from
anywhere in the space. This was solved by ensuring that the
rectangular tables were positioned so that all four students in a
group could sit with two facing the other two and also turn side-
ways to face forward toward the projector screen. These issues
may seem trivial, but for PBL-style tutorials to be successful it
is very important that the students are able to "huddle" in their


Vol. 44, No. 1, Winter 2010


groups and exchange ideas and information as well as come to-
gether as a whole class to discuss ideas and gain feedback. This
is the essence of the communication and collaboration needed
for PBL to take place.
By the end of 2006/07, the new course-delivery style of
PBL was deemed a success, due to the improvement in the
failure rate to only 2% and the continued positive feedback
from students. Initially, it had been envisioned that further
developments would involve rolling out the same format
across other courses. The combined impact of the running
costs and infrastructure limitations meant that this type of
widespread rollout was not yet an option, however. But les-
sons learned from this course on how to deliver a PBL-style
tutorial have been transferred to other small-group teaching in
the department, such as modules in our process design strand
throughout Years 1 to 3.

Blended Learning PBL Approach (2007-onwards)
Despite generally positive student feedback for the PBL style
of delivery, there was evidence from the peer evaluation data
that some students were being "carried" by the members of their
groups. Since the summative assessment was 50% group assign-
ments and 50% individual examination there was the danger
that students could succeed in the course but not have the core
technical skills required for courses later in their undergradu-
ate program. Constraints on staff time and increasing student
numbers (see Table 1) meant that providing additional remedial
group tutorials and/or one-to-one support for developing the
students' problem-solving skills were not practical options to
support weaker students. It was necessary to find an alternative
approach that would "blend" with the PBL approach.
It is noted by Woods11l that students need to be skilled in
problem solving before embarking on PBL. So, it was im-
portant to provide an effective mechanism for supporting the
weaker students that was not staff intensive. The development
of the computer-aided learning resources, particularly a set of
online formative quizzes, seemed an ideal strategy to meet this
need. The University's Centre for Inquiry-based Learning in
the Arts and Social Sciences (CILASS) was approached for
funding for development and evaluation of the online forma-
tive quizzes, and a grant was awarded in 2007.
The online resources were developed in time for use during
the 2007/08 academic year, and were provided viaWebCT Vista
(managed learning environment).81 The aim of introducing
online self-assessment resources (or quizzes) alongside PBL
was to enable the students to self-assess their weaknesses and
strengths in the core chemical engineering principles and to
practice their problem-solving skills. We expected that students
who used these resources would come to PBL classes more
prepared and better able to contribute to the group work.
The online formative quizzes allow students to get instant
feedback on whether their answers are correct. If they select
the wrong answer then additional feedback is provided directing

25


















Teach, ., 06 6. . .. 3,i,.. ,. ..,� - , .," " -
Ae.yK Akglcli I "t (kgol) X


S 0lr.nuIlun I,,-l. -

SAssignmlntls,
3I Calendar

i3 Learning Modules
fi Profile
SSeaorch
, Web Links
15 WIho Orlrini-
[H) = Hidden

SMNanage Courq w
..J AsspsqmPnr Manag]pi
.;, Assignment Dropbox
14 6rarlp AmkY
R Gradin, Forms
4 Group Manager
C Trakrira l
0 Notes
c SelectiAe Release


A(kg/h) P(kgfh)
Mixer


B(kg/h)


Student Value Correct
Response Answer
A + a - P .. F

sr. + 0-

Score: 0%


" Question Status
O Unanswered
rD Answer not saved

/ Answered
1i 2 3 4 5

6 7 S 9 10

11


Feedback



ir,.. i.n .,.| p n. .nrl ., JT ur '_riC, n,


i~.-r -,ntal F-..- .r ly -r r -. .-.,rn-llr,..r.- ,r..j ,r1. ...(r .-.r,, rr.- .- . rrill n.fir-,iil r la,-,.-
i- ir_.r ,. '.?.- rr., ., r ,: I., r e I ,.t ] n'r ,.lJl
14'\I Ou llOar


Finish :


I Help


Figure 1. Online formative quiz-multiple-choice question


Figure 1. Online formative quiz-multiple-choice question.


them to the relevant part of the lecture notes and/or textbook
by Felder and Rousseau[9] (see Figures 1 and 2). The computer
aided assessment tool makes this possible without requiring
large amounts of staff time. Although the quizzes were deliv-
ered within WebCT Vista,�'1 some of the development work was
carried out within Respondus,[10� a third-party tool for creating
online assessments. (For detailed discussion of the development
of the online quizzes see Rossiter and Biggs.J111) These online
quizzes cover five core technical skills:
1) Unit conversions using the unity brackets approach
2) Mass to mole conversions
3) Calculations and definitions ; i,,,,, to material balances
4) Material balance calculations without reactions (e.g. Fig. 1)
5) Material balances with reactions (e.g., Fig. 2).

Figure 1 shows an example of a typical multiple-choice
question for a mixer within the question databank of the online
quizzes. The student here has selected the wrong answer so is
given specific feedback on why their chosen answer is wrong,
as well as being told the correct answer and being given more
general feedback on what was required. Within the question
databank, as well as multiple-choice questions, there are many

26


calculation-type questions where a range can be specified for the
variables so that a different set of values is presented each time
the question is encountered. Thus, the students can repeatedly
carry out the same calculation but with different number sets
until they have mastered the problem (see Figure 2).
In the academic year 2007-08, for the first time, the ex-
amination for the course was held at the end of Semester 22
whereas the blended PBL approach was used in Semester 1.
This meant that there was a gap of several months between
2 The examination of the course content changed to being at the end of
the year because a detailed program review by the Departmental Cur-
riculum Committee concluded that the students were being over-assessed
and developing a surface learning approach to their studies as they had
10 separate 1.5 hour examinations during the year. So to promote deep
learning of the content the number of examinations was reduced to six
separate 3-hour examinations. With the CPE1002 Chemical Process
Principles-material balances being assessed at the end of Semester 2 as
part of a 3-hour exam including the Semester 2 course content on energy
balances. The timing of this exam meant the students had also engaged
in a week-long PBL-style design study relating to mass and energy bal-
ances, thus reinforcing the course content delivered in Semester 1. Given
only a slight increase in the failure rate for the summative assessment of
CPE1002 (see Table 1), the change of timing of the examination does
not appear to have had a detrimental effect on the students'outcome.
Chemical Engineering Education


W406 a














IB H Z U |x x� T | P 1n < Powe& |


Preview
SBefore pubishng your Respondus file to the server, i is
vPreview ieW ecommnended that you "preview" the fie, in the preview made,
9gt rhe FIlq the Re pondus ile remns n your had drive and an Internet
I - b Iferlie,o is u=toeIt
If you want to make changes o the Respondus fie, click the
"modiy itenn" btton from within the preview task You wi be
returned to the preview task once the change has been made.


Appeaance in In(ernet Explorer (other bowsers may vary slightly] od fy]
Changes made in Edit will not appear until you Save the document !...... | 41.i

The molar flowrate of air into a furnace for the complete combuson reaction of propane (C3Hg) is 1727 kmol/h. If the air is fed
at 55% excess, what is the theoretical amount of oxygen needed for the reaction. Assume the composition of air is 21mol%
oxygen (O0) and 79mol% nitrogen (N2). Give you answer to 1 decimal place
Answer: I

General Feedback
See F & R page 145 for calculations of excess air.
The chemical reaction is C3H8 + 502 gives CO2 + H20
If the amount of excess air is [y], then the amount of excess 0, is 0.21*[y].
The theoretical amount of 0, is less than the excess 02
If the air i 50% in excess, then this is 1.5 times thetheoretical amount of 02
So moles 02 ed = .5*moles of theoretical 02
Correct Ans2wer















Figure 2. Online formative quiz-calculation question.


the material being delivered and the examination. So an added
benefit of the online quizzes was that they were available
throughout the year to assist students when they studied for
the examination and carried out a detailed PBL-style design
study in Semester 2 relating to mass and energy balances.

EVALUATION RESULTS OF BLENDED
LEARNING APPROACH
As part of our CILASS-funded learning and teaching project
during 2007-08, a detailed evaluation study was carried out
to assess the impact of introducing the online quizzes within
the PBL framework, on the students, the academic staff, their
department, the university, and the wider community. This
evaluation study provided a rich source of data. The data was
collected via classroom observation by Petrulis, a focus group
with students, interview with academic project staff, and a
questionnaire to all students involved with the course. Some
of this data is discussed in the following sections.

What Did the Students Say?
The students were asked in the questionnaire and the focus
group about all aspects of the course including the PBL-style

Vol. 44, No. 1, Winter 2010


tutorials and their use of the online quizzes. The comments in
Table 2 (next page) suggest that by working on the PBL group
tasks and assignments the students were seeing the connection
between chemical engineering practice and what they were
learning in the course. This was reinforced by the question-
naire data3 where:
* 93% of respondents found the PBL activities enjoyable
and ,,.,, ii.,,, Student quotes: "Working as a group
was fun and I worked extra hard because I did not want
to let my group down," "Made me feel as if I was a 'real'
chemical engineer and made me feel mature," "It was
hard work but was enjoyable."
* 98% of respondents found their experience of PBL
helped them at least to some extent to develop confidence
and skills in working collaboratively. Team working is an
essential skillfor engineers.
* 96% of respondents found their experience of PBL
helped them at least to some extent to develop confidence
and skills in problem solving. Problem solving is also an
essential skill for engineers.

3 Questionnaire, End of Semester 1 Feedback by Year 1 students,
Feb. 15, 2008, 54 responses out of 69.











TABLE 2
Extract From Student Feedback at Focus Groupt on Their Learning (2007/08)
Interviewer: Do you think the module (CPE1002) changed the way you approach learning or problem-solving activities?
Student: "In the first semester, we were learning things and I could see why we were learning them because I could see how to apply them. This
semester, we're learning a lot, but I'm not always sure why. I wish we had more practical group assignments throughout the course."
Interviewer: If you had your choice between the group format and what you're getting now (predominantly lectures), which would you prefer?
Student: "You actually felt like an engineer when you were doing the group assignments. Now I just feel like a student, learning a lot of things."


TABLE 3
Extract From Student Feedback at Focus Groupt on Feedback in Quizzes (2007/08)
Student: "They (Quizzes) were beneficial because of the immediate feedback. With the homework, it took a week or so. The online quizzes
also referred you to the book and page for more information. If you just get a grade, that's meaningless."
Student: "The online self-test quizzes gave INSTANT feedback. So if you didn't get the question right, I understood why and did not do the
same mistake again. The quizzes were unlimited and this helped me practice."


TABLE 4
Extract From Student Feedback at Focus Groupt on Use of Quizzes (2007/08)
Interviewer: A couple of you said you didn't use the quizzes at all. Could you say why?
Student 1: "I tried the quizzes a couple times, and they weren't very hard. The homework were more challenging, and you could talk to the
teachers about them. That felt so much more helpful."
Student 2: "The examples in the quizzes weren't as difficult as the homework."
f Eight students were present at the focus group. - Interviewer R. Petrulis (CILASS) on March 3, 2008


In relation to the online quizzes, 81.5% of students who
responded said the quizzes had helped them to some extent
in developing their core technical skills for the problem-based
learning activities. (The rest either didn't use the quizzes,
didn't know, or didn't respond.) The comments provided in
Table 3 show that the immediate feedback on the quiz answers
was also found to be beneficial.
The online quizzes were provided to help the less-able
students to develop their problem-solving skills and this
seems to have been achieved judging by the quote from this
student: "The homework made us work hard, but the quizzes
really helped us learn how to do the homework." From the
quotes in Table 4, however, students 1 and 2 don't seem to
need this type of support. Biggs stated when initcI i" .d4
"The point of the quizzes was to help those who needed the
basics; not to challenge those who needed to be challenged,
because the homework (and) assignments were there to do
that. I think this shows that we were right to set this up in
the first place." Rossiter commented at the same interview:
"It reinforces that you could describe something in three dif-
ferent ways and it would have meaning to different people.
It's not that the quizzes, assignments, and homework cover
different things-they don't. They make it accessible to dif-
ferent types of learners."
What Did the Staff Say?
Overall, blending online formative quizzes with off line
PBL-style tutorials has proved successful for this course.
This has helped to provide a mechanism with instant feed-
28


back for the less-able students to get help in developing their
problem-solving skills in preparation for the PBL group tasks
and assignments. It has also resulted in helping all students
to be more actively engaged in the group work.
Providing these online quizzes did involve significant de-
velopment time; however, this was offset by a reduction in
students' requests for remedial one-to-one support. Some time
was also gained through some of the coursework (an online
test worth 10%) being automatically marked by WebCT. Also,
the homework sheets were modified since some of the ques-
tions formed the basis of the online quizzes. Hence, this led
to some reduction in the weekly homework-marking load.

CONCLUSIONS
There have been several major challenges to address relat-
ing to this first-year chemical engineering course, such as
increasing numbers of students, a widening range of student
abilities and learning needs (including, in some cases, lack
of well-developed problem-solving skills), constraints on
provision of additional academic staff, and need for adequate
learning spaces for PBL-style tutorials. Creative solutions had
to be found to deal with these challenges, and this paper has

4 CILASS Project leaders' interview of Catherine Biggs and Diane Ros-
siter by Robert Petrulis (Evaluator) carried out 20 March 2008. The
interview was carried out as part of the CiLASS project triangulated
evaluation study and its purpose was to have the academic project
team commenton the collated studentfeedbackfrom the questionnaire
and focus group. Hence "close the loop" on the student feedback.


Chemical Engineering Education











outlined the blended-learning approach adopted for course
delivery and support. In 2005, it was envisioned that the
PBL approach in CPE 1002 would lead to major change in
favor of PBL for the entire program. This has not occurred
to the extent originally hoped. The process-design strand of
the undergraduate programs has been influenced by the PBL
approach, however, since we (Rossiter and Biggs) have both
been involved in design-project supervision.
We originally set out to address the issue of student failure
in this course, knowing that the high failure rate could be
exacerbated by steadily increasing numbers of students. The
innovations described in this paper-particularly the imple-
mentation of online quizzes-have indeed improved student
learning and success. They have had ancillary benefits as well,
including promoting student self-motivation and engagement,
improving problem-solving skills among weaker students, and
helping students develop transferable skills in such areas as
teamwork and communication. We might also mention that
the experience of teaching the course, despite the increas-
ing class size, has become progressively more rewarding as
student results have improved.

ACKNOWLEDGMENTS
The authors are grateful to: Professor Paul Lant of Chemical
Engineering at the University of Queensland for his inspira-
tion, initial provision of the PBL content, and training that was
invaluable for starting a major process of change within our
department; and the Leverhulme Trust for funding his sabbatical
to Sheffield. The authors also wish to acknowledge the Centre
for Inquiry-based Learning in the Arts and Social Sciences
(CILASS) at the University of Sheffield for project funding and
support to carry out some of this educational development work.
Author Biggs would like to acknowledge the Engineering and
Physical Sciences Research Council (EPSRC) for the provision
of an Advanced Research Fellowship (EP/E053556/01).


REFERENCES

1. Woods, D.R., "Problem-Based Learning, Especially in the Context of
Large Classes," available from htm> accessed July 6, 2009.
2. Crosthwaite, C., I. Cameron, P. Lant, and J. Litster, "Balancing Cur-
riculum Processes and Content in a Project-Centered Curriculum: in
Pursuit of Graduate Attributes," Education for Chemical Engineers
Trans. IChemE, Part D, March, pp. 1-10 (2006)
3. King, A., "From the Sage on the Stage to the Guide on the Side,"
abstract of journal article in College Teaching, 41, accessed Sept.
15 2008. Available from qst?docld=94305197> (1993)
4. Newman, M., i .. .-l..i . .lIi I ....... I "in Higher EducationAcad-
emy Imaginative Curriculum Guide, accessed July 6, 2009. Available
from tive_Curriculum_Guide_Problem_Based_Learning> (2008)
5. Biggs, C., CPE112 Studentfeedback2005/06, The University of Shef-
field, Department of Chemical and Process Engineering, Sheffield S1
3JD (2006)
6. Biggs, C., CPE112 Chemical Process Principles I - A Review, The
University of Sheffield, Department of Chemical and Process Engi-
neering, Sheffield S1 3JD, March, Internal Report (2006)
7. CILASS, CILASS spaces for learning and teaching, The University
of Sheffield, Centre for Inquiry-based Learning in the Arts and Social
Sciences. Accessed Sept. 15 2008. Available from ac.uk/cilass/learning-spaces> (2008)
8. WebCT Vista, Blackboard Learning System - Vista Enterprise License,
E-learning platform, accessed Sept. 14, 2008. Available from www.blackboard.com/products/Academic_Suite/Learning_System/
vista.htm> (2008)
9. Felder, R.M., and R.W Rousseau, Elementary Principles of Chemical
Processes, 3rd Ed., John Wiley and Sons, Inc., USA (2000)
10. Respondus, Respondus version 3.5 - Assessment Tool for Learning
Systems, Respondus, Inc. spondus.shtml> Accessed Sept. 15, 2008 (2008)
11. Rossiter, D., and C.A. Biggs, "Development of Online Quizzes to
Support Problem-Based Learning in Chemical Engineering," In
Proceedings Third International Blended Learning Conference 2008,
University of Hertfordshire, UK, 18-19 June, pp. 113-123, University
of Hertfordshire Press, ISBN 978-1-905313-58-7 (2008) 1


Vol. 44, No. 1, Winter 2010









r.19.1 AIChE special section


ASYNCHRONOUS

DISTANCE-EDUCATION COURSE

FOR NONSCIENTISTS
Coordinated Among Three Universities








TAMARA FLOYD SMITH,' DAVID BAAH,1 JAMES BRADLEY,2 MICHELLE SIDLER,2 ROSINE HALL,3
TERRELL DAUGHTREY,2 AND CHRISTINE CURTIS2'4
Tuskegee University * Tuskegee, AL
2 Auburn University * Auburn, AL
3 Auburn University at Montgomery * Montgomery, AL
4 University of South Carolina * Columbia, SC
ith the goal of exposing non-science-and-engineer- Tamara Floyd Smith is an associate professor of chemical engineering
ing (NSE) students to the principles and ethical and a 3M scholar at Tuskegee University in Tuskegee, AL. She is also
issues of nanotechnology, the course "Concepts of affiliated with the Tuskegee University Center for Advanced Materials.
Nanoscience" began as a proposal-"Ethics of the Nanoscale" David Baah is a Ph.D. student in the Materials Science and Engineering
-to the National Science Foundation. The proposal included Program at Tuskegee University.
several educational components including, but not limited to: James Bradley is the W. Kelly Mosley Professor of Science and Hu-
manities in the Department of Biological Sciences at Auburn University
1) exposing freshman non-science majors to nanotechnology, in Auburn, AL.
an emerging technological field; 2) incorporating ethics into Michelle Sidleris an associate professor of English at Auburn University
science courses; 3) intra- and inter-university team teach- in Auburn, AL.
ing; as well as 4) exploring the benefits and challenges of Rosine Hall is a professor of biology at Auburn University at Montgomery
multi-university asynchronous and synchronous distance in Montgomery, AL.
education (SDE) formats. This discussion is limited to the Terrell Daughtrey is a media support technologist at Auburn University
details of offering the course in SDE format jointly among in Auburn, AL.
Auburn University (AU), Tuskegee University (TU), and Christine Curtis was formerly a professor of chemical engineering and
associate provost at Auburn University in Auburn, AL. She is currently
Auburn University at Montgomery (AUM). Details related to vice provost for Faculty Development at the University of South Carolina
course content and other aspects of the program are discussed in Columbia, SC.
elsewhere.1m
� Copyright ChE Division of ASEE 2010
30 Chemical Engineering Education










The advent of the Internet and ubiquitous high-speed data
transmission have made SDE an attractive educational for-
mat. The SDE format is one in which data are transmitted to
students in real time as opposed to an asynchronous format,
which typically involves recordings. Advantages of SDE when
compared to traditional "brick and mortar" classrooms are the
obvious time and energy savings associated with individuals
not being required to gather in one location. Studies suggest
that students taught the same course in traditional and SDE
formats perform similarly.[2] Thus, choosing an SDE format
is a neutral choice with respect to student outcomes. Two
disadvantages of any distance education format, however, are
limited direct contact with the instructor and the potential for
technical complications, both difficult to overcome.
An SDE format may use video, audio, graphics, and combi-
nations of the three.[3] Standard videoconferencing equipment
or Internet-based software can facilitate two-way communica-
tions for SDE. Multi-point (three or more transmitting sites)
efforts are more complex, however, and may require a hub
or bridge. Another feature of SDE is that students may be
gathered in two or more classrooms, sitting alone at remote
computers, or combinations of the two.
Various disciplines have investigated the SDE format.[4, 5]
This discussion, however, is limited to science and engineer-
ing courses and programs. One prominent example of an
SDE effort is the Singapore-MIT Alliance for Engineering
Education, which focuses on professional master's programs
and also Ph.D. educational programs.[6] The alliance began
in 1998 and has expanded to include a more research-centric
phase. The alliance includes three institutions: the Massachu-
setts Institute of Technology (MIT), the National University
of Singapore, and Nanyang Technical University. Typically,
students are gathered in classrooms at the three institutions
where video, audio, and graphics data are transmitted. In
addition to typical coordination and technical difficulties in-
herent in this type of effort, the alliance faces the exceptional
challenge of a 12-hour time difference. Despite challenges,
the alliance has been very effective and emerged as a leader
in international distance education.]71
The Electrical and Computer Engineering Department
at the University of West Florida in Pensacola offers SDE
courses to the Fort Walton Beach Campus.[2] The courses are
two-way transmissions between a classroom on each campus.
The distance-education effort, which began in Fall 2002,
involves the simultaneous transmission of video, audio, and
graphics data using Polycom videoconferencing systems and
an interactive pen display and multimedia lectern manufac-
tured by SMART Technologies, Inc. An assessment of the
SDE program indicated that students at the main campus and
off-site campus passed at similar rates of 67.9% and 66.7%,
respectively. Additionally, students at the off-site campus
were administered a survey to gain feedback on their experi-
ence in the SDE course. The survey indicated that 1) students


preferred synchronous distance education to asynchronous
distance education, 2) one drawback of SDE was lack of
direct interaction with the instructor, and 3) students valued
the availability of SDE.
The School of Information Technology and Engineering
at George Mason University has offered SDE since 1994.[8, 9]
The number of SDE courses has grown from one course in
2000/2001 to 24 courses in 2003/2004.[9] Moreover, George
Mason's experience has provided the following observations
related to SDEO8':
1) Most students would prefer a traditional course format
but, for those who chose SDE, the disadvantages of the
SDE format do not outweigh disadvantages of traveling
to a traditional classroom.
2) In the absence of the inconvenience of travel, some stu-
dents still prefer SDE because of their learning styles.
3) Consistent with other groups,f2l studies that compare
SDE to traditional classrooms suggest no significant
difference with respect to student outcomes.

Lastly, the Georgia Institute of Technology School of Engi-
neering has been involved in SDE since as early as 1991 with
the offering of an online master's program in electrical and
computer engineering from both the main campus and a satel-
lite campus in Metz, France. 10] Georgia Tech also participates
in an academic collaboration with Georgia Southern Univer-
sity, Armstrong Atlantic State University, and Savannah State
University to offer students at those campuses engineering
degrees using several educational modes including SDE.
Clearly, the SDE format is not unique within science and
engineering disciplines, but the course that the authors de-
scribe is unique because it targets freshman-level, non-sci-
ence-and-engineering majors, whereas most efforts emanating
from science and engineering departments target science and
engineering majors. The motivation for the SDE course format
for this course was fourfold:
1) Real-time interaction of instructors and students on
three different campuses
2) The efficient use of resources on the three campuses
associated with combining three classrooms into one
classroom
3) The optimal use of instructor expertise from the three
campuses - the most qualified instructor from among
the three universities was chosen to lecture on a given
topic
4) The SDE format is on par with traditional styles with
respect to student outcomes

RESOURCE REQUIREMENTS

SDE efforts can be resource-intensive during the initial
roll-out phase. For example, the purchase of a single video-
conferencing unit can represent a significant capital invest-


Vol. 44, No. 1, Winter 2010










ment of approximately $10K. Also, network staff resources
are critical to address transmission issues related to firewall
settings. Finally, there is a significant time investment by
instructors to modify lecture content so that it is suitable for
the SDE format.
Equipment
Distance education equipment was purchased (as necessary)
and configured for all three universities. AU used a Tandberg
director system which consisted of a 3000i Smart Board rear
projector 67" display touch screen, a Tandberg 6000 Codex,
audio ceiling microphones with electronic sound cancella-
tion (eliminates microphones picking up the sound from the
far end of the classroom and returning it as an echo), two
wide-angle wave cameras, and a 12" Centronic touchscreen
control monitor. TU used a Tandberg 770 MPX Portable unit
that included one wide-angle wave camera, a 32" monitor,
and a roll cart. AUM used a Vitel Video Conference System
that included two 32" monitors, two wide-angle cameras, and
12 table microphones.
All three institutions had access to views of the other
two institutions during lectures but, typically, the lecturing
institution was viewed unless another institution was asking
a question. Because the course was viewed in real time, it
could be and was very interactive. This opportunity for an
improved extended-classroom dynamic couldn't be realized
for a distance education course that is asynchronous.
Facilities
Figure 1 shows the configuration for the SDE transmissions.
The Intercampus Interactive Telecommunication System
Office at the University of Alabama at Birmingham (UAB)
facilitated the three-way interaction of the participating insti-
tutions and provided streaming archiving for asynchronous
lecture viewing. Special classrooms were not required, but
access is critical. Most universities schedule classrooms to be
occupied most of the day. Consequently, if transmission issues
need to be resolved, limited access to the exact Internet port
that is used can cause unnecessary course delays.
Staffing
Auburn University, the lead institution for the course,
provided a media support instructional technologist who
attended all lectures and was the technical coordinator and
contact person for technical issues from all three campuses.
Also, initially, network staff from all three campuses were
integral to the course to address firewall issues and other
technical issues that arise during transmission. The Singa-
pore-MIT Alliance found that the best practice is to move
the course transmission outside of the firewall.[61 If network
administrators are not comfortable with operating outside of
the firewall, however, satisfactory transmission can still be
achieved. After the initial resource-intensive phase, network
staff should still be available for emergencies to prevent inter-
ruption in course instruction.
32


MULTI-UNIVERSITY INSTRUCTION:
STRUCTURE AND EXECUTION
Several logistical issues needed to be addressed related to
multi-university SDE instruction. First, each university is on a
different class schedule. Graduate student schedules are typi-
cally very flexible and permit deviations from standard class
start times (e.g., on the hour) and course blocks (50 minutes,
80 minutes, etc.), but undergraduate schedules are much more
constrained. As a result, course scheduling was a significant
challenge. BothAU and TU offer Monday/Wednesday/Friday
(MWF) and Tuesday/Thursday (TTh) courses, but AU starts
on the hour and ends at 10 minutes until the hour, whereas TU
starts at 10 minutes after the hour and ends on the hour. AUM
does not have class on Friday. The compromise was that the
course would be offered MWF with 40 minutes of core con-
tent. AU handled issues like homework and announcements
for 10 minutes before class, and TU handled those issues for
10 minutes after class. All sessions were recorded, and AUM
viewed the Friday lecture off-line.
Another issue was the scheduling of institutional breaks.
Each institution had different spring breaks, semester start/end
dates, holidays, etc. Long breaks such as spring break were co-
ordinated by viewing recorded lectures during those periods.
The semester start/end dates in some cases were close enough
for all three institutions to coordinate and in other cases were
handled by temporary asynchronous viewing.
Course Offerings and Enrollment
The course was offered during the Spring 2007 and Fall
2007 semesters. Course enrollment data are provided in


Figure 1. Multi-University Synchronous Distance
Education transmission configuration.
Chemical Engineering Education


TU Tandberg
Video


AU Tandberg Video
Conferencing


AUM Vtel
Video










Table 1. Enrollment (pre-test participation) was significantly
higher at Auburn University because the course was one
section of an established course. At Tuskegee University,
the course was acceptable for "science elective" credit but,
despite heavy advertising, students and advisors were accus-
tomed to more traditional courses and chose those. Enrollment
at AUM was affected by the lack of a laboratory offering,
since all majors must have two laboratory science classes
to meet basic curriculum requirements. At AU, the lower
division "Concepts of Science" course, which is targeted at
non-science majors, has a recitation hour instead. The cur-
riculum committee at AUM would not allow a recitation to
be substituted for a laboratory.

Student Outcomes
Student learning for the purpose of assigning a grade was
assessed using four in-class exams and a comprehensive final
exam. The impact of the course, however, was assessed by
administering pre-course/post-course tests to the students.
The results of the pre-test and post test are outlined in Table
1. The pre-test was administered to establish the baseline for
student knowledge of the subject matter. Typically, the post
test was administered after the final lecture but prior to the
final exam. The pre/post test consisted of 32 questions (24
True/False type and eight short-answer). Table 1 shows the
number of students participating from AU, TU, and AUM
and their corresponding pre/post test average scores. For AU,
all students who completed the pre-test did not complete the
post test, and the pre/post assessments were not matched in
the end because of Institutional Review Board (IRB) restric-
tions. Consequently, it was possible that the students who
scored the lowest on the pre-test did not take the post test
and thus inflated the score difference. To remove this error,
the pre-test results reflect both the average of all the students
tested and the average of the students scoring highest on
the pre-test corresponding to the same number of students
who took the post test at AU. The second number reported
in the score difference column gives the most conservative
estimate of student learning because it is calculated from the
arbitrarily higher pre-test scores. Another issue is that 24/32
questions were True/False type, implying a baseline of zero


knowledge at a score of 12/32 or 37.5% for random guessing.
Despite the aforementioned challenges with the assessment
exercise, it is clear that the students' knowledge of the subject
matter improved significantly, ranging from 7.8 to 29.2%. In
addition to increased knowledge of nanoscience, students
were also able to benefit from the expertise of faculty from
multiple campuses and gained insight into the culture of
other campuses.
Clearly, the assessment data revealed that the students'
knowledge of the concepts of nanoscience improved. The
overall course drop rate, however, was 41% for the first se-
mester and 33% for the second semester. In addition, overall
enrollment dropped by 43% from the first to the second
semester. Based on anecdotal evidence, a number of factors
including but not limited to course difficulty, unbalanced
course content, and technical difficulties contributed to the
decrease in enrollment. Because multiple factors influenced
course enrollment, it is difficult to isolate the contribution of
the SDE format in the absence of survey data.
For the SDE course described, students gathered in one
location at their respective campuses where traditional classes
were also offered. Thus, the common SDE benefit of saving
the time, energy, and inconvenience of traveling to a distant
location was not realizable, and the primary benefit to students
was the optimization of faculty expertise from three campuses.
It is the opinion of several faculty, however, that the benefit
of optimized faculty expertise may not outweigh the chal-
lenges of the SDE format for freshman non-science majors
because the students are not advanced enough to appreciate
the optimized expertise.

CONCLUSIONS
A synchronous distance education course joint among Au-
burn University, Tuskegee University, and Auburn University
at Montgomery was successfully offered for two semesters to
introduce non-science majors to the concepts of nanoscience.
The majority of the lectures were conducted in real time so
that students from all three campuses could interact with the
various lecturers and students at other campuses. Although
several logistical and technical issues were encountered, the


TABLE 1
Concepts of Nanoscience Enrollment and Assessment Data
Pre-Test Post Test
School Term % Diff
# Students Avg. score # Students Avg. score
_______________ (%) (%)________
AU Spr 07 31 68.1/75.7 16 91.2 23.1/15.5
AU Fall 07 18 67.8/72.8 11 89.1 21.3/16.3
TU Spr 07 4 62.5 4 70.3 7.8
TU Fall 07 2 60.9 2 90.1 29.2
AUM Spr 07 2 72.4 2 92.6 20.2
AUM Fall 07 1 1

Vol. 44, No. 1, Winter 2010











course ran satisfactorily for two semesters with the support
of networking staff and limited asynchronous viewing of
recorded lectures. Analyzing the results of assessment tests
given to students revealed that their knowledge of the con-
cepts of nanoscience improved by 7.8% to 29.2% as a result
of completing the course.

ACKNOWLEDGMENTS

The authors would like to gratefully acknowledge the
National Science Foundation (SES-0532340) for funding.
The following faculty members are acknowledged for their
participation in the course: Robert Ashurst, Guy Beckwith,
Virginia Davis, Roderick Long, and Christopher Roberts (Au-
burn University); Leonard Ortmann (Tuskegee University).

REFERENCES
1. Floyd-Smith, T., D. Baah, J. Bradley, M. Sidler, R. Hall, and C. Curtis,
"Concepts of Nanoscience for Non-Scientists: A Distance Education
Course Coordinated Among Three Universities," AIChE National
Meeting Conference Proceedings, Nov. 16-21 (2008)


2. Matthews, C., "Sychronous Distance Delivery of an Electrical and
Computer Engineering Program,"ASEE/IEEE Frontiers in Education
(2005)
3. Pullen, J., ,..lih Il..I1. of Internet Video in Distance Education For
Engineering," Proceedings of IEEE/ASEE Frontiers in Education
(2001)
4. Podgor, E., "Teaching a Live Synchronous Distance Learning Course:
A Student Focused Approach," University of Illinois Journal of Law,
Technology and Policy, 2006(2), 263-272 (2006)
5. Webster, J., and P Hackley, "Teaching Effectiveness in Technology-
Mediated Distance Learning," The Academy of Management Journal,
40, 1282-1309 (1997)
6. MIT-Singapore Alliance: Retrieved
7/28/2009.
7.
8. Pullen, J., "Synchronous Internet Distance Education: Wave of the
Future or Wishful Thinking?," Proceedings of the 2002 eTEE Confer-
ence, 174-179 (2002)
9. Pullen, J., and P McAndrews, "Low-Cost Internet Sychronous Distance
Education Using Open-Source Software," Proceedings of the 2004
ASEE Annual Conference (2004)
10. Jackson, J., M.HayesIII,A. Saad, andT. Bamwell, "Frameworkfor Coopera-
tive Synchronous and Asynchronous Distributed Engineering Education,"
Proceedings of the 2002 ASEE Annual Conference (2002) 1


Chemical Engineering Education












[ey1 %AIChE special section


A Survey of the Role of

Thermodynamics and Transport Properties


IN ChE UNIVERSITY EDUCATION

in Europe and the USA


PETER AHLSTROM, KAREL AIM, RALF DOHRN,
J.RICHARD ELLIOTT, GEORGE JACKSON,
JEAN-NOEL JAUBERT, EUGENIA A. MACEDO,
JUHA-PEKKA POKKI, KATI RECZEY,
ALEXEY VICTOROV, LJUDMILA FELE ZILNIK,
AND IOANNIS G. ECONOMOU
University of Bords * SE-50190 Bords, Sweden,
and as listed in the biographical information box.
Thermodynamics and Transport Properties (TTP) is a
central subject in the majority of chemical engineering
university curricula worldwide, and it is thus of interest
to examine how it is taught today in various countries. The


Peter Ahlstrim is an associate professor for chemical physics at the Univer-
sity of Boras, Sweden. He got his M.Sc. in chemical engineering in 1983 after
studies at Lund University and ETH Zrrich and his Ph.D. at Lund University.
in 1988. His research interests are molecular simulations of liquids, including
developing methods for the treatment of polarisable molecules. Recently his
focus has been on macromolecules and phase equilibria.
Karel Aim graduated with honors in technical, analytical, and physical
chemistry from the Institute of Chemical Technology in Prague in 1971 and
received his Ph.D. in physical chemistry from the Institute of Chemical Process
Fundamentals (ICPF) of the Academy of Sciences (ASCR) in 1977. Since
1971 he has worked at ICPF and spent longer periods with the Technical
University of Denmark in Lyngby and the University of Trieste. His main re-
search interests include experimental and applied statistical thermodynamics
of fluid systems.
Ralf Dohrn studied industrial engineering and received his Ph.D. in chemical
engineering from the Technical University Hamburg, in Harburg, Germany.
Since 1998 he has been the head of the Thermophysical Property Group in
the Process Development Department of Bayer Technology Services GmbH.
He teaches thermodynamics at TU Hamburg Harburg, where he holds an
honorary professorship, and at Institute Francais du Petrole.
loannis G. Economou is the research director of the Molecular Thermodynam-
ics and Modeling of Materials Laboratory at the National Center for Scientific
Research "Demokritos" in Aghia Paraskevi, Greece. He holds a Diploma of
Chemical Engineering from the National Technical University of Athens (1987)
and a Ph.D. in chemical engineering from Johns Hopkins University (1992).
Richard Elliott is a professor of chemical and biomolecular engineering at
the University of Akron, where he has taught since 1986. He is a coauthor of
the text Introductory Chemical Engineering Thermodynamics with Carl Lira
of Michigan State University, published by Prentice Hall.
George Jackson is a professor of chemical physics in the Molecular
Systems Engineering Group at the Chemical Engineering Department of
Imperial College London. He earned his D. Phil. (PhD) in physical chemistry


content and the organization of the courses implicitly reflect an
unexpressed "thermodynamics philosophy." The discussion
of different learning styles"' and their implication on teaching
methods has also spurred us to investigate which methods are
used for TTP teaching, especially since it is often regarded as
a "difficult subject." Our ultimate aim is to improve chemical
engineering education for the benefit of the graduates and the
industries that will hire them.
A survey on graduate thermodynamics education exclu-
sively in the United States was performed a few years ago
by Dube and Visco.[21 As far as we know, no systematic study
of the undergraduate TTP education has been undertaken, at
least in recent years.


at Oxford University.
Jean-Noel Jaubert is a professor of chemical engineering thermodynamics in
the Laboratory of Thermodynamics for Multiphase Processes (LTMP) at INPL,
France. He is a graduate of the Ecole Superieure de Chimie Marseille (France)
and has a Ph.D. from Universite Paul Cezanne. His teaching and research
interests are thermodynamics, calculation of phase equilibrium, prediction of
physicochemical properties, and improvement of thermodynamic cycles.
Eugenia A. Macedo is an associate professor of chemical engineering at the
University of Porto. She graduated in chemical engineering from the University
of Porto (Portugal) in 1978, and received her Ph.D. from the same University
in 1984. Her research interests are in chemical thermodynamics and separa-
tion processes.
Juha-Pekka Pokki is a lecturer and a researcher at Helsinki University of
Technology where he received M.Sc. (Tech.) in 1995 and D.Sc. (Tech.) in
2004 in chemical engineering. His research area is modeling, computation,
and measurement of phase equilibrium.
Kati Reczey is an associate professor at Budapest University of Technol-
ogy and Economics, BME. She graduated as chemical engineer in 1972 at
Technical University of Budapest, Hungary. She got a dr. techn. degree in
1980 and a Ph.D. in 1991. Her field of research is biotechnological utilization
of lignocellulosics.
Alexey I. Victorov is a professor of physical chemistry at the Department of
Chemistry, St.Petersburg State University. He holds a Ph.D. (1987) and Dr.Sci.
(1997) from the same university. His studies in the area ofmolecularthermody-
namics are focused on phase equilibria modeling, macroscopic behavior, and
structure of soft matter, and self-assembly on nanoscale (branching micelles,
asphaltenes, ion-exchange membranes, block copolymer gels, etc.).
Ljudmila Fele Zlnik received her Ph.D. in chemical sciences from the Uni-
versity of Ljubljana, Slovenia, where she is an assistant professor. She is a
research fellow at the National Institute of Chemistry, Ljubljana, Slovenia. Her
research areas are separation processes, process design and optimization,
and thermophysical properties.


� Copyright ChE Division ofASEE 2010
Vol. 44, No. 1, Winter 2010 3.











In the present study, a survey about TTP education in both
undergraduate and post-graduate programs in Europe and
the United States is presented. Responses received from 136
universities from 20 different European countries and the
United States were thoroughly analyzed and the major find-
ings are presented here.
The study differs from the earlier one of Dube and Visco
in that:
i. Both Europe and the United States were included in the
study and a comparison is performed between Europe
and the United States regarding certain educational
aspects.

ii. Both undergraduate and graduate education are examined.

iii. The teaching methods were .,i. ,,1,i.i

A survey regarding education in EU (European Union)
countries is especially timely in light of the Bologna process.
The Bologna process seeks to establish standards of compari-
son for curricula that have developed independently in many
countries over many years. The unification envisioned by the

TABLE 1
Number of universities / colleges per country that
were contacted and that responded to the survey*
Country number of answers
(number of inquiries sent)
Austria 1 (2)
Belgium 3 (12)
Bosnia and Herzegovina 1 (1)
Croatia 2 (3)
Denmark 2 (4)
Estonia 1(1)
Finland 4(4)
France 6 (24)
Germany 28 (36)
Greece 3 (3)
Hungary 2 (3)
Italy 5 (14)
Netherlands 1(13)
Norway 1(6)
Portugal 5 (7)
Russia 2 (14)
Serbia 1 (1)
Slovenia 2 (2)
Sweden 4(7)
United Kingdom 7 (18)
USA 55 (150)
TOTAL 136 (325)
* Countries from which no response was received include (the number
of institutions contacted is shown in parenthesis): Iceland (1), Ireland
(3), FYR Macedonia (1), Moldova (1), Switzerland (1), Poland (3).


EU means that career mobility and training must be taken into
consideration. Performing this survey at this time provides
a snapshot of the thermodynamics curriculum that can serve
to advance the Bologna process while simultaneously docu-
menting the status of chemical engineering education in both
Europe and the United States.

SURVEY METHODOLOGY
The survey was conducted by an international team of
chemical engineering professionals from academia and
industry using a Web-based surveying system.I31 Invitations
were sent out by e-mail to universities and colleges offering
an accredited chemical engineering program. The e-mail was
normally sent out by one of the co-authors of this paper, in
most cases from the same country as the contacted university
or from a neighboring country. The corresponding addresses
were collected with personal knowledge or based on informa-
tion from the Web pages of the institutions. In each case, the
invitation was sent either to a teacher responsible for TTP
teaching or to the head of the chemical engineering program,
department, or school. In a few cases, no such information
could be found and the invitation was sent to the general e-
mail address of the institution. Several reminders were sent
out to increase the final response rate; nevertheless signifi-
cant variation in the frequency of responses per country was
observed. A summary of the institutions contacted and the
responses received per country is shown in Table 1. Overall,
the response rate in Europe was 46% whereas in the USA
a lower rate of 36% was recorded. About one-third of the
European responses were from Germany whereas from most
other countries the response rate was much lower.

RESULTS AND DISCUSSION
TTP Teaching With Other Disciplines
More than 70% of the universities that responded offer a
B.Sc. in chemical engineering, 65% offer an M.Sc., and 55%
offer a Ph.D. Most universities offer at least two courses of
TTP in the chemical engineering curricula.
About half of the courses are taught to chemical engineers
exclusively whereas the rest are taught together with other
disciplines of engineering, mainly mechanical and/or process
engineering. The first course is often studied together with
other disciplines of engineering, especially in Europe: In 39%
of the cases (10% in the United States), this second discipline
is mechanical engineering, in 29% of the cases (2% in the
United States) it is process engineering, and in 19% of the
cases (0% in the United States) is energy engineering. Other
programs with joint TTP teaching include chemistry/applied
chemistry (14% in Europe and 4% in the United States), ma-
terials science (11% in Europe and 6% in the United States)
and bio engineering (10% in Europe and 9% in the United
States). In some cases one thermodynamics course is studied
together with several other disciplines.
Chemical Engineering Education









The secondTTP course is studied together with other programs
to a muchlesser extent: In 27% of the cases in Europe (6% in the
United States) it is studied together with mechanical engineering,
in 28% of the cases in Europe (0% in the United States) with
process engineering and in 22% of the cases in Europe (3% in
the United States) with energy engineering. These results indicate

50

40

S30 -


10
& 20




1 2 3 4 5 >5
Number of courses
Figure 1. Number of TTP courses reported by the various
universities. (Black bars: Europe. Gray bars: USA.)


Physical chemistry


Independe


60


III..-
None <3 3 7 8-12 13 18 19 24 >24
Chemical Engineering


II i


Other cou


60


.I lI .
None <3 3-7 8-12 13-18 19 24 >24
Weeks


I'll
" I W
None <3 3-7 8-12
Weeks


that in many instances the first TTP course has to be kept at a
general level (and does not specialize in chemical thermodynam-
ics) to accommodate the various fields of study of the students.
TTP Teaching in Terms of Quantity
The extent of TTP that is taught has been analyzed both with
respect to the number of courses and their size. The number
of courses reported from each university is given in Figure 1
where it can be seen that the majority of European universi-
ties report more than two courses each whereas the majority
of U.S. universities reports at most two courses. Hence, most
of the following discussion is based on the first two courses
reported from each university.
An issue that caused much confusion among the respondents
was the definition of the size of a course since no unambiguous
measure of the length and the workload per course exists. We
have chosen to use the workload measured by the amount of
full-time study weeks per course, i.e., the intention is that if a
course was expected to be studied as the only course during a
given period, the value given should be the number of weeks
that course was expected to fill the student's time. If the student
in a given program was expected to follow more than one course
at the same time, the week should be split between the courses
according to the generated work
nt course load. An example: If two courses
are given in parallel during 10 weeks
and both of them are expected to
generate the same workload, each
of them is regarded to as being of
* five weeks' length. To simplify the
calculation for European universi-
ties, we introduced a transforma-
tion based on the European Credit
I . Transfer System (ECTS) introduced
1318 19 with the Bologna process in Europe:
1.5 ECTS units correspond approxi-
rses mately to one week of work since
one year usually contains about 40
study weeks corresponding to 60
ECTS units. Judging from the reac-
% Eu, tions of the respondents, however,
SUSA such a course-size measure is not
yet familiar in many countries, and
thus some care has to be exercised
- when interpreting the results.
13-18 19 24 24 Both in Europe and the United
States, just over 40% of the courses


Figure 2. The extent of TTP (as full-time study weeks) taught as part of different coursess.
Frequently thermodynamics is taught as a part of many different courses, like physical chem-
istry and applied chemical engineering courses, and the amount of thermodynamics in those
courses is shown here. Sometimes, however, a pure thermodynamics course is given and
those courses are presented as "Independent course" in this paper. (In this context "Chemical
Engineering" does not include physics, physical chemistry, and similar fundamental courses
but only the amount of TTP in the applied chemical engineering courses.)
Vol. 44, No. 1, Winter 2010


spend at most seven weeks on ther-
modynamics. In Europe the courses
are generally less than 19 weeks
whereas in the United States, one-
fifth of the respondents spend more
than one semester on TTP. As seen
from Figure 2, for the chemical en-
37












TABLE 2 Contents of thermodynamics Course 1 (percentage of total number of responses)
Topic Central Treated in some detail Mentioned Not part of course
Europe USA Europe USA Europe USA Europe USA
1st law 90 91 8 7 0 2 2 0
2ndlaw 88 80 10 11 1 2 1 7
Entropy 80 74 14 13 4 6 3 7
Molecular/Statistical 9 9 24 15 35 48 32 28
interpretation of entropy
Free energy and quality of energy 44 43 22 26 22 19 11 13
3rd law and absolute entropy 26 33 21 18 35 35 18 33
Thermodynamic cycles 55 50 28 37 8 7 10 6
Heat expansion of solids and liquids 14 18 34 30 33 35 20 17
Equations of state 45 56 36 32 10 11 9 2
Phase equilibria 39 48 26 15 16 9 19 28
Vapor Liquid Equilibria 30 46 21 18 21 4 28 32
Liquid Liquid Equilibria 15 22 18 19 15 19 52 41
Heat transfer 9 7 19 11 20 39 52 43
Thermochemistry 21 9 16 20 6 30 56 41
Statistical thermodynamics 5 2 4 6 26 30 65 63
Molecular simulation 1 0 1 7 14 15 84 78
Kinetic theory of gases 8 0 15 9 32 22 45 68
Non-equilibrium thermodynamics 3 2 12 2 12 9 78 87
Thermodynamics for biological systems 4 0 3 6 16 37 78 58


TABLE 3 Contents of thermodynamics Course 2 (percentage of responses for course 2)
Topic Central Treated in some detail Mentioned Not part of the course
Europe USA Europe USA Europe USA Europe USA
1st law 33 43 17 17 27 20 23 20
2nd law 36 46 14 20 25 17 25 17
Entropy 28 49 22 23 23 14 27 14
Molecular/Statistical 11 11 16 26 34 34 39 29
interpretation of entropy
Free energy and quality of energy 36 34 19 37 22 14 23 14
3rd law and absolute entropy 19 11 25 23 22 29 34 37
Thermodynamic cycles 34 6 5 17 13 34 48 43
Heat expansion of solids and liquids 11 11 27 17 19 43 44 29
Equations of state 56 51 16 40 8 6 20 3
Phase equilibria 59 78 11 11 13 9 17 3
Vapor Liquid Equilibria 52 78 14 11 8 6 27 6
Liquid Liquid Equilibria 42 54 12 17 8 20 38 9
Heat transfer 22 6 12 17 14 20 52 56
Thermochemistry 36 29 17 31 6 23 41 17
Statistical thermodynamics 8 14 9 14 16 40 67 31
Molecular simulation 3 3 3 9 17 46 77 43
Kinetic theory of gases 8 3 8 14 23 26 61 57
Non-equilibrium thermodynamics 3 0 9 9 11 14 77 77
Thermodynamics for biological systems 3 3 8 17 6 31 83 49


Chemical Engineering Education

















































gineering courses, two sets of course lengths were observed,
corresponding either to at least a full semester of full-time
studies or to less than half a semester. Most students meet
thermodynamics in physical chemistry and chemical engineer-
ing courses. In Europe, but not in the United States, a pure
thermodynamics course is also included in most programs.

Contents of TTP Courses
A list of selected items was made and the respondents were
asked to fill in how central that was in each course, given four
alternatives: "Not part of the course, \ k. m i, n.Ld," "Treated
in some detail," and "Central." The results for courses 1-3
are given in Tables 2-4.
In the first course, the first and second laws of thermody-
namics as well as entropy are central in both regions. It should
be noted that in 7% of the U.S. universities entropy is not
discussed. Normally the statistical interpretation of entropy is
mentioned as well as the third law and absolute entropy, but
not in significant depth. One reason for this can be that in many
cases a TTP course has to cover the educational needs for other
disciplines as well. The second course frequently concentrates
more on phase equilibria. In the main, both of these courses
consist of classical thermodynamics whereas the molecular
interpretation often is touched upon. Statistical thermody-

Vol. 44, No. 1, Winter 2010


namics and molecular simulation as well as thermodynamics
for biological systems are not core topics in any of the two
courses either in United States or in Europe, although they
are more frequently mentioned in the United States. Equally,
non-equilibrium thermodynamics is not part of any of the two
first courses in any of the regions. An interesting detail is that
about half of the two main courses in the United States at least
mention thermodynamics of biological systems.
A third TTP course is taught at about half of the European
universities but only at one-fifth of the universities in the
United States. This course is mainly spent on phase equilibria
(but also on entropy in the United States). Statistical Thermo-
dynamics also forms a part of the majority of the third course
in the United States and in 42% of the courses it is central or
treated in some detail. In Europe, however, there is no mention
of statistical thermodynamics or molecular simulation in most
courses. Further details are provided in Table 4. These results
partly reflect the fact that many of the first thermodynamics
courses are taught to general engineering students whereas
the latter courses normally are taught to chemical engineering
students and hence are more specialized. Another observation
is that atomistic perspectives are encountered earlier in the
United States than in Europe, where classical thermodynamics
normally is the central theme in all the first three courses.


TABLE 4 Contents of thermodynamics Course 3 (percent of responses for course 3)
Topic Central Treated in some detail Mentioned Not part of the course
Europe USA Europe USA Europe USA Europe USA
1st law 33 36 21 43 23 14 23 7
2ndlaw 30 21 14 21 23 21 33 36
Entropy 28 50 28 29 12 14 40 7
Molecular Statistical 5 43 12 21 21 21 63 14
interpretation of entropy
Free energy and quality of energy 26 36 21 29 7 21 47 14
3rd law and absolute entropy 12 21 12 14 23 29 54 36
Thermodynamic cycles 16 14 9 21 21 43 54 21
Heat expansion of solids and liquids 7 7 19 21 14 29 61 43
Equations of state 49 21 19 7 12 57 21 14
Phase equilibria 51 64 14 29 9 0 26 7
Vapor Liquid Equilibria 33 57 7 29 9 0 51 14
Liquid Liquid Equilibra 44 36 9 21 7 29 40 14
Heat transfer 35 7 7 21 12 7 47 64
Thermochemistry 9 14 12 57 26 7 54 21
Statistical thermodynamics 9 21 5 21 9 21 77 36
Molecular simulation 7 7 7 7 12 36 74 50
Kinetic theory of gases 12 0 14 29 16 21 58 50
Non-equilibrium thermodynamics 9 0 9 0 12 29 70 71
Thermodynamics for biological 2 0 2 14 16 14 79 71
systems











TABLE 5 The most popular textbooks for Course 1 and 2 (percentage of each course).
Books by the same author or team of authors have not been separated since the exact version is often unclear from the
answers. Books listed are those that were used by at least 4% in at least one of the continents. This limitation together
with the large number of courses where locally produced material (about 10%) and books published only in the national
languages leads to the numbers not summing up to 100%.
Author(s) Course 1 Course 2
Europe USA Europe USA
Atkins & de PaulaEs 78] * 18 23 3
Baehr, et al.E9 8 8
C I,.l" ,I 4 2
Elliott & Lira"121 14
Felder & Rosseau[131 11
Gmehling, Kolbe[141 6
K ...1. . :- ,-1 7 6
Prausnitz, et al.[161 6 4
Sandlert6' 13 14
Smith, van Ness, & AbbottE41 11 39 8 43
* Including translations into German and Greek. One instance of Reference 8 was reported for course 1 and at least one instance of Reference 7
was reported for course 2. The rest is mainly Reference 5, but References 5 and 7 were not always discerned by the respondents.


Textbooks Used in TTP Courses
Another issue that reflects the approach to thermodynamics
is the choice of textbooks. The most popular (i.e., used by at
least 4% of the courses in one of the continents) textbooks
in the first two courses are listed in Table 5. Clearly, there is
a difference in the choice of course books between the two
continents although a few popular books are common, as for
example Chemical Engineering Thermodynamics by Smith
et al.41--clearly the most popular TTP book in the United
States. The book was first published about 60 years ago but it
has been thoroughly revised several times to date. The same
applies to the most popular book in Europe, which is the
physical chemistry book by Atkins and co-authors first pub-
lished more than 30 years ago. 51 The fact that it is a physical
chemistry book can be seen as an indication of the emphasis
put on TTP courses in many European universities, or of the
background of the corresponding teacher. The popularity of
the book by Sandler61l may possibly be coupled to the study
of biochemical and biological systems.
A striking fact is that many respondents (about 10%) men-
tion a compendium written for the course as main literature.
This is especially frequent in Europe. It is an indication of
the published textbooks not being appropriate for the course
and could be due to non-overlapping contents between the
available textbooks and the course, lack of textbooks in
the national language, or the professor regarding available
textbooks to be non-pedagogical or too comprehensive (or
perhaps just too expensive).
Structure of TTP Courses
An interesting issue is what methods are used in thermo-
dynamics teaching and whether there are any differences

40


between the continents (cf. Reference 1 for a discussion of
teaching methods). Therefore we asked questions about the
use of different teaching methods in the thermodynamics
courses in the two continents. The answers for the first two
courses mentioned by each respondent are summarized in
Tables 6 and 7.
The teaching of the first two courses appears to be tra-
ditional in both continents. Courses are centered around
lectures and exercise classes with little or no laboratory work
whereas home assignments are given in the vast majority of
the courses. The teaching methods for Course 1 and 2 are
similar except for the case of Problem-Based Learning (PBL)
in the United States, cf. below. An interesting observation
is the fact that for the first course, no work outside class
coupled to the lectures and exercise classes is expected in
about half of the universities in the United States. Instead,
students tend to have home assignments. It can also be noted
that a rather large amount of time is used in class for home
assignments (actually this must be going through the task
and discussing the outcome afterwards). In Europe, stu-
dents appear to be expected to study by themselves without
special assignments as indicated by the amount of time the
students are expected to spend "outside" class on lectures
and exercise classes.
In the first course, there is a significant component of PBL in
more than half of the universities in the United States whereas
it is used in only one-third of the European universities. An
interesting observation is that PBL is more prevalent in the
first course whereas one may have expected it to be used more
in later courses when the students would be expected to have
a greater potential to assimilate such a teaching method.


Chemical Engineering Education











TABLE 6 Time (in hours/course) used for different forms of teaching in Course 1 in Europe
(Values for the USA are given in parenthesis)
Percentage of answers; "(outside)" means "expected student work outside class," PBL= Problem-Based Learning, cf. [1]
Type Oh 1-20 h 21-40 h 41-60 h >60 h
Lectures (in class) -(-) 16(11) 48 (54) 25(30) 11(6)
Lectures (outside) 16(43) 39(33) 28(11) 11(7) 5(6)
Exercise classes 8(11) 44(59) 40(22) 6(7) 2 (-)
Exercise class (outside) 20 (48) 36 (35) 32 (13) 9 (2) 1(2)
PBL etc. (in class) 70 (26) 25 (56) 5 (13) -(4) -(2)
PBL etc. (outside) 66 (44) 29 (32) 4(15) 1(7) -(2)
Home assignment (in class) 41 (20) 4 (37) 8 (28) 6 (9) 1(6)
Home assignment (outside) 34(11) 35(28) 16(20) 1(20) 3 (20)
Laboratory classes 66(83) 25(15) 4 (-) 3 (2) 2 (-)
Lab classes (outside) 78 (83) 18 (13) 3 (4) 1 (-) 1(-)

TABLE 7 Time (in hours/course) used for different forms of teaching in Course 2 in Europe
(Values for the USA are given in parenthesis)
Percentage of answers for Course 2 in Europe and the USA, respectively, cf. Table 6
Type Oh 1-20 h 21-40 h 41-60 h >60 h
Lectures (in class) 3 (-) 19(16) 46 (48) 22 (25) 10(11)
Lectures (outside) 18(16) 35 (39) 25 (28) 3 (11) 5(5)
Exercise classes 10(8) 38 (44) 38 (40) 10(6) 5 (2)
Exercise classes (outside) 24(20) 32 (36) 27 (32) 13 (9) 5(1)
PBL etc (in class) 67 (70) 27 (25) -(5) 5 (-) 2 (-)
PBL etc (outside) 73 (66) 22 (29) 5 (4) - (1) - (-)
Home assignments (in class) 48 (41) 40 (44) 8 (8) 3 (6) 2 (1)
Home assignments (outside) 38 (34) 32 (35) 18(16) 2(1) 11 (3)
Laboratory classes 73 (66) 16 (25) 6 (4) 2 (3) 5 (2)
Lab classes (outside) 81(78) 16 (18) 3 (3) - (1) - (1)


CONCLUSIONS
Classical thermodynamics is (and will probably continue
to be) a core discipline for chemical engineers and it is re-
flected in the invariability of the relevant university courses
for several decades now. Also, the fact that there have been
no profound changes in classical thermodynamics during the
past decades is reflected in this invariability. The most popu-
lar textbook had its first edition 60 years ago and most other
textbooks follow the same outline. More "modem" atomistic
or molecular viewpoints are normally found in the courses
in the late stage of studies (if present at all) where they often
are combined with statistical thermodynamics. In the USA
atomistic/molecular descriptions or explanations seem to be
somewhat more popular than in Europe. The high fraction
of the first thermodynamics course that is studied with other
disciplines of engineering in Europe probably limits the use
of an atomistic approach. Even though the results for teaching
methods are quite similar for the United States and Europe,
a notable difference is the higher amount of problem-based
learning and home assignments in the United States.

Vol. 44, No. 1, Winter 2010


The results presented here may reflect the needs for ther-
modynamics from an industrial perspective. In this respect,
there is an ongoing investigation within the Working Party of
Thermodynamics and Transport Properties of the European
Federation of Chemical Engineering

ACKNOWLEDGMENTS
This study was performed under the auspices of the Work-
ing Party of Thermodynamics and Transport Properties of the
European Federation of Chemical Engineering, wp-ttp.dk>.
Technical support by Peter Sigr6n (at the Centre for Learn-
ing and Teaching, University of Boras) regarding the SPSS
program "MrInterview" is gratefully acknowledged.

REFERENCES*
1. Ramsden, P, Learning to Teach in Higher Education, 2nd ed., Rout-
ledge, London (2003)
2. Dube, S.K., and D.P Visco, "A Survey of the Graduate Thermodynam-

* For textbooks also the year of thefirst edition is given, ifknown.












ics Course in Chemical Engineering Departments Across the United
States," Chem. Eng. Ed., 39(4), 258 (2005)
3. SPSS Dimension, SPSS Inc., 233 S. Wacker Drive, Chicago, Ill.
60606
4. Smith, J.M., H.C. Van Ness, and M.M. Abbott, Introduction to Chemical
Engineering Thermodynamics, McGraw-Hill, Boston (2005) (1st Ed.
1948)
5. Atkins, PW, and J. de Paula, Atkins' Physical ( ...... 8th Ed.,
Oxford University Press, Oxford (2006) (1st Ed. 1978)
6. Sandler, S.I., Chemical, Biochemical, and Engineering Thermodynam-
ics, Wiley, Hoboken, N.J. (2006) (1st Ed. 1977)
7. Atkins, PW, and J. de Paula, Elements ofPhysical ( ...... 4th Ed.,
Oxford University Press, Oxford (2005)
8. Atkins, PW., and L. Jones, Chemical Principles: The Questfor Insight,
Freeman,. New York (2002)
9. Baehr, H.D., and S.Kabelac, Thermodynamik, 13th Ed., Springer, Berlin
(2006) (1st Ed. 1962)
10. Cengel, Y.A., Introduction to Thermodynamics and Heat Transfer, 1st
Ed., McGraw-Hill, Boston (1997) (in 2008 a 2nd Ed. was published)
11. Cengel, Y.A., Thermodynamics: An Engineering Approach, McGraw-
Hill, Boston (2008) (1st Ed. 1989)
12. Elliott, J.R., and C.T. Lira, Introductory Chemical Engineering Ther-
modynamics, Prentice Hall, Upper Saddle River, N.J. (1999)
13. Felder, R.M., and R.W. Rousseau, Elementary Principles of Chemical
Processes, Wiley, Hoboken, N.J. (2005) (1st Ed. 1978)
14. Gmehling, J., and B. Kolbe, Thermodynamik, 2nd Ed., VCH-Verlag,
Weinheim (1992) (1st Ed. 1988)
15. Koretsky, M., Engineering and Chemical Thermodynamics, Wiley,
(2004)
16. Prausnitz, J.M., R.N. Lichtenthaler, and E.G. deAzevedo, Molecular
Thermodynamics ofFluid-Phase Equilibria, Prentice Hall PTR, Upper
Saddle River, N.J. (1999)

APPENDIX: RESPONDING UNIVERSITIES
Austria
Graz University of Technology/Technische Universitit Graz
Belgium
University College Ghent
Katholieke Hogeschool Kempen
University Libre de Bruxelles
Bosnia and Herzegovina
University of Tuzla
Croatia
University of J.J. Strossmayer of Osijek
University of Zagreb
Denmark
Technical University of Denmark
University of Southern Denmark
Estonia
Tallinn University of Technology
Finland
Helsinki University of Technology
Lappeenranta University of Technology (LUT)
University of Oulu
Abo Akademi University
France
Ecole national superieure de chimie de Paris (ENSCP)
University de Paris Sud
University Claude Bernard Lyon 1
Ecole Polytechnique de l'Universit6 de Grenoble 1 - Polytech'Grenoble
Ecole Nationale Superieure des Industries Chimiques - Institut National
Polytechnique de Lorraine (ENSIC - INPL)


Germany
Universitit Duisburg-Essen
Clausthal University of Technology
Universitat Kassel
Brandenburgische Technische Universitat Cottbus
(Brandenburg State University at Cottbus)
University of Kaiserslautern
Dortmund University of Technology
Georg-Simon-Ohm Hochschule Nirnberg
Universitat Stuttgart
Universitaet Karlsruhe
University of Erlangen-Nuremberg
Universitat Bremen
Technische Universitat Dresden
(Dresden University of Technology)
Universitat Bayreuth
Cologne University of Applied Sciences / Fachhochschule Kbln
Hamburg University of Applied Sciences
Technical University Berlin
Fachhochschule Flensburg
(Flensburg University of Applied Sciences)
Hochschule Merseburg (FH)
Technische Universitat Carolo-Wilhelmina zu Braunschweig
RWTH Aachen University
Technische Universitat Hamburg-Harburg
(Hamburg University of Technology)
Hochschule Niederrhein
University of Applied Sciences
University of Siegen
Helmut-Schmidt Universitat der Bundeswehr Hamburg
(Helmut-Schmidt University of the Federal Armed Forces Hamburg)
Leibniz Universitat Hannover
Technische Universitaet Darmstadt
Universitat Kassel
Otto-von-Guericke-Universitat Magdeburg
Ruhr-University Bochum
Greece
Aristotle University of Thessaloniki
University of Patras
National Technical University of Athens
Hungary
University of Debrecen (2)
Italy
University of Pisa
University of Trieste
University di Cagliari
University di Palermo
Netherlands
Delft University of Technology (TU Delft)
Norway
Bergen University College
Portugal
Universidade de Aveiro
University of Minho
Institute Superior Tecnico, Universidade Tecnica de Lisboa
Universidade de Coimbra
University of Porto
Russia
St.Petersburg State University
Lomonosov Moscow State University
Serbia
University of Belgrade


Chemical Engineering Education












Slovenia
University of Maribor
University of Ljubljana
Sweden
Milardalens hbgskola/Malardalen University
Royal Institute of Technology, KTH
Karlstads universitet/Karlstad University
Higskolan i Boras/University of Boras
United Kingdom
Imperial College London (2)
University of Manchester
Newcastle University
University of Birmingham
University College London
London South Bank University
United States ofAmerica
University ofAkron
University of Colorado
Virginia Commonwealth University
Mississippi State University
University of Virginia
University at Buffalo, The State University of New York
University of California Santa Barbara
University of South Alabama
University of Toledo
Case Western Reserve University
South Dakota School of Mines and Technology
Purdue University
Brigham Young University
Villanova University
New Mexico State University
South Dakota School of Mines and Technology
Clarkson University
Iowa State University
University of Rhode Island


Northwestern University
University of Maine
Auburn University
The Ohio State University
UC Davis
Wayne State University
Bucknell University
University of Louisville
University of Maryland, Baltimore County
Polytechnic University
University of Notre Dame
Yale University
University of South Carolina
University of Missouri
Texas Tech University
Stanford University
University of Pittsburgh
University of Kansas
University of Minnesota
Michigan State University
California Institute of Technology
U of Arizona
University of Utah
Rowan University
Cleveland State University
Rice University
Tennessee Technological University
Kansas State University
California State University, Long Beach
Johns Hopkins University
University of Nevada, Reno
Rose-Hulman Institute of Technology
Tuskegee University
West Virginia University
Louisiana Technical University
Vanderbilt University 1


Vol. 44, No. 1, Winter 2010











MR! t classroom
---- --- s_____________________________________


Integrating Academic and Mentoring Support


FOR THE DEVELOPMENT OF FIRST-YEAR


ChE STUDENTS IN HONG KONG



EDMOND I. Ko AND YING CHAU
The Hong Kong University of Science and Technology * Clear Water Bay, Kowloon, Hong Kong


In recognition of the difficulties beginning university
students face in handling the transition from second-
ary school to university,E11 many resources have been
developed to help them deal with the academic, social, and
emotional adjustment issues.[2] Engineering students, in
particular, are often at risk because of a technically demand-
ing curriculum. The situation is even more severe in Hong
Kong, where the normative length of an engineering degree
is three years. This means engineering students face a heavy
workload right from the start, just when they are trying to
adapt to university life. It is thus important to help students
develop effective learning skills and habits early.
To address this issue, the School of Engineering at The
Hong Kong University of Science and Technology has made
provisions for its departments to offer introductory courses
that are specifically targeted at the academic and professional
development of beginning students. In response, the De-
partment of Chemical and Biomolecular Engineering-the
only such department in Hong Kong-has offered a newly
designed course, CENG001, since Fall 2005 to meet the
following key objectives:
1. to help students deal proactively with the transition
from secondary school to university,
2. to familiarize students with the intended learning
outcomes of university and engineering education,
3. to develop professional skills such as the ability to
communicate, to work on multidisciplinary teams,
and to engage in lifelong learning within a conceptual
framework, and
4. to build a learning community among new and senior
students in the department.

To date, the newly designed CENG001 has been offered
four times. About 250 first-year chemical engineering students
44


have taken the course, including every student who is cur-
rently in the department.
In the teaching of CENG001, we have made an attempt to
make reference to other chemical engineering courses that
students are taking in the same semester in order to place the
learning of professional skills in context. Furthermore, we
have integrated the course with the department's mentoring
program in order to provide peer support to the first-year
students. This integration has the added benefit of fostering
a learning community, thus enhancing students' out-of-class
learning and their sense of belonging to the department.
Taking such a holistic approach to promote the personal and
professional development of first-year students is innovative
and (as discussed below) effective.


Edmond I. Ko is an adjunct professor of
chemical and biomolecular engineering at
The Hong Kong University of Science and
Technology. He received his B.S. from the
University of Wisconsin-Madison and his
M.S. and Ph.D. from Stanford University, all
in chemical engineering. His research inter-
ests are in hetero-
geneous catalysis
and engineering
education.

Ying Chau is an assistant professor of
chemical and biomolecular engineering
at The Hong Kong University of Science
and Technology. She received her B.S. in
agricultural and biological engineering from
Cornell University, M.S. in chemical engi-
neering from the University of Pennsylvania,
and Ph.D. in chemical engineering from the
Massachusetts Institute of Technology. Her research interests are in
drug delivery systems, biomaterials, and bionanotechnology.


� Copyright ChE Division of ASEE 2010
Chemical Engineering Education











TABLE 1
Learning Outcomes of CENG001
At the end of this course you (students) should be able to:
1. Explain to a layman (someone without expertise on this subject) the skills and competencies a university graduate in general, and an engi-
neering graduate in particular, should possess.
2. Use your knowledge of Myers-Briggs Type Indicator (MBTI) to develop skills and competencies in areas such as time management,
communication, and teamwork.
3. Identify your strengths and weaknesses in relation to the skills and competencies that are important in a modern workplace.
4. Make concrete plans for personal and professional improvement, especially in areas in which you are currently weak.
5. Develop the habits to become a reflective, self-regulated learner.


COURSE DESCRIPTION
Learning outcomes and course contents of CNEG001 are
shown in Tables 1 and 2, respectively. Briefly, the course fo-
cuses on the academic and social adjustment issues commonly
faced by first-year university students. Topics include the
purposes of university and engineering education, learning,
time management, teamwork, communication and interper-
sonal skills, and goal setting.
The sequence of topics is based on meeting the students'
needs. For example, time management is discussed at about
one month into the semester, just when students are feeling
the pressure to manage their time well in order to deal with
the first round of examinations. Teamwork and presentation
skills are introduced at a time when students have to make
presentations based on group projects in the other chemical
engineering courses they are taking in the same semester. This
coordination puts the learning of these skills in context, and
students have found it useful to be able to apply what they
learn in one course to another.
An important intended learning outcome of the course
(which is also expected of engineering graduates accredited
by signatories of the Washington Accord) is for students to
become self-regulated lifelong learners. To this end the Myers-
Briggs Type Indicator (MBTI), a widely used and researched
instrument on personality types,[] is adopted for students to
gain a better self-understanding. In particular, students learn
about the strengths and blind spots of each MBTI type in situ-
ations such as participating in teamwork, resolving conflicts,
and communicating with people. MBTI thus provides a useful
tool for them to monitor their behaviors and to identify areas
for improvement, which is important for self-regulation.
As the key in self-development rests on a better understand-
ing of oneself, students in CENG001 are asked to participate
in many reflective activities in class, either individually or
in small groups. To the extent possible, these activities in-
volve the applications of MBTI and make reference to other
chemical engineering courses (e.g., CENG152, Introduction
to Environmental Engineering) students are taking in the
same semester, in order to place students' learning in context.
Below are three examples of these activities:


TABLE 2
Course Contents of CENG001
Session # Topic
1 Introduction/Setting expectations
(what are we going to do?)
2 University education
3 Engineering education
4 Myers-Briggs Type Indicator (MBTI)
5 Time management
6 Effective learning
7 Introduction to teamwork
8 Communication skills
9 Presentation skills
10 Conflict resolution skills
11 Making the most out of your university educa-
tion: setting personal and professional goals
12 Reflection/Evaluation
(how well have we done?)

1. Based on a realization ofyour ,.. i,, i i' and weaknesses
as a learner in terms ofyour learning style, identify an area of
learning that you would like to improve the most and formu-
late an action plan. Think on your own first, write your plan
down, then discuss with your neighbor for afew minutes. Be
prepared to share your discussion with the whole class.

2. Think about the team experience you have in your
CENG152 project so far, what has .... , .. .. il and what could
have been done better? In what ways could your knowledge
of MBTI help your team to be more effective in the future?

3. Think of a conflict situation you had in your CENG152
team. Identify the main reason for that conflict, explain it in
terms of your knowledge of MBTI, and propose a solution
should you find yourself in a similar situation again.
These activities serve to promote active and collaborative
learning in class, which is particularly important for begin-
ning university students as such experience may be new to
them. Furthermore, students are asked to submit monthly
reflective statements, taking stock of their experiences in
the past month and making plans for the coming one. The


Vol. 44, No. 1, Winter 2010





























A brief reflective statement of what I have learned from completing this form I


(a) The threefigures represent, respectively, CENG001 students in Fall 2007/
students in Fall 2008/ U.S. students.1/6
(b) I = introversion, E = extroversion, S = sensing, N = intuition, T = thinking, F
J =judging, P =perceiving
(c) Percentages are rounded to the nearest integer and thus do not necessary

form that helps them to develop the habit of learning from
reflection is shown in Table 3.
The course instructor reads all statements and provides
feedback to students both individually and collectively in
class. Faculty tutors and senior student mentors also receive
statements of their mentees. In general, students' reflective
statements reveal the main challenges they face at different
points in the semester (which contains 14 teaching weeks).
In the first month, the key issues are making social adjust-
ments and managing their time. In the second month, they are
trying to establish the right balance between academic and
social activities as they face the first round of examinations.
It is not until the third month that many are able to find their
own rhythm and become comfortable with university life.
Throughout the semester, however, time management remains
their biggest concern.

ASSESSMENT AND VERIFICATION OF MBTI
TYPES
The approach used in assessing and verifying MBTI types
in CENG001 is as follows:


1. Students are asked to complete an online as-
sessment provided by Human Metrics.'41 This
particular instrument was chosen after a pilot
study, in which participants rated its ease of
use and clarity of results to be better than
another online instrument.
2. Students are required to attend a debriefing
session, during which they are asked to iden-
tify their "self-estimated" type.
3. Any differences between the online result and
the self-estimated type are resolved by using a
verification procedure described elsewhere.'s
Verification, which is necessary for identify-
ing the best-fit type, basically involves reading
type descriptions in Introduction to Type and
identifying the one with which the student
agrees the most.


For the students we have worked with, about 55% found
their online type and self-estimated type to be identical, and
another 25-30% reported differences in only one dichotomy.
Type verification thus turned out to be straightforward for most
of the students. For those whose online type and self-estimated
type were different, about half identified their best-fit type to
be the online type; the other half, the self-estimated type. This
shows that a self-estimated exercise is as good as this particular
online instrument. The type distributions of CENGOO1 students
in Fall 2007 and Fall 2008 are shown in Table 4, along with
data for a group of sophomore chemical engineering students
in the United States.[6] These data are shown for reference pur-
poses only because for the sample sizes in this study, there is a
noticeable variation in the two batches of CENG001 students.
The type distributions should thus be seen as snapshots of three
specific groups and not a study comparing chemical engineering
students in Hong Kong and the United States.

PEER MENTORING PROGRAM
The Department of Chemical and Biomolecular Engineer-
ing has established a peer mentoring program since 2005 to


Chemical Engineering Education


TABLE 3
Reflective Statement
Name
Month/Year
Things that I have done well this month
Things that I could have done better this month
The most difficult challenge I faced this month
The most valuable thing I learned this month
My main goals for next month
Activities needed to achieve my goals
Resources needed to achieve my goals
Evidence and measure of success


TABLE 4
Comparison of MBTI Type Distributions(a)
ISTJ(b) ISFJ INFJ INTJ
, , 19%/16%/5% 2%/18%/3% 6%/7%/8%
2%/12 2' I
ISTP ISFP INFP INTP
2%/2%/3% 2%/7%/2% 6%/6%/2% 0%/0%/9%
ESTP ESFP ENFP ENTP
2%/0%/7% 6%/3%/2% 8%/3%/10% 2%/2%/4%
ESTJ ESFJ ENFJ ENTJ
10%/0%/11% 12%/7%/7% 14%/13%/2% 10%/4%/6%


I










Among those who made suggestions

for improvement, over half expressed

that the frequency of meetings could

be increased. This suggested to us

that the problem encountered by some

mentoring groups was due to a lack of

meetings. Overall, the evidence shows

that the nature and content of the

informal gatherings, together with the

discussion on the reflective statements,

was able to increase the students'

awareness and their adaptive ability in

making the transition to university life.



help incoming students adjust academically and socially to
university life. Starting in Fall 2007, activities in the peer
mentoring program have been coordinated with those in
CENG001. Via this partnership, we expect to build a learn-
ing community among new and senior students and, in the
process, extend the benefits of the program and CENG001
from the first-year students to student mentors.
For a class of around 60 students in the first year, we re-
cruited 15 volunteers from the second and third year as peer
mentors. These student mentors were selected by the faculty
coordinator based on their maturity, good communication
skills, and strong commitment to helping first-year students.
Before the academic year began, student mentors were given a
training session, in which they learned about peer mentoring,
the roles and responsibilities of mentors, the overall structure
of the department's peer mentoring program, and more im-
portantly, the common adjustment issues faced by beginning
students. The latter part was extracted from students' feedback
for CENG001 in the past.
Each mentoring group contained one student mentor, four
first-year student mentees, and a faculty tutor. Student mentors
are expected to initiate informal gatherings with their mentees
at least once a month. They are encouraged to discuss adjust-
ment issues with the first-year students using the reflective
statements from CENG001 (Table 3) as a guideline. Their
secondary role is to act as a bridge between mentees and their


faculty tutor. In addition, student mentors are responsible for
proposing and organizing three departmental events during the
first semester that can meet the objectives of the mentoring
program. In Fall 2007 and Fall 2008, the events held were: i)
an orientation session to provide practical tips about university
life for the beginning students; ii) a departmental BBQ mixer
-a social event for all first-year students, student mentors,
and faculty tutors; and iii) a dinner event with chemical en-
gineering alumni.
The department has put several mechanisms in place to
monitor the program. In addition to the availability of faculty
tutors and the mentoring program coordinator, student men-
tors can report to the program coordinator the progress and
problems in a mid-semester review. Suggestions for improve-
ment are followed up either in that semester or in subsequent
years. As an example, the sharing of students' reflective
statements with tutors and mentors was implemented upon
learning that mentors found it difficult to start conversations
with their mentees. The number of students assigned to each
mentor has also been reduced based on the feedback from
mentors. To assess the effectiveness of the mentoring pro-
gram as a whole, both mentors and mentees are surveyed at
the end of the semester with questionnaires and focus group
meetings. The results, discussed below, are used to improve
the program in the following year.

DISCUSSION
Assessment of Learning Outcomes
In the spirit of outcome-based education, the real success of
CENGOO1 must depend on a direct demonstration of students
having attained the intended learning outcomes. Since the
development of skill and affective outcomes takes time, it
is made clear to students (in the course syllabus) that having
an awareness of the importance of a skill is not equivalent to
the actual possession of that skill. Nonetheless, as pointed out
by Kirkpatrick,[7] reaction and learning lay the foundation for
behavior and results; in that light the preliminary feedback
from students has been encouraging. In the end-of-course
evaluations, CENG001 consistently received course and
instructor ratings in the range of 80-90 (out of 100), which
compared favorably with the average scores (70-80) of the
department and of the school. Perhaps more important are the
enthusiastic students' (voluntary) comments, some of which
are produced below.
"(This course) teaches us many ;l,,ir. outside the textbook.
Such ;1t,,. , are probably the most important ;1i,,i. in our
lives. I personally feel that this course lets us know our inner
self better."
"Students who have really paid attention ;Ii. .i . 1,..i the
course should be able to learn how to learn, which is one of
the course e(l... i,., I believe this is crucial."
"Since there is a sudden change from secondary school to


Vol. 44, No. 1, Winter 2010










university, there are a lot ofi;lo,,. , that need to be adapted.
From this course, I can learn a lot of useful skills to deal
with the difficulties faced and have the chance to know more
about myself."
Similar positive comments were made by students in our
regular Faculty-Student Liaison Committee meetings. An-
other source of feedback was available when students were
asked to share what they had learned in completing the reflec-
tive statements. Their responses are revealing, as illustrated
in the following examples.
"Usually I do not sit down and think about how I am cop-
ing with life as an undergraduate. I think this exercise helps
me do so."
"Always think about what we have done, if it can be im-
proved, and set goals for the coming days."
"I have learned 'action' should be done once you have
made a decision. Just 'ild,,i l,. 'of what you are going to do
is not enough."
"Ail.. o. ,:i this reflective statement is the last assignment
of the course, I will continue the habit of learning from re-
flection."
In the reflective statements many students also made explicit
mention of their learning in terms of MBTI, as shown by the
following examples.
"The most important ;,, it. I have learned in this course is
the concept and applications of MBTI. It is really useful for
me to understand myself and even other people more."
"I now know what my MBTI type is and consequently my
preferences in dealing with all sorts of il,, ,.. "
\ ! ) knowledge of MBTI has enabled me to work with oth-
ers more effectively."

Integration With the Peer Mentoring Program
In the end-of-semester evaluation in Fall 2007,46 mentees
(close to 80%) responded to the survey. According to the
student feedback, the mentoring program has been able to
achieve its main objectives: 85% and 91% of the respond-
ing students indicated positively that the overall program
has guided them to make academic and social adjustments,
respectively. There also appeared to be a consensus among
the respondents in identifying the benefits of the program.
On the academic level, the program let them see a clearer
picture of university education in the coming years, and learn
the prospects of chemical engineering. On the social level,
students enjoyed the friendship with their mentors and other
mentees. Through the program, students were able to develop
a sense of belonging to the department in only one semester.
As one student pointed out, "I am not alone."And another has
learned "how to plan for the future and set goals."
A large majority of the first-year students found the infor-
mal gatherings with mentors and departmental events helpful


(83% and 85%, respectively). As CENG001 reflective state-
ments were discussed in some of the informal gatherings, we
polled students whether this kind of mentor-mentee conversa-
tion helped them to discover and solve adjustment problems,
and 83% gave a positive response. Among those who made
suggestions for improvement, over half expressed that the
frequency of meetings could be increased. This suggested to
us that the problem encountered by some mentoring groups
was due to a lack of meetings. Overall, the evidence shows
that the nature and content of the informal gatherings, together
with the discussion on the reflective statements, was able to
increase the students' awareness and their adaptive ability in
making the transition to university life.
According to the feedback of student mentors, the program
was successful in creating a learning community. As effec-
tive communication is a top criterion for establishing a good
relationship with mentees, most student mentors expressed
that the most important thing that they have learned was com-
munication skills. A mentor wrote that he/she "learned to give
advice without bias" and "learned to listen rather than just to
hear." Through organizing the departmental events, students'
leadership skills were also sharpened. Interestingly, like the
mentees, some mentors indicated that the program led them to
better understand themselves and university education. At the
same time, they have developed a stronger sense of belonging
to the department through their service as mentors.
There is an obvious synergy between CENG001 and the
peer mentoring program. To further improve the program
quality, this partnership could be further strengthened; for
example, by providing more targeted mentor training, asking
mentors to attend CENG001 lectures, and relaying the course
news from CENG001 to student mentors on a regular basis.
A quarter of student mentors felt that the mentee to mentor
ratio should be lowered to three such that each mentee could
receive more attention. More active faculty participation in the
mentoring program would also provide different perspectives
for students and increase its impact.

SUMMARY
In this paper we report a simple and effective model, namely
the integration of a first-year development course and a peer
mentoring program, that proactively deals with the adjustment
issues of first-year chemical engineering students. All avail-
able data show that participating students, both mentors and
mentees alike, learn and grow from a purposeful combina-
tion of in-class and out-of-class activities and, in the process,
develop a stronger sense of belonging to the department. As a
result of these preliminary but encouraging findings, the de-
partment has initiated an assessment project to follow a cohort
of students in their development of teamwork competence
throughout the undergraduate curriculum (in particular in
laboratory courses and final-year projects where students work
in teams). This project should lead to more direct evidence


Chemical Engineering Education











of the attainment of key learning outcomes of CENG001, in
particular teamwork skills.
The positive responses of students to CENG001 are similar
to a professional development course one of the authors (Ko)
previously taught. 8" It thus appears that there is room for "non-
technical" courses in an engineering curriculum, especially
in view of the desire to educate well-rounded engineering
students for the modern workplace. Since these courses deal
with generic learning outcomes that are not discipline-specific,
they can be easily adopted by other academic departments
both within and outside engineering.
On a personal note, we have enjoyed being part of this
integrative effort as it has enabled us to interact with many
students at a personal level and thus get to know them and
the problems they face better. It is particularly pleasing that
the approach described in this paper is innovative because
it adheres to an old-fashioned ideal, namely that teaching at
its best is a human activity-in particular about enhancing
human capacity.


REFERENCES

1. Terenzini, PT., The Transition to College: Easing the Passage, a Sum-
mary of the Research Findings of the Out-of-classroom Experience
Program, University Park: National Center on Postsecondary Teaching,
Learning and Assessment (1993)
2. The National Resource Center for The First-Year Experience and
Students in Transition
3. Provost, J.A., and S. Anchors, Using the MBTI Instrument in Colleges
and Universities, Center of Applications of Psychological Types,
Gainesville, FL (2003)
4. Web site for online assessment (an instrument of 72 yes/no questions
for MBTI type assessment) JTypes2.asp>
5. Myers, I.B., Introduction to Type, 6th Ed., CPP, Inc. Palo Alto, CA
(1998)
6. Felder, R.M., G.N. Felder, and E.J. Dietz, "The Effects of Personality
Type on Engineering Student Performance and Attitudes," Journal of
Eng. Ed., 91(1), 3-17 (2002)
7. Kirkpatrick, D.L., and J.D. Kirkpatrick, Evaluating Training Programs,
3rd Ed., Berrett-Koehler Publishers, Inc. San Francisco (2006)
8. Ko, E.I., "A Seminar Course on Professional Development," Chem.
Eng. Ed., 32(3), 234 (1998) 1


Vol. 44, No. 1, Winter 2010











MR]!1 -class and home problems


Development of Problem Sets for K-12 and Engineering on

PHARMACEUTICAL PARTICULATE


SYSTEMS




MARIANO J. SAVELSKI, C. STEWART SLATER, CHRISTOPHER A. DEL VECCHIO,


ADRIAN J. KOSTELESKI, AND SARAH A. WILSON
Rowan University * Glassboro, NJ 08028
Over the past several years, there has been a signifi-
cant effort to develop curricula and coursework for
chemical engineers in biochemical engineering
and biotechnology,1'] but little effort has been undertaken to
develop courses and coursework in the pharmaceutical engi-
neering field. This field involves the manufacture of the active
pharmaceutical ingredients and drugs in the final dosage form.
A successful development of curriculum materials requires
a new approach to integrating concepts of batch processing,
solid-liquid separation techniques, solid-solid particulate
processing, and technology at the nano-scale. The interface
of pharmaceutical science and chemical engineering is crucial
for understanding the basis of structured organic particulate
systems sopsS), a term that describes the multicomponent
organic system that comprises a drug, nutraceutical, or medi-
cine formulation.
It might be argued that a separate curriculum (at a graduate
level) is necessary for a full understanding of this technology,
but one approach fostered by Rowan University and other
institutions is to integrate concepts of new technologies into
the traditional undergraduate chemical engineering curricu-
lum through laboratories/demonstrations, in-class/homework
problems, and case studies. Rowan faculty members have
successfully incorporated the fundamentals and applications
50


Mariano J. Savelski is an associate professor in the Chemical Engineer-
ing Department at Rowan University. He has seven years of industrial
experience in the area of design and optimization of chemical plants. His
research and teaching interests are in optimizing processes for water and
energy reduction; lean manufacturing in food, consumer products, and
pharmaceutical industry; and developing renewable fuels from biomass.
He received his Ph.D. in chemical engineering from the University of
Oklahoma, M.E. in chemical engineering from the University of Tulsa, and
B.S. in chemical engineering from the University of Buenos Aires.
C. Stewart Slater is a professor of chemical engineering and founding
chair of the Chemical Engineering Department at Rowan University. He
has an extensive research and teaching background in separation pro-
cess technology with a particular focus on membrane separation process
research, development and design for green engineering, and pharma-
ceutical and consumer products. He received his Ph.D., M.S., and B.S.
in chemical and biochemical engineering from Rutgers University. Prior
to joining Rowan University he was a professor at Manhattan College.
Christopher A. Del Vecchio is currently an ensign in the U.S. Navy
training to become a submarine officer. He worked on this project during
his junior and senior years at Rowan University. He received his B.S. in
chemical engineering with a concentration in material science.
Adrian J. Kosteleski is currently a senior in the chemical engineering
program at Rowan University, where he has worked on this project.
Sarah Wilson is currently in the doctoral program in chemical engineer-
ing at the University of Massachusetts - Amherst. She worked on this
project during her junior and senior years at Rowan University where she
received a B.S. in chemical engineering. She participated in the ERC
Research Experiences for Undergraduates summer program at Rutgers
University in 2008.

� Copyright ChE Division of ASEE 2010
Chemical Engineering Education


The object of this column is to enhance our readers' collections of interesting and novel prob-
lems in chemical engineering. We request problems that can be used to motivate student learning
by presenting a particular principle in a new light, can be assigned as novel home problems, are
suited for a collaborative learning environment, or demonstrate a cutting-edge application or
principle. Manuscripts should not exceed 14 double-spaced pages and should be accompanied
by the originals of any figures or photographs. Please submit them to Dr. Daina Briedis (e-mail:
briedis@egr.msu.edu), Department of Chemical Engineering and Materials Science, Michigan
State University, East Lansing, MI 48824-1226.










of green engineering into the chemical engineering curriculum
and are currently working to integrate principles of phar-
maceutical particulate systems.[23] This paper describes the
first step in that process: to develop in-class and homework
problems that present key principles of this field that are ap-
propriate to K-12 through chemical engineering courses.
The NSF-sponsored Engineering Research Center for
Structured Organic Particulate Systems (SOPS) led by Rutgers
University () is striving to become a
focal point in pharmaceutical processing. Rowan University
is an outreach partner school of the Center. The work involves
developing coursework that integrates fundamental concepts
and research advances in SOPS into K-12 and undergraduate
engineering curricula. More than 20 problem sets have been
developed and will be integrated into the Center Web site for
dissemination. A study by the AmericanAssociation of Pharma-
ceutical Scientists (AAPS) found that nearly 60% of practicing
industrial scientists believe that there is alack of qualified appli-
cants available with backgrounds in product development and
pharmaceutical technology.[4] The development of educational
outreach programs and course modules could increase student
interest in the area of pharmaceutical engineering from basic
research to manufacturing technology.
A review of the literature shows some K-12 and college-
level educational activities in specific areas of pharmaceutical
science and particle technology. Some experimental methods
have been developed, but synthesis into coursework is not
widespread. Laboratory exercises have been developed to
educate lower-level engineering students on fundamentals
of drug delivery. [5] Experimental methods developed by Nash
illustrate the bulk density, compressibility, flowability, and
other properties of cohesive and non-cohesive pharmaceuti-
cal powders.[6] Simple demonstrations of particle sampling,
flow, and segregation have also been created to display the
unique properties of particles.7- 9] The need for the study of
particle technology in the chemical engineering curriculum
has led to the development of undergraduate and graduate
coursework.[1013]
This paper presents our first steps in our educational
partnership with the Center. The overall goal of this phase
of the project is to develop problem sets that have both an
educational objective that fits into existing coursework and
also a direct link to an area of pharmaceutical research at the
Center. Some of the education topics that have been covered
include: converting values to fractions, decimals, and percent-
ages; solid, liquid, or gas phase characterization; calculating
molecular weight; relative standard deviation calculations;
and solving simple and complex mass and energy balances in-
volving particulate systems. The problems have also covered
the following pharmaceutical processing topics: conversion
from batch to continuous tableting; particle behavior and
segregation; continuous mixers, granulators, and milling
machines; high-pressure homogenization and characterizing


nanoparticles; and drop-on-demand technology. The follow-
ing section will introduce several problems that have been
developed for elementary school to college-level students.
All problems are available on the Resources / Educational
Modules page of along with other
educational materials from the ERC.

PROBLEM 1: WHY POWDERS ARE UNIQUE
Granular materials are unique because they can display the
properties of different states of matter. Particles are able to
support weight, which is a property of a solid; flow from a
container, which is a property of a liquid; and be in compres-
sion, which is a property of a gas. The problem presents the
distinctive properties of particles and asks the student to iden-
tify the property being displayed and connect it to a physical
state. The exercise emphasizes the properties of the different
states of matter while introducing relevant pharmaceutical
engineering concepts. The different properties of powders
can apply directly to pharmaceutical engineering production
techniques. The compressibility of powders is important
for producing pharmaceutical tablets, where differences in
compressibility could alter the ability of the tablet to dissolve
in the body or cause it to be brittle and easily broken. The
powder flowability is important for transportation of powders
during the production process, where batches of powder must
be transported throughout the pharmaceutical plant. This
problem is for a middle school science class.
Problem Statement
Imagine you are walking down the beach. With each step
that you take, the sand moves beneath your feet, but how is
this sand unique? If sand were considered a liquid or a gas,
it would not be able to support the weight of your step. As a
result, in this case, sand must be classified as a solid. Now,
picture building a sand castle. You push the sand down into a
bucket and feel it compress into a harder powder. Are solids
able to be compressed? The only form of matter that can be
compressed under force is a gas. Finally, you have a bucket
full of dry sand. You tip the bucket over to empty it and watch
the sand pour out. If sand were a solid or a gas, it would not
simply flow out of the bucket. Under this condition, the sand
acts as if it were a liquid as it flows out of the bucket. Particle
behavior is difficult to understand because it cannot easily be
classified as a liquid, solid, or gas.[141
Determine the state of matter with properties similar to those
being displayed by particles in the following situations:
a) You pour salt into a salt shaker.
b) You go to shake the salt, but it was packed into the
bottom of the salt shaker when it was filled.
c) You jump up and down on a gravel parking lot, but
your feet do not sink into the gravel.
d) Can you think of any other examples displaying the
complex properties of powders?


Vol. 44, No. 1, Winter 2010











Problem Solution
The problem requires that the student have a strong
knowledge of the properties of a physical state of matter.
Using this knowledge, the student will classify the proper-
ties being displayed by the particulates as similar to a solid,
liquid, or gas.
a) The salt is flowing into the salt shaker, which is a
property of a fluid. This property can be used during
transportation of pharmaceutical powders. The flow of
powders .-,,. , i, pipes and hoppers can , if .. Jiil
affect processing.

b) The salt has been compressed into the bottom of the
salt shaker. The ability to be compressed is a property
of a gas. This property is used to create pharmaceuti-
cal tablets, where powders are compressed to form a
solid dosage form.

c) The gravel can support the weight of your body. This
is a property of a solid material. This understand-
ing of the unique properties of granular materials is
important in specific industrially relevant situations
including powder transportation .n-. -, 1, chutes and
.-, ii, i1,, operations where powders are compressed
into compact pills.

PROBLEM 2: EXAMINING PARTICLE
SEGREGATION WITH THE RSD
The active pharmaceutical ingredient (API) in a pharma-
ceutical tablet is often significantly smaller than the particles
being used as fillers and binders. As a result, the segregation of
the small API particles can detrimentally effect the production
of pharmaceuticals by creating inconsistent tablet composi-
tions. Currently, the Center is using computer simulations to
determine the effects of particle properties on the segregation
of granular mixtures. The relative standard deviation (RSD) is


used to quantify
the degree of seg-
regation within
the mixture. This
problem intro-
duces the concept
of segregation
in pharmaceuti-
cal engineering,
while teaching
students about
RSD calculations.
Particle technol-
ogy concepts are
not always incor-
porated into a tra-
ditional chemical
engineering cur-
riculum. A better
understanding of


granular behavior is important because it can greatly affect
the composition and therefore the efficacy of pharmaceutical
products.[4] This problem is suitable for a freshman or sopho-
more engineering or applied math course.
Problem Statement
When manufacturing pharmaceutical products, it is impor-
tant to limit the segregation of materials to prevent differences
in pharmaceutical powder concentrations. Segregation occurs
when property differences between the particles cause them to
demix. To illustrate segregation in powder a vibrating cylinder
system containing pharmaceutical powders of differing sizes
can be used (Figure la). This geometry is similar to the trans-


Figure 1. Figure la (left) shows a diagram of vibrating
cylinder system with grey API particles and black filler
particles. Figure lb (right) shows the position of the
samples taken from the vibrating cylinder system.


TABLE 1
Number of API Particles in Samples Over Time
Time (s)
0 10 20 30 40 50 60
Sample Number of API Particles
Number
1 10 3 2 0 0 0 0
2 10 6 3 1 0 0 0
3 10 8 5 2 1 0 0
4 10 8 5 3 2 3 1
5 10 9 8 5 7 5 4
6 10 10 10 12 12 11 8
7 10 12 12 15 13 14 16
8 10 15 14 17 16 17 19
9 10 12 18 20 23 24 25
10 10 17 23 25 26 26 27
Sum 100 100 100 100 100 100 100

Chemical Engineering Education










portation of powder products in drums in the pharmaceutical
industry where segregation can occur.
Students are presented with an example of a vibrating
system where segregation occurs between the small gray
particles that represent the API and the large black particles,
that represent the binder and filler particles. By analyzing
the segregation in the system over time, the behavior of the
particles can be examined. One method of analyzing particle
segregation is through the use of the RSD. The equation for
the RSD can be seen in Equations (1), (2), and (3).

# of API Particles in Sample
Total # of Particles
Total # of Particles


RSD =
C


SN -l
N-1


1.2
1

0.8

) 0.6
0.4
0.2


0 10 20 30 40 50 60
Time (s)


Figure 2. RSD of the API particles showing the degree of
segregation as the cylinder vibrates.


0 02 04 06
Height of System (m)


where C is the fully mixed concentration of API particles
(time = 0), C is the concentration of API particles in sample
i, and N is the total number of samples.
Based on the data in Table 1, produce a graph of the RSD
vs. time for the cylindrical system seen in Figure lb.
a) What does it mean when the RSD is equal to zero?
b) What do you think the chart says about the amount of
, ,i ,.,1,. .- over time?
c) Each of the samples shown in Table 1 represents all of
the particles in one disk of the cylinder, as seen in Fig-
ure lb. If the particles reach a height of I m, produce
a graph containing the concentration ofAPI particles
with respect to the height at 0, 30, and 60 seconds.
What happens to the API particles over time?


Problem Solution
The students will convert the number of particles to the
3) concentration of particles in each sample using Equation
(1). Table 1 shows the concentration of API particles in each
sample over time. The concentrations can then be used to
determine the RSD using Equations (2) and (3) and can be
tabulated (see Table 2, on page 57). The students can then
graph the RSD value over time to determine the relationship
between the RSD and the level of segregation in the drug
system (Figure 2). When the RSD is equal to zero, it means
that the system is perfectly mixed and the concentration of
API (small particles) is equal in every sample. As time goes
on, the RSD becomes higher, which means that the system is
becoming segregated. Therefore, if the drugs were formulated
from this mixture, they would have too little or too much API
to meet specifications.


I


08 1


The students can then change the sample number to the
height of particles in the cylinder, with each sample disk being
0.1 m in height, and further calculations can be made. The
students can graph the concentration of API particles vs. the
height of particles to see the distribution of particles over the
height of the system (Figure 3). The graph shows that the API
particles move toward the bottom of the system over time.
The system starts fully mixed, but as it vibrates, the smaller
particles are able to fit through the void spaces between the
filler (larger particles). This type of segregation is called per-
colation and occurs between small APIs and larger filler and
binder particles used to make pharmaceutical tablets.
Drums similar to the cylinder discussed in this problem are
used to transport pharmaceutical particles from site to site
in an industrial setting. It is important that the particle size
distribution remains homogeneous throughout the process to
prevent inconsistencies in processing. Also, when transporting
pharmaceutical powders between equipment (for example
from a mixer to the tablet press) this type of segregation can
occur. The resulting nonhomogeneous mixture can be detri-
mental to the manufacturing process, therefore understanding
this phenomenon is important.


Figure 3. Change in the distribution of API particles over
the height of the vibrating cylinder.

Vol. 44, No. 1, Winter 2010












The goal is to

increase student

interest in research

and development in

the pharmaceutical

industry by

introducing

students to basic

pharmaceutical

engineering

concepts within the

framework of

existing courses.

Educational topics

include the states


of matter,

unit conversions,

material balances,

and relative stan-

dard deviations.


PROBLEM 3: CONTINUOUS POWDER FLOW MIXING
Introductory-level chemical engineering students often learn about material balances
through problems that involve gas and liquid process streams. The problem aims to
emphasize material balances for multi-component solid-phase systems while intro-
ducing the pharmaceutical manufacturing concept of continuous tablet production.
The problem requires the student to perform material balances around a continuous
dry blender. The problem exposes introductory-level chemical engineering students
to novel solids blending technologies being developed in the pharmaceutical industry.
This problem can be used in an introductory sophomore material balance or process
principles class.
Problem Statement
Research is under way to convert traditional batch processes to continuous-flow
operation for pharmaceutical tablet production. The continuous processing techniques
will decrease the costs of labor and time while increasing the uniformity of the final
drug tablets. One of the research areas in pharmaceutical powder flow technologies
examines the cohesive properties in continuous powder flow blending.[15] One of these
experiments examines the properties of pharmaceutical formulations after they have
been blended using a continuous-flow mixer.
The production of acetaminophen tablets used for pain relief requires many unit op-
erations to process the active ingredients and excipients (other ingredients) into tablet
form. One of these unit operations is a powder-blending stage in which powders are
mixed together in preparation for the final tableted product. These powders include:
starch, a binder; acetaminophen and codeine phosphate, the active ingredients in pre-
scription-strength pain medicine; and Avicel microcrystalline cellulose, a filler. Three
feed streams enter a continuous blender; the first contains pure starch at 28 mg/min,
which is combined with a stream of pure cellulose. The third feed stream contains a
mixture of acetaminophen and codeine phosphate in a 10:1 mass ratio, respectively.
The output flow leaves at a rate of 9.3 X 104 lb/min and has a composition of 71.0
wt% acetaminophen. In the next process, magnesium stearate, a lubricant, is blended
with the output mixture.
a) What is the input flow rate of the third stream and what is its composition?
b) What are the compositions of the other components of the output stream?
c) If each tablet contains 300 mg of acetaminophen, how many tablets are being
produced per hour?


Cellulose
mMcc


Starch

mst = 28 mg/mir

API
mAce= 10 mod


Mixing Process


423 mg/min
- mTot

Acetaminophen 71 wt%


Figure 4. Flow diagram of the continuous mixing process.
Chemical Engineering Education










Problem Solution
The student will need to perform a mass balance to solve for the flow rates and compositions of the inlet and outlet streams. A
flow diagram of the process is shown in Figure 4. From the diagram, an overall material balance can be performed to determine
the flow rate of materials in and out of the system (with appropriate unit conversions). From the problem it is known that the
mass of acetaminophen to codeine is 10:1.
m + m2 +m3 +m4 =m5 (4)


28mg mAoe m od mMCC = 9.3x 10-4 lb g = 423g (5)
min mA d min 0.00221b min

The output stream composition from the blender can be used to find the input stream flow rate of acetaminophen.

mA =423mg (0.71)=300 mg (6)
min' min

Ae = 300 mg= 1OmCod ==) C = 30 mg
mmCoin min (7)
Using the total mass balance we can now find the input flow rate of Avicel microcrystalline cellulose.

28mg 11(30 mg +m = 9.3x104 lb g =423 (8)
in min cc min 0.0022 lb min


mMc = 65 (9)
min

Now that the flow rate of each component is calculated, the composition of the output stream from the ribbon blender can be
calculated and number of pills determined.

65 mg
mMcc = un x 100 = 15.4% (10)
423 mg
min

30 mg
mCod = x 100 = 7.0% (11)
423 mg
min

S= 300 mg 1 pill 60 min pills (12)
m = 300 x x - 1 60 (12)
A min 300 mg hr hr

PROBLEM 4: ACETAMINOPHEN DRYING RATE DETERMINATION
Drying is a process not normally taught in the traditional chemical engineering curriculum, but is important in pharmaceuti-
cal manufacture. In this problem, students perform material balances on a dryer and determine the time required to fully dry a
pharmaceutical powder mixture. The problem can be used in material and energy balance and heat transfer classes.
Problem Statement
The pharmaceutical industry currently uses a series of batch processes in the full-scale production of tablets. In one of the
initial steps in manufacturing a pain medicine, acetaminophen and hydroxypropyl cellulose are granulated to create small drug
particles, which form a powder. The hydroxypropyl cellulose is a water-soluble binder that is dissolved in water prior to granu-
lation. After granulation, the water must be removed so that only 2.0 wt % water remains in the powder.[16 Initially, the powder
contains 7,560 g of acetaminophen, 247 g of hydroxypropyl cellulose, and 15.0 wt % " .lt ''i The water is then removed in a
tray dryer, where the powder is placed on trays and hot air is circulated until only 2.0 wt % water remains.
a) Determine how much water must be removed from the powder.
b) The powder is placed on five trays that are 24 inches wide and 30 inches long. The thickness of the powder is 20 mm and the


Vol. 44, No. 1, Winter 2010










bottom and sides of the tray are assumed to be insulated. Dry air at 75 �C (T) and 8.5% humidity flows parallel to each tray at a
rate of 5 mis. Estimate the rate of drying for the constant-rate drying period of the powder.1
c) What will be the minimum time required for the drying of the powder?
Problem Solution
Total Mass = 7,560 g Ace + 247 g Hyd + x g Water (13)

To calculate the amount of water in the initial and final powder stream use the mass fraction values to calculate the grams of water.
x
-77 = 0.15 ==) x = 1,378 g Water (14)
7,807+x

Xf = 0.02 ==) x, = 159 g Water (15)
7,807 + xf

1,378 g water - 159 g water = 1,218 g water is removed (16)

At 8.5% humidity, it was determined that the air absolute humidity (H) is 0.010 kg H20/kg air, the wet bulb temperature (T,)
was found to be 40 �C, and the absolute humidity at saturation (Hw) was 0.05 kg H20/kg air. The humid volume of the air (vH)
was then determined.

m3 22.41 1 1 j m+=1.00 3 (17)
V H M' 22.4TK +H = 1.00 (17)
kg dry air 273 28.97 18.02 kg dry air

Using the humid volume, the density (p) and the mass velocity (G) of the humid air were calculated.
1.0 kg dry air + 0.010 kg water kg
P==1.1 k(18)
m m
1.00
kg dry air

G= vp= 19,800 kg (19)
hr m2

The convective heat transfer coefficient (h in W/m2K) for parallel air flow across the tray was then calculated.

h = 0.0204 G8 = 55.8 (20)
K m2

Using steam tables, it was determined that the latent heat of the vaporization at T, ( w) was equal to 2,574.3 kJ/kg. This can
then be used to find the rate of drying per unit area (Ro) during the constant-drying period.

kg water h T/ 3600 s kg water
Shm2 I hm2

The total rate can then be determined by multiplying the rate of drying per unit area by the area of all of the trays.

kg water . 1 ft 0.3048 m 1ft 0.3048 m
Total Rate = 2.73 k5 trays x 24 mx - x x 30 inx x (22)
h m2 12 in ift 12 in ft

Total Rate = 6.34 kg H20 / hour (23)

The time required for drying can be determined by
Dryg Te Amount of Water to be Removed (
Drying Time = (24)
Constant Drying Rate

. 1218 g HO 1 kg 60 min
Drying Time 1= xg x -1 = 11.5 minutes (25)
6.34 kO 1000 g 1 h


Chemical Engineering Education












SUMMARY

Basic problems on particle technology and processing
relevant to pharmaceutical manufacturing have been devel-
oped for integration into engineering curricula. Educational
topics include the states of matter, unit conversions, mate-
rial balances, and relative standard deviations. The goal is
to increase student interest in research and development in
the pharmaceutical industry by introducing students to basic
pharmaceutical engineering concepts within the framework
of existing courses.


ACKNOWLEDGMENTS

We gratefully acknowledge the assistance of Frank Ro-
manski and Bill Engisch of Rutgers University in reviewing
the course materials. This project has been supported by
National Science Foundation Engineering Research Center
Grant #ECC 0540855.


REFERENCES
1. PR. Westmoreland, "Chemistry and Life Sciences in a New Vision of
Chemical Engineering," Chem. Eng. Ed., 35(4), 248 (2001)
2. Slater, C.S., and R.P Hesketh, "Incorporating Green Engineering into
a Material and Energy Balance Course," Chem. Eng. Ed., 38(1), 48
(2004)
3. Slater, C.S., R.P Hesketh, D. Fichana, J. Henry, A.M. Flynn, and M.
Abraham, "Expanding the Frontiers for Chemical Engineers in Green
Engineering Education," J. Eng. Ed., 23(2), 309 (2007)
4. Augsburger, L.L., "Ensuring the Supply of Highly Qualified Phar-
maceutical Scientist Specialists in Product Development and Related
Technologies for Present and Future Needs -Report of the 2004 PT
Section Education Committee," AAPS PharmSciTech, 8, p. El (2007)


5. Farrell, S., and R.P Hesketh, "An Introduction to Drug Delivery for
Chemical Engineers," Chem. Eng. Ed., 36(2), 198 (2002)
6. Nash, R.A., "A Laboratory Experiment in Pharmaceutics: Macro-
Physical Properties of Pharmaceutical Powders," American J. of
Pharmaceutical Ed., 48, 143 (1984)
7. Ross, M.R., "A Classroom Exercise in Sampling Technique," J. Chem.
Ed., 77, 1015 (2000)
8. Iveson, S.M., and G.V. Franks. "Particle Technology Demonstrations,"
Chem. Eng. Ed., 37(4), 274 (2003)
9. Fritz, M.D., "A Demonstration of Sample Segregation,"J. Chem. Ed.,
82, 255 (2005)
10. Bucala, V., and J. Pifia, "Teaching Population Balances for ChE
Students: Application to Granulation Processes," Chem. Eng. Ed.,
41(3), 209 (2007)
11. Ehrman, S.H., P. Castellanos, V. Dwivedi, and R.B. Diemer, "A
Population Balance Based Design Problem in a Particle Science and
Technology Course," Chem. Eng. Ed., 41(2), 88 (2007)
12. Sinclair, J.L., "A Survey Course in Particle Technology," ( I,.... I .
Ed., 33(4), 266 (1999)
13. Schmid, H., and W Peukert, "Novel Concepts for Teaching Particle
Technology," Chem. Eng. Ed., 36(4), 272 (2002)
14. Geldart, D., "Powder Processing-The Overall View," in Principles
of Powder Technology, M.J. Rhodes, Ed. John Wiley & Sons, New
York (1990)
15. Portillo, PM., M.G. Ierapetritou, and FJ. Muzzio. "Characteriza-
tion of Continuous Convective Powder Mixing Processes," Powder
Technology, 182, 368 (2008)
16. Uhumwangho, M.U., and R.S. Okor, "Modification of Drug Release
fromAcetaminophen Granules by Melt Granulation Technique- Con-
sideration of Release Kinetcis," Pharmaceutical Science, 19(1), 27
(2006)
17. Holinej, J., J.L. Johnson, and L.J. Miller, "Evaluation of Hydroxypro-
pylcellulose as a Dry Tablet Binder in Granulation," Pharmaceutical
Technology Report, Hercules Incorporated (2002)
18. Niazi, S., Handbook of Pharmaceutical Manufacturing Formula-
tions: Compressed Solid Products, Vol 1., CRC Press, New York
(2004) [


TABLE 2
Calculation of RSD from the concentration of API particles

Time (s)
0 10 20 30 40 50 60

Sample Number -_ C1)2

1 0 4.90E-03 6.40E-03 1.00E-02 1.00E-02 1.00E-02 1.00E-02
2 0 1.60E-03 4.90E-03 8.10E-03 1.00E-02 1.00E-02 1.00E-02
3 0 4.00E-04 2.50E-03 6.40E-03 8.10E-03 1.00E-02 1.00E-02
4 0 4.00E-04 2.50E-03 4.90E-03 6.40E-03 4.90E-03 8.10E-03
5 0 1.00E-04 4.00E-04 2.50E-03 9.00E-04 2.50E-03 3.60E-03
6 0 0.00E+00 0.00E+00 4.00E-04 4.00E-04 1.00E-04 4.00E-04
7 0 4.00E-04 4.00E-04 2.50E-03 9.00E-04 1.60E-03 3.60E-03
8 0 2.50E-03 1.60E-03 4.90E-03 3.60E-03 4.90E-03 8.10E-03
9 0 4.00E-04 6.40E-03 1.00E-02 1.69E-02 1.96E-02 2.25E-02
10 0 4.90E-03 1.69E-02 2.25E-02 2.56E-02 2.56E-02 2.89E-02

K- C-)2 S 0 1.56E-02 4.20E-02 7.22E-02 8.28E-02 8.92E-02 1.05E-01

Sum/(N-1) 0 1.73E-03 4.67E-03 8.02E-03 9.20E-03 9.91E-03 1.17E-02
[Sum/(N-1)]12 = 0 4.16E-02 6.83E-02 8.96E-02 9.59E-02 9.96E-02 1.08E-01

/C = RSD 0 4.16E-01 6.83E-01 8.96E-01 9.59E-01 9.96E-01 1.08E+00
Vol. 44, No. 1, Winter 2010 57











|i1 laboratory


TEACHING TECHNICAL WRITING IN


A LAB COURSE IN ChE



STEPHEN J. LOMBARDO
University of Missouri * Columbia, MO 65211


Have you ever been faced with a paragraph of technical
writing such as the following?
An understanding of the characteristics of fluid flow
within cylindrical pipes is an important aspect in the design
and operation of equipment. Conveniently for the engineer,
the flow offluid within a pipe can be generalized in terms of
a single dimensionless number called the Reynolds number
The two general types of flow behavior are called laminar
flow and turbulent flow. Qualitatively, laminar flow cor-
responds to low flow rates in which the streamlines of the
fluid flow are parallel to the line of the bulk flow. As the flow
rate is increased, however, an unstable pattern is eventu-
ally observed in which eddies are present moving in all
directions and at all angles to the bulk flow; this is termed
turbulent flow.
Simply stating to the student "Be more direct" or "Try to be
more cohesive" may not be sufficient for the student to un-
derstand how to improve the paragraph. Pointing out passages
and specific ways to revise them, however, may lead to writing
that is clearer, more concise, and more coherent.
This paper explains how technical writing1 '] is taught in a
laboratory course[6 8] in which 16-48 juniors are organized in
groups of two to four, depending on overall class size. Over
the span of one semester, the students perform the seven
experiments listed below:
" Temperature Measurement and Response Time
" Pressure and Vacuum Measurements
E Viscosity
E Determination of the Compressibility Factor, Z
" Comparison of Flow Measuring Devices for an Incom-
pressible Fluid
E Calibration of an Orifice Meter for a Compressible Fluid
E Laminar and Turbulent Flow
and then present the analysis of data in the form of written
technical reports.


The teaching of writing in a student's major field of uiid ' '"' i
"part of a 30-year-old trend in U.S. higher education known as
'writing across the curriculum' or 'writing in the disciplines'." [10]
At the University of Missouri-Columbia, a Writing Intensive
(WI) course "requires revision as a way of improving critical
thinking ... and ... each course should include at least one
revised writing assignment addressing a question for which
there is more than one acceptable interpretation, explanation,
analysis, or evaluation."E11 To meet the objective of improv-
ing their writing and critical thinking, students are required
to take two 3-credit-hour courses designated as WI, one of
which is an upper division course in their major field of study.
In the Department of Chemical Engineering, WI courses have
been frequently offered either in laboratory courses or in the
capstone design course.
As a consequence of the need for revision combined with
large class sizes, I have developed several methodologies to
teach technical writing in a classroom setting. One goal has
been to transfer to the classroom some of the effort devoted to
commenting on each student's paper, and thereby teach aspects
of good writing collectively rather than individually to each
student via written comments. The methodologies presented
herein pertain to the written component of technical reports
and not to the conventions of technical writing, such as proper
format and correct usage for equations, tables, and figures.
Although these conventions are extremely important and are
treated in the course, other resources are available.11, 12]

Stephen J. Lombardo received his B.S.
degree from Worcester Polytechnic Institute
and his Ph.D. degree from the University of
California at Berkeley, both in chemical engi-
neering. He worked for seven years in industry
in the areas of ceramic materials and ceramic
processing before joining the University of
SMissouri-Columbia in 1997.


� Copyright ChE Division of ASEE 2010


Chemical Engineering Education











LESSONS IN TECHNICAL WRITING
To teach principles of writing, I use the book Style: Lessons
in Clarity and Grace, by Joseph M. Williams.[13] Fundamental
to Williams' approach is that good writers make informed
choices. Although his book is not specifically focused on
technical writing, the principles embodied within the book
are general in nature. The sole caveat is that the writing prin-
ciples may be too far afield from the technical writing of the
students for direct or immediate incorporation into their own
writing. To remedy this, I have adopted lessons covered by
Williams but rewritten them using examples from technical
writing, and, in fact, often with examples that the students
have recently used or will use soon in their own reports. The
balance of this paper explains how examples from Williams'
book are modified to teach chemical engineering students
elements of good writing and good technical writing.
Williams' book is divided into lessons, six of which are
covered in the lab course:
Lesson 3: Actions
Lesson 4: Characters
Lesson 5: Cohesion and Coherence
Lesson 6: Emphasis
Lesson 7: Concision
Lesson 8: Shape
These six lessons also comprise the majority of the two sec-
tions of the book covering "clarity" and "grace," as given in
the book title.
As a first example, Williams teaches in "Lesson 3: Actions,"
that "Sentences are clearer when actions are verbs."[131 An
example he uses is1131:
The Federalists'argument in regard to the destabilization
of government by popular democracy WAS BASED on their
belief in the tendency of factions to FURTHER their self-
interest at the expense of the common good.

In this sentence, actions (in boldfaced type) are not verbs (in
capitalized type) but are nouns. Williams offers an improved
version (indicated by the /) where many of the actions, for-
merly nouns, have now been converted to verbs 131:
/The Federalists ARGUED that popular democracy DE-
STABILIZED government, because they BELIEVED that
factions TENDED TO FURTHER their self-interest at the
expense of the common good.

The original sentence highlights the pitfall of nominalization,
whereby verbs have been converted into nouns. Such writ-
ing, although grammatically correct, often comes across as
abstract and indirect and can be improved by making better
choices as in the revised version.
Williams' lesson, although instructive, may be too far re-
moved for all students to recognize it in their own writing or
to incorporate it directly into technical writing. To bridge this
gap, the pedagogy of Williams' examples has been retained


but now in sentences that are recast as technical writing found
in chemical engineering:
Determination of the accuracy and precision WAS an
important basisfor the selection of a temperature-measur-
ing device.
iWe SELECTED a temperature-measuring device based
on accuracy and precision.
IThe selection of a temperature-measuring device WAS
BASED on accuracy and precision.
The second improved example is offered for writers who es-
chew use of the first person in technical writing. I personally
use the first person in my own writing and allow students to
as well, as long as it is not overdone. I find that use of the first
person is especially effective and efficient when writing more
informal office-type correspondence and industrial technical
reports. I do recommend, however, that superfluous use of the
first person be avoided, such as changing "our data, our equip-
ment, our results" to "/the data, the equipment, the results."
In "Lesson 3: Actions," Williams also addresses how actions
can be hidden in adjectives such as in (adjectives boldfaced,
verbs capitalized)`13]:
The data ARE indicative of the problem.
/The data INDICATE the problem.
The modified versions presented to the students are
The results ARE indicative that the measured values ARE
representative of the bath temperature.
/The results INDICATE that the measured values REP-
RESENT the bath temperature.
In Williams' sentences above, the ideas may not be such a
stretch for the students to understand and apply, but both sets
of examples reinforce the idea that avoiding forms of "to be"
(a weak verb) with adjectives and replacing them with active
verbs (strong verbs) leads to more direct and clearer writing.
In "Lesson 4: Characters," Williams addresses the impor-
tance of having short, specific, and concrete subjects appear
as what he terms the "characters" of sentences. Four examples
are given that demonstrate a progression in the appearance of
the subjects as characters 131:
1. There was a fear that there would be a recommendation
for a budget reduction.
2. The fear of the CIA was that a recommendation from
the president to Congress would be for a reduction in its
budget.
3. The CIA hadfears that the president would send a recom-
mendation to Congress that it make a reduction in its
budget.
4. The CIA feared the president would recommend to Con-
gress that it reduce its budget.
These four sentences, although exhibiting some similari-
ties, differ markedly in directness and specificity. Version
1 exaggeratedly begs the questions"11: Who fears? Who


Vol. 44, No. 1, Winter 2010











The main areas of improvement I have

observed are in concision, addition of

technical content, and in applying
the Old-to-New strategy.

recommends? Who reduces? The second version provides
the characters-the CIA, the president, the Congress-but
they do not appear as subjects of verbs but rather as objects
of prepositions. Version 3 places the characters in the subject
position but remains loaded with nominalization. The final
version remedies this last deficiency...and an interesting
change has taken place. Although Version 4 is about the same
length as the first, it has much more content and provides an-
swers to all of the questions: Who fears? Who recommends?
Who reduces?
To parallel the above example, I present four sentences that
arise from the Viscosity experiment:
1. The solution used was subjected to measurementfor its
viscosity behavior.
2. The solution of 30 mole% glycerol in water used by our
lab group was measured by a capillary viscometer for its
viscosity.
/3a. The viscosity of a 30 mole% glycerol in water solution
was measured with a capillary viscometer.
/3b. A capillary viscometer was used to measure the vis-
cosity of a 30 mole% glycerol in water solution.
/3c. We measured the viscosity of a 30 mole% glycerol in
water solution with a capillary viscometer.
Version 1 begs the questions: Which solution? Used by whom?
Measured by " lha 'Version 2 provides some of these answers
by introducing specific technical content, albeit as objects of
propositions. Versions 3a -3c provide much more content
and have eliminated some low-level content that may not be
necessary. Although we see that in these latter three versions,
the sentences are slightly longer than Version 1, this arises
primarily because of the lengthy but specific description of
the solution.
"Lesson 4: Characters" also treats the relative merits of
passive vs. active voice with an example germane to technical
writers (verbs capitalized, nouns boldfaced) 131:
To determine ifmonokines elicited a response, preparations
.. WERE ADDED....
From a strictly grammatical viewpoint, the introductory clause
is a dangling modifier, because it has an implied subject (I
or we) that is not the same as the subject of the main clause
(preparations). In fact, however, the usage above is so com-
mon, that most readers and writers of technical writing are
perfectly comfortable with such dangling modifiers, especially
if the alternative is�131
To determine if monokines elicited a response, I ADDED
preparations....


To address this issue (but not to resolve it!), I present the
following:
By , ii- i,,,,, into Eq. (1), the viscosity of the mixture WAS
DETERMINED.
/By ,vi- ivimi,, into Eq. (1), we DETERMINED the vis-
cosity of the mixture.
/The viscosity of the mixture WAS DETERMINED from
Eq. (1).

Although these sentences only parallel the ones of Williams,
they do highlight another common weakness in students'
technical writing, namely, low technical content, which can
be remedied as follows:
By ,i. iiii,,,,,, pure component viscosities into Eq. (1), the
viscosity of the glycerol-water solution was determined to
be 10 cP at 298 K.

/By ,ii ii ii,,,,, pure component viscosities into Eq. (1),
we determined the viscosity of the glycerol-water solution to
be 10 cP at 298 K.

In "Lesson 5: Cohesion and Coherence," Williams ex-
plains how to lend a sense of cohesion to writing. He offers
two sentences, one with an active verb and one with a pas-
sive verb"31:
A) The collapse of a dead star into a point perhaps no
larger than a marble creates a black hole.
B) A black hole is created by the collapse of a dead star into
a point perhaps no larger than a marble.

He then asks which sentence, A or B, fits better in the passage
given below131:
Some astonishing questions about the nature of the Universe
have been raised by scientists exploring black holes in space.
[A or B] So much matter compressed into so little volume
changes the fabric of space around it in puzzling ways.
As indicated below, sentence B lends a sense of cohesion
to the passage:
JSome astonishing questions about the nature of the Uni-
verse have been raised by scientists exploring black holes
in space. A black hole is created by the collapse of a dead
star into a point perhaps no larger than a marble. So
much matter compressed into so little volume changes the
fabric of space around it in puzzling ways.
The reason is that readers find text more coherent when
sentences end with new information (in bold) and then be-
gin sentences with old information (in bold italics). This is
called the Old-to-New strategy. My experience is that very
few undergraduate students are familiar with this strategy for
achieving coherence in writing. To apply this Old-to-New
technique, I present to the students the five-sentence passage
which began this article:
'An understanding of the characteristics of fluid flow
within cylindrical pipes is an important aspect in the
design and operation of equipment. 2Conveniently for


Chemical Engineering Education










the engineer, the flow offluid within a pipe can be general-
ized in terms of a single dimensionless number called the
Reynolds number. 3The two general types of flow behavior
are called laminar flow and turbulent flow. 4Qualita-
tively, laminarflow corresponds to low flow rates in which
the streamlines of the fluid flow are parallel to the line of
the bulk flow. 5As the flow rate is increased, however, an
unstable pattern is eventually observed in which eddies are
present moving in all directions and at all angles to the bulk
flow; this is termed turbulent flow.

The first sentence (Sentence 1) introduces a number of pos-
sible new ideas (bold type) as to the topic of the paragraph,
and in fact introduces too many ideas. Sentence 2 violates
the Old-to-New strategy by not beginning with Old infor-
mation (bold italics), but instead introduces additional New
information, namely a new character "the engineer," in the
form of a dangling modifier. Sentence 2 next proceeds to the
"flow of fluid," which is Old information, and then introduces
New information, the Reynolds number. In sentence 3, the
Reynolds number should now represent Old information, but
instead this topic is dropped and two types of flow are now
introduced, which leads to further new information, namely
laminarr flow" and "turbulent flow." Sentences 4 and 5 then
proceed to follow the Old-to-New strategy by defining laminar
and turbulent flow. The location of turbulent flow at the end
of the fifth sentence, however, is where readers expect to find
new information.
An improved version of the paragraph, which more closely
adheres to the Old-to-New strategy, is given below:

/'Two general types offluid flow behavior are observed
and these are referred to as laminar and turbulent flow.
2Qualitatively, laminar flow corresponds to low flow rates
in which the streamlines of the fluid flow are parallel to the
line of the bulk flow. 3As the flow rate is increased, however,
an unstable pattern is eventually observed in which eddies
are present moving in all directions and at all angles to the
bulk flow; this is termed turbulent flow. 'These two types
of flow behavior can be predicted by a single dimensionless
number called the Reynolds number.
In the version above, the topic sentence focuses on fluid flow
behavior, and introduces the New information of laminar and
turbulent flow. Laminar flow now appears as Old information
in the sentence 2, and is explained as New information by
"low flow rates." In sentence 3, the phrase "as the flow rate
is increased" is the Old information in the form of a subtle,
indirect link back to "low flow rates" from the preceding
sentence. In sentence 4, the connection to previous informa-
tion is straightforward, and the paragraph is then summarized
with the introduction of the concept of the Reynolds number.
(Presumably, the paragraph could be extended with a defini-
tion of the Reynolds number and the corresponding regimes
for laminar and turbulent flow. Alternatively, a new paragraph
of the same content could begin with a paragraph transition
back to the idea of the Reynolds number.)


The Old-to-New Strategy thus reduces each paragraph to the
introduction of new information in a topic sentence followed
by a succession of sentences that proceed
0 New. First Sentence (topic sentence with transi-
tion from preceding paragraph)
Old - New.

Old - New. Last sentence (concluding sentence)

In "Lesson 7: Concision," several strategies are presented
for achieving concision, such as deleting meaningless words
and doubled words. I combine this aspect of writing along
with adding technical content to sentences, as indicated in
the following examples.
The viscosity was basically measured with an accurate
device called the Brookfield rheometer.

/The viscosity of the glycerol-water solution was measured
with a Brookfield rheometer from 10-60 rpm.
The full and complete data are in Table 1 for each and every
shear rate.

iThe viscosity data for the glycerol-water solution at 25 �C
are in Table 1.

In "Lesson 8: Shape," Williams teaches that writers should
start sentences with subjects, next get to the verb quickly, and
then get to the object. I use the following examples to make
the same points with subjects boldfaced, verbs capitalized,
and objects underlined:
A pump for the process based onflow rate and efficiency
WILL BE SELECTED.

/A pump WILL BE SELECTED for the process based on
flow rate and efficiency.
We WILL SELECT for the process a pump based on flow
rate and efficiency.

JWe WILL SELECT a puma for the process based onflow
rate and efficiency.

The first pair of sentences above highlights how the verb can
be moved closer to the subject, and the last pair indicates how
the object can be brought closer to the verb.

STUDENT FEEDBACK AND ASSESSMENT
As is indicated in Table 1 (next page), students had a num-
ber of positive comments to the approach presented herein
on teaching technical writing, and these reflect the majority
of student responses. Another type of receptive feedback is
that students have asked me to provide them, via the Internet
or other means, with the examples herein.
Table 1 also presents some negative comments, which
reflect the students' desire for exposure to more examples of
technical writing and for the instructor to be more accepting of
different writing styles. Several of these comments highlight


Vol. 44, No. 1, Winter 2010











the peril of attempting to modify students' writing without
sufficient appreciation for the difficulty of the task and for
the sensitivity of the students.
Although the teaching of writing is difficult, both students
and I generally notice an improvement in their writing. In
response to the query "Please rate, in your own estimation,
how much your technical writing improved" the students
rated their improvement 3.4 on a scale of 1=improved a lot
to 9=improved very little. The main areas of improvement I
have observed are in concision, addition of technical content,
and in applying the Old-to-New strategy.

SUMMARY
In this paper, examples are presented for teaching technical
writing in a laboratory course to undergraduate students. The
lessons presented to the students are adapted from the book
Style: Lessons in Clarity and Grace, by Joseph M. Williams.
Williams emphasizes the importance of informed choices in
writing. His lessons are modified to bring them nearer to the
technical writing the students are learning and practicing in
the course, with the hope that the students will thus better be
able to incorporate these ideas into their own writing.
As a side note, the medium for presenting examples of
technical writing in the classroom has evolved from overhead
transparencies to use of an interactive whiteboard to computer
projection of word processing documents. The interactive
whiteboard was especially effective in that I could edit and
highlight, in color, in front of the students. Most recently, I
have been using computer projection with colored text to
illustrate the principles.
The writing lessons covered herein are generally presented
to the students by first showing them the uncorrected versions
and then eliciting their comments on ways for improvement.
After this group discussion, the principle of the lesson is
presented and further practiced interactively with additional
examples. The exception to this purely interactive approach
is for the Old-to-New strategy. Because this strategy is un-
familiar to many students and may appear complicated at
first glance, I normally spend parts of two lectures teaching
it. Sometimes I distribute paragraphs to the students in class
and ask them to analyze the writing to see if it adheres to or
violates the Old-to-New strategy. Other times, I have them
analyze one or more of their own paragraphs.

ACKNOWLEDGMENT
The content of this paper draws heavily from Williams'
book, and he additionally deserves credit for the formatting
conventions I have used here to help illustrate the ideas in
black-and-white print. I am also indebted to the Campus Writing
Program at the University of Missouri-Columbia and to Martha
D. Patton, who always pointed me in the right direction.


TABLE 1
Selected Comments From Students
on Writing Instruction.
Positive:
It was an excellent class for writing within the major.
The balance between technical learning and writing technique
allowed for the development of important application skills that
had remained unaddressed until then.
Explanations and examples of writing techniques were helpful.
Giving specific examples of certain writing to aid our writing
really helped.
Excellent instruction on writing techniques.
Good examples of better writing and revision.
Learning how to technically write seemed like a very useful tool
for not only later classes, but also future jobs.
Examples of good vs. bad writing. Active discussion and in-class
revision practice.
Negative:
I recognize that the writing techniques are very good, but I feel
there could be a greater acceptance of other writing styles.
Put notes (examples of writing) on a Web site.
The text bothered me, I can't say that I read it, but some of the
concepts it presented seemed odd.
Would like to see more examples of students' reports presented and
criticized in class.
Be more accepting of other people's writing.

REFERENCES
1. Gopen, G.D., and J.A. Swan, "The Science of Scientific Writing,"
Amer. Sci., 78, 550 (1990)
2. Day, R.A., Scientific English: A Guide for Scientists and Other Profes-
sionals, Oryx, Phoenix (1995)
3. Friedly, J.C., 'Top Ten Ways to Improve Technical Writing," Chem.
Eng. Ed., 38, 54 (2004)
4. Sharp, J.E., B.M. Olds, R.L. Miller, and M.A. Dyrud, "Four Effec-
tive Writing Strategies for Engineering Classes," J. Eng. Ed., 88, 53
(1999)
5. Prausnitz, M.R., and M.J. Bradley, "Effective Communication for
Professional Engineering Beyond Problem Sets and Lab Reports,"
Chem. Eng. Ed., 34, 234 (2000)
6. Newell, J.A., D.K. Ludlow, and S.PK. Sternberg, "Development of
Oral and Written Communication Skills Across an Integrated Labora-
tory Sequence," Chem. Eng. Ed., 31, 116 (1997)
7. 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)
8. Sharp, J.E., "Teaching Strategies for Integrating Communication in the
Chemical Engineering Lab," ASEE Annual Conference Proceedings,
4555(2003)
9. , accessed June 2009
10. , accessed June 2009
11. Beall, H., and J. Trimbur, A Short Guide to Writing about ( f.......
Longman, New York (2001)
12. Dodd, J.S., ed., The ACS Style Guide, American Chemical Society,
Washington, D.C. (1997)
13. Williams, J.M., Style: Lessons in Clarity and Grace, 9th Ed., Pearson
Education, New York (2007), plus earlier editions 7


Chemical Engineering Education











Random Thoughts...



HARD ASSESSMENT OF SOFT SKILLS


RICHARD M. FIELDER
North Carolina State University
REBECCA BRENT
Education Designs, Inc.
o, you've been told that as part of your department's plan
for addressing the ABET (or Bologna or Washington
Accord) accreditation criteria, you've got to teach your
students how to communicate effectively and/or discuss engi-
neering solutions to social problems and/or analyze and resolve
ethical dilemmas. You just have two small problems to solve.
First, how do you teach those skills when (if you're like most
of us) no one ever taught them to you? Second, how do you
assess how well your students have mastered the skills?
Let's look at the assessment question first. In most engineer-
ing and science courses, the things we grade are mainly solu-
tions to quantitative problems, short answers to closed-ended
questions, and multiple-choice test items. You can grade those
things objectively without much difficulty as long as the ques-
tions are clear, the correct answers are not a matter of opinion,
and the grader awards points consistently. It's a different story
when it comes to grading essays and written and oral project
reports. Since there are no unique "correct answers," subjective
and inconsistent judgment calls often contaminate the grading.
When that happens, student resentment and complaints can
quickly get out of hand, and many students never learn the skills
you are trying to teach because they don't really understand
the criteria they are being graded by.
The challenge in evaluating "soft" student products is to find
a grading process that is reliable (a given product would get
almost identical marks from two or more expert graders and
from the same grader at different times), andfair (the students
know the grading criteria- what counts, and by how much; the
grading is based entirely on the criteria; and the students have
been adequately instructed in the methods and skills required
to meet the criteria). Two types of instruments -checklists and
rubrics-can provide both reliability and fairness.
A grading checklist is a form that lists the instructor's
grading criteria and the maximum points allocated to each
criterion. The instructor assigns up to the maximum points
for each criterion and totals the points to determine the final
assignment grade. Table 1 shows an illustrative checklist
for written reports."1 A grading rubric also lists the grading
criteria, but now the instructor scores each one on a discrete
scale (e.g., 5-4-3-2-1 or 4-3-2-1) and gives brief descriptions
of what each numerical rating represents. The overall product
Vol. 44, No. 1, Winter 2010


grade is determined as a weighted sum of the points given
for each criterion, with each weight representing the relative
importance of that criterion to the instructor. Table 2 shows
an illustrative excerpt from a rubric used for rating individual
team member performance in group projects.[2] Another good
example is a rubric designed to evaluate both individual and
team performance on an oral project report in an engineering
design course.[31 Creating rubrics is made easy by a free online
tool called I.' ni.,i ,' ().
Once you have a checklist or rubric, grading student work
becomes much more efficient than the usual procedure in
which detailed feedback is provided on each student product,
and more reliable because the breakdown of points by criteria
makes it more likely that products of the same quality will
get the same grade. The students can quickly see why they
got the grades they did and where their work fell short of
your expectations, and they should emerge with a clear idea
of what they need to do to raise their grades on subsequent
assignments. The system is even more effective if you use two
raters (e.g., you and a trained teaching assistant, or two teach-
ing assistants) to grade all student products. If the raters fill
out their forms independently and then reconcile their ratings,
the grading will be at least as objective and reliable as what
we normally do for quantitative problem-solving tests.
Checklists and rubrics are also excellent tools for teach-
ing students the procedures and skills required to meet the
instructor's expectations. For example, suppose you plan to
give several writing assignments in your class. Before the
students begin work on the first one, give them two sample
completed assignments-one that makes most of the mistakes
you anticipate your students will make in their initial efforts
(inadequate background discussion, unsupported conclusions,
bad grammar, sloppy graphics, etc.), and one much better. In
or out of class, have them individually use your checklist or
rubric to grade the first sample assignment, and then in class
have them work in pairs to reconcile their ratings. If they can't
agree on a rating, have them average their individual scores.
Poll them to get their reconciled ratings for each criterion,
and then tell them the ratings you would have given and why.
Repeat the exercise with the second assignment. At that point
the students will have a clear idea of what you are looking
for, and their products will on average be far better than those
you are used to seeing.
Similarly, before your students give oral reports, give a short
illustrative talk yourself and deliberately make the common
@ Copyright ChE Division of ASEE 2010












mistakes you know many of them will make (read directly
from a manuscript, lecture to the board and mumble, rapidly
show lots of PowerPoint slides with garish backgrounds and
cluttered text, and so on). The students will quickly see what
you're doing and start laughing and maybe throwing things at
you. Stop after a couple of minutes and have them individually
rate your talk using your checklist or rubric and then reconcile
their ratings in pairs. After you discuss the ratings and give
yours, give a better talk that still has mistakes, and repeat the
exercise. Prepare to be pleasantly surprised at the quality of
the oral reports most of them subsequently give. 41


REFERENCES
1. Checklist designed by Professor Lisa Bullard, North Carolina State
University, and reproduced with permission.
2. CATME (Comprehensive Assessment of Team Member Effectiveness)
Web site,
3. Welch, H., D. Suri, and E. Durant, "Rubrics forAssessing Oral Commu-
nication in the Capstone Design Experience: Development, Application,
Analysis, and Refinement," Intl. J. Engr. Ed., 25(5), 952 (2009)
4. For additional tips on teaching and assessment strategies that address
each of ABET Outcomes 3a-3k, see Felder, R.M., and R. Brent,
"Designing and Teaching Courses to Satisfy the ABET Engineering
Criteria," J. Engr. Education, 92(1), 7 (2003) felder-public/Papers/ABETPaper (JEE).pdf> 7


TABLE 1
Grading Checklist for Written Reports"'
Student: Project Phase:
Date: Evaluator:
Max. Score Comments
Technical Content (60%)
Abstract clearly identifies purpose and summarizes principal content 10
Introduction demonstrates thorough knowledge of relevant background and prior work 15
Analysis and discussion demonstrate good subject mastery 30
Summary and conclusions appropriate and complete 5
Organization (10%)
Distinct introduction, body, conclusions 5
Content clearly and logically organized, good transitions 5
Presentation (20%)
Correct spelling, grammar, and syntax 10
Clear and easy to read 10
Quality of Layout and Graphics (10%) 10
TOTAL SCORE 100

TABLE 2
Excerpt from a Peer Rating Rubric for Team Projects[21
TEAM NAME / NUMBER:
Your
- Write the names of the people on your team.
name
Contributing to the Team's Work
* Does more or higher-quality work than expected.
S 5 * Makes important contributions that improve the team's work.

* Helps teammates who are having difficulty completing their work.
4 4 4 4 4 Demonstrates behaviors described in both 3 and 5.
* Completes a fair share of the team's work with acceptable quality.
3 3 3 3 3
* Keeps commitments and completes assignments on time.
* Helps teammates who are having difficulty when it is easy or important.
2 2 2 2 2 Demonstrates behaviors described in both 1 and 3.
* Does not do a fair share of the team's work. Delivers sloppy or incomplete work.
1 1 1 1 1
* Misses deadlines. Is late, unprepared, or absent for team meetings.

* Does not assist teammates. Quits if the work becomes difficult.
Interacting with Teammates
5 5 5 5 5 ...

Chemical Engineering Education











classroom
----- --- s___________________________________________


TWO UNDERGRADUATE


PROCESS MODELING COURSES

Taught Using Inductive Learning Methods


MASOUD SOROUSH AND CHARLES B. WEINBERGER
Drexel University * Philadelphia, PA 19104
Application of process models in the process industries
has increased enormously in the past two decades.
Significant advances in computer hardware and soft-
ware, process modeling, and numerical methods have made
possible the sharp rise in the development and application
of process models. Among these advances, the availability
of increasingly fast computers at very low prices stands out.
Process models are currently used in process design, process
optimization, model-based control, process monitoring,
trouble shooting, safety and flexibility analyses, design of
experiments, and personnel training, among other areas. With
an increasing use of process models in the process indus-
tries, changes have been made in our chemical engineering
undergraduate curriculum to equip graduates with adequate
process-modeling skills.
Efforts have also been made in other chemical engineering
departments and engineering disciplines to respond to the
increasing use of process models in the process industries.
Foss and Stephanopoulos[11 developed an approach to pro-
cess modeling in which students are led to crafting a process
model before writing any equations. To this end, students
are led through a structured modeling methodology with
which the physics and phenomena of the process are identi-
fied and engineering science concepts placed into a model
structure using their developed software. High and Maase[2]
developed a graduate process modeling course that involves
using MATLAB, which students need to solve equations
governing process models. Their course uses case studies,
active problem solving, teamwork, and experimentation to
promote creative and critical thinking in the students. No-
cito-Gobel, et al.,[3] developed a multidisciplinary modeling
course on mathematical modeling. Pang[4] proposed teaching
chemical engineering concepts using plant models. Rappin,


et al.,[5] developed a software tool that allows students to
model and simulate chemical engineering processes. Rabb
and Chang[6] described an interdisciplinary dynamic model-
ing and control course with students and instructors from
several engineering disciplines. Layton 71 used modeling and
simulation projects to improve learning objectives in a course
on systems dynamics.
Inductive teaching methods have received more attention
in recent years. In inductive teaching, specific observations,
case studies, or examples are presented first, and the general
theory is then taught or discovered by the students with the
instructor's help after the need to know the theory has been


Masoud Soroush received his B.S. in chemi-
cal engineering (1985) from Abadan Institute
of Technology, Iran. He received his M.S.
(chemical engineering, 1988, and electrical
engineering: systems, 1991) and Ph.D.
(chemical engineering, 1992) degrees all
from the University of Michigan, Ann Arbor.
He is a professor of chemical and biological
engineering at Drexel University. His cur-
rent research interests are in biomedical
engineering, process systems engineering,
quantum chemistry, polymer engineering,
and fuel cells.
Charles B. Weinberger received his B.S. in
chemical engineering (1963) from the Univer-
sity of California, Berkeley, and M.S. (1964)
and Ph.D. (1970) degrees both in chemical
engineering from the University of Michigan,
Ann Arbor. He is a professor of chemical and
biological engineering at Drexel University.
His research interests are in polymer process-
ing, fluid mechanics of multiphase systems,
extensional-flow theology, and pneumatic
transport of particles.

� Copyright ChE Division of ASEE 2010


Vol. 44, No. 1, Winter 2010










established.[8] Inductive teaching methods include project-based
learning, case-based teaching, discovery learning, inquiry
learning, problem-based learning, and just-in-time teaching.
Inductive methods have been found to be in general more ef-
fective than traditional deductive methods, in which theories
are first given, followed by applications of the theories.[8] Ac-
tive learning methods,[9] which actively engage students in the
learning process, are increasingly used in classes. In contrast,
in traditional learning methods, students receive information
from the instructor passively. The two courses described herein
employ both active learning and inductive methods.
This manuscript describes two core chemical engineer-
ing courses, each four quarter-credit hours, that were
developed and introduced into the Chemical Engineering
Undergraduate Curriculum at Drexel University in 1996.
The objectives were threefold: first, to fill the gap that
existed between the mathematics courses taken during the
freshman year and the chemical engineering courses taken
in the following years; second, to improve engineering
judgment of chemical engineering students; and third, to
provide the students with a strong, lasting background in
process modeling that enables them to attack and solve
open-ended process modeling problems systematically.
The two courses, Process Modeling I and II, have proven
successful in achieving the aforementioned objectives.
They have been offered in pre-junior and junior years
(Years 3 and 4 of our 5-year undergraduate program),
respectively; the students take the first modeling course
after they have just taken Material and Energy Balances.
Some 95% of Drexel undergraduate chemical engineer-
ing students select a five-year program including one and
one-half years of co-op experience in industry. The open-
ended nature of homework problems and the richness of
lecture contents, including simple, physically meaningful
examples from different disciplines, are among the major
features of the two courses. These two courses differ from
previous courses taken by the students in several aspects,
but perhaps the most important is that the students derive
equations from physical problem descriptions, rather than
plug numbers into previously derived equations. This equa-
tion derivation task represents an active learning process,[9]
which requires the students to make a number of engineer-
ing judgments to solve process modeling problems in class.
Many students view this as a major challenge initially in
Process Modeling I, but they soon meet the challenge.
The students' in-class activity and engagement in solving
modeling problems included in lectures represents an ac-
tive learning process.
This paper is organized as follows. Section 1 describes
common features of the two courses. Sections 2 and 3 then
explain the specifics of Process Modeling I and II, respec-
tively. Assessment data and sample students' comments on the
courses are included in section 4. Finally, concluding remarks
are given in section 5.
66


1. INSTRUCTION METHOD

An instruction method that is flexible, interactive, and
hands-on is used in the two courses. The key instruction
focus in these courses is in-class learning rather than cov-
ering a priori-set amount of topics in a lecture. In lectures,
students think through concepts, solving a set of simple,
carefully chosen and arranged, physically meaningful prob-
lems. Students actively and directly participate in in-class
problem-solving efforts, either individually or as a team
(students decide), although they may work at their own rate.
Process Modeling I students do this by applying process
modeling knowledge or skills learned earlier in Process
Modeling I, and Process Modeling II students by knowledge
gained in Process Modeling I, Process Modeling II, and
other previously taken chemical engineering courses. Note
that the same method of process modeling taught in Process
Modeling I is used in Process Modeling II. Mathematics is
kept in the background until model predictions are needed;
it comes into play when process-model equations should
be solved to predict the process behavior. The only partial
differential equation solved in Process Modeling II is that
of a semi-infinite body, for which there exists a closed-form
analytical solution in terms of an error function. These
courses are offered twice each year, as Drexel has four
quarters in each year. Process Modeling I is offered in Fall
and Spring Quarters, while Process Modeling II is offered
in Winter and Summer Quarters. The number of students in
each class ranges from 25 to 40.
Students solve process modeling problems in class without
the instructor's direct guidance. The instructor often serves
as a "referee" in the class discussions and provides hints in
the form of questions (if such hints are necessary). After most
students have gone through all of the process modeling steps,
one or more student volunteers are chosen by the instructor to
write their solutions) on the board. Other students are then
asked to evaluate the correctness of the process models)
written on the board. The volunteer students should explain,
defend, and correct their modelss, if a correction is necessary.
At the end of each process modeling problem, the instructor
summarizes the class discussions to highlight the key concepts
in the problem.

2. PROCESS MODELING I
This course begins with a review of the applications of
process models in the process industries and a description of
a mathematical model; that is, a set of equations that describe
the relations among relevant variables (that are of interest to
the user) in a process. The rest of the course is devoted to
how to derive the set of model equations by taking students
through derivations of model equations for about 50 process
examples carefully selected from several disciplines. Some
of these examples are listed in Table 1. Each process model
has four main components:
Chemical Engineering Education











L Conservation equations

II. Constitutive equations

III. Constraints on process variables

IV Conditionsfor the dependent variables
In developing macroscopic models, the conservation laws
(species, mass, momentum, and energy), which serve as the
main pillars of the models, are closed through constitutive
equations (e.g., reaction rate equations, ideal gas law, Fick's
law of diffusion, Fourier's law of conduction, Newton's law
of viscosity, and Stefan-Boltzmann law of radiation). Given
the same input (information) that is provided for the process
under consideration, these equations should allow one to de-
scribe/predict the process behavior of interest. Constraints on
process variables define the regions in which process model
predictions are physically meaningful; they provide the model
equations with additional physical and chemical realities.
For example, a steady state model, consisting of a set of al-
gebraic equations, can have a solution with a negative steady
state concentration, which is not physically meaningful. The
constraints are also helpful in solving the model equations
numerically. The conditions allow one to solve governing dif-
ferential equations and predict uniquely the process behavior.
Principal topics covered in lectures include:

* Macroscopic first-principles models for simple lumped-
parameter dynamic processes

* Steady state behavior

* Numerical solution to a set of algebraic equations (New-
ton-Raphson and secant methods)

* Macroscopic first-principles models for simple spatially
distributed steady state processes

* Numerical .,, i ,i,, ii. - (Euler's, implicit modified Euler's,
and explicit modified Euler's methods)

* Compartmental modeling of spatially distributed pro-
cesses

* Lumped-parameter modeling of processes with imperfect
mixing
Coverage of these numerical methods ensures that all stu-
dents have adequate knowledge of those algorithms behind
equation solvers available on computers. Students in Process
Modeling II use computers extensively to solve model equa-
tions numerically.
The roughly 50 examples covered during lectures are care-
fully selected from several different disciplines and arranged
in a proper sequence, so that each example is used to teach a
concept that is required in process model development of a
subsequent example. A list of some of these examples is given
in Table 1. The examples' diversity emphasizes to the students
that the modeling method is systematic and general; that is, the
method can be applied to any process, whether it is a chemical
plant, the human body, or a collection of living organisms.
Vol. 44, No. 1, Winter 2010


Process Modeling I is offered before Fluid Dynamics,
Mass Transfer, Heat Transfer, and Reaction Kinetics. An
inductive teaching method&81 is used in this course. The con-
stitutive equations needed in Process Modeling I but taught
in detail later in these four chemical engineering courses are
derived in Process Modeling I by calling on the students'
intuition. Their intuition is invoked frequently to derive
the functional or specific form of the constitutive equations
that the students have not seen previously. For example,
the functional form of a constitutive equation describing
the dependence of flow rate of a fluid inside a pipe on pres-
sure difference between the two ends of the pipe, length,
roughness, and diameter of the pipe, and fluid viscosity, is
derived via intuitive qualitative prediction of the effects of
fluid and pipe properties and pressure difference between
the two ends of a pipe on the flow rate of a fluid through the
pipe. A specific (linear) form of the constitutive equation is
derived using the analogy between flow of a fluid (driven
by pressure difference) through a pipe and electric current
(driven by electric potential difference) through a resistor.
Students who take Process Modeling II are already familiar
with different flow regimes and correlations for pressure drop
in a pipe, as they take this course after the Fluid Dynamics
course. This intuitive approach is needed only in Process



TABLE 1
Sample Process Examples Covered in
Process Modeling I Lectures
Liquid level in a tank with an open top and two pipes at the bottom
Liquid level in a tank with a fixed closed top, trapped gas above
liquid, and two pipes at the bottom
Liquid level in a tank with a moving closed top attached to an out-
side spring, trapped gas above liquid, and two pipes at the bottom
Heating tank
Mixing tank
Stirred tank chemical reactor
Continuous stirred tank fermentor
Tubular chemical reactor
Flow of a river over a salty bed
One-pass, shell and tube heat exchanger
Growth of fish in a lake
Prey and predator example
Female and male populations of a species
Mixing tanks with imperfect mixing
Tank reactors with imperfect mixing
Intravenous (IV) injection (drip) of a drug over a time period (zero
order infusion)
IV bolus injection/shot of a drug
Drug shot under skin (subcutaneous injection of a drug)
Oral administration of a drug












Example 1. Consider the tank shown in Figure El. Develop a model to describe level of
water in the tank, h (m), as a function of upstream and downstream pressures (P1 and P2).
FI and F2 are volumetric flow rates of water (m3/s) through the two pipes connected to the


H Water
F 1 P ,


E "> L- < I P2


A
Figure El: Tank process of Example 1.
bottom of the tank. For a moment, let us assume that water flows into the tank through
the left pipe and leaves the tank through the right pipe. P (kg/m/s2) is the pressure at the
bottom of the tank. P, (kg/m/s2) is the gas pressure at the top of the tank.
Assumptions
Constant tank cross-sectional area (A); No evaporation of water; Air does not dissolve in
water; Constant water density (pa); Constant temperature (T); Constant tank volume (AH)
Process Model Development
I. Conservation Equations
-Mass balance on water in the tank


Fip - F2p + 0 - 0 = [m]
dt
Mass balance on air in the tank


(kg of waters)


dt
where m is the mass of water in the tank, m. is the mass of air in the tank, and t is
time (s). The second equation implies that m. is constant.


II. Constitutive Equations
m = pAh (tank has a constant cross-sectional area); F1 fi(P1 - P); F2
P = pagh + P,; Ideal gas law (assumption): A(H - h)Pg = m0RT/M,


R f2(P P2);


III. Constraints
h > 0, t > to (reference time), P > 0, P > 0, P2 > 0 (absolute pressures)

IV. Conditions
The model has one 1st-order ordinary differential equation; one condition is needed. For
example, h(to) ho.
Process Model


dh ( m RT
dt MoA(H - h) Ph
h 0, t >to, P 0, P> 0, P2 > 0


MOA(H ) gh - P2 , h(to) = 0ho


Given the pressures Pi(t) and P2(t), the model can predict h(t).

Figure 1. Sample Process Modeling Ilecture notes (size reduced)-Part I.


Modeling I, which is offered before our chemical engineering
undergraduate courses except for Material Balance.
The interactive hands-on learning approach implemented in
Process Modeling I is described using the two sample examples
from Process Modeling I lecture notes given in Figures 1 and
2. Roughly a one-hour lecture is devoted to solving one to
two of such process modeling problems. First, the instructor
describes the process modeling problem using a schematic of
the process. Second, to help students understand the problem,
the instructor asks students to predict qualitatively and sketch
the responses of the process variables-which the model
should predict-to changes in other variables of the process.
68


For example, the instructor asks them
to predict whether water level in the
tank of Example 1 or 2 (Figure 1 or 2)
rises or drops when the downstream
pressure increases. Third, the instruc-
tor asks students to develop a model
of the process (carry out steps I-IV
listed at the beginning of this section).
Students can discuss their ideas with
each other and the instructor. The
instructor "answers" questions and
interacts with students through ask-
ing questions; direct answer is not
provided. Fourth, one or two students
voluntarily come to classroom board
and write their solutions) on the
board. A reward (1-3 points) is given
to the students who come to the board.
The points are counted towards 5% of
the final grade. If no student is willing
to come to the board, the instructor
solves the trivial parts of the process
modeling problem and then increases
the level of the reward (number of
points) and allocates more time to the
problem, until one student volunteers
to solve the challenging parts) of the
problem. Fifth, the instructor asks the
sitting students to comment on the
correctness and completeness of the
solutions) written on the board. The
standing students) should answer the
questions without any instructor's in-
terference. If the solutions written on
the board are incorrect or incomplete
and the seated students cannot detect
these deficiencies, then the instructor
asks questions that indirectly point
to the deficiencies in the solutionss.
Finally, the instructor provides a
summary of key issues in the process-
model development after a complete


and correct process model is developed.
Leading students through derivation of model equations for
about 50 carefully selected process examples in the lectures
allows the students to improve their engineering judgment
and to learn how to systematically integrate information on a
process to form a process model. Students' grades in Process
Modeling I are based on scores in three exams, each with a
weight of 28%, and on homework scores. The students are
provided with typed lecture notes; notes for a week of lectures
are made available to students in the following week (only
after the lectures are given). Several textbooks[2 4] are recom-
mended as references for background information.
Chemical Engineering Education


~t~L~


^^












TABLE 2
Sample Examples Covered in
Process Modeling II Lectures
Transient temperature profile in a semi-infinite
solid
Transient temperature profile in a cylindrical
solid
Transient temperature profile in a spherical
solid
Liquid level in two tanks (with significantly
different cross sectional areas) in series
Arnold cell
Batch reactor with two series reactions of
significantly different rates
Aging of human vs. that of a fly
Profile of temperature inside a fin
Transient concentration profile inside a solid
sphere


TABLE 3
Process Modeling II Sample Projects
Position vs. time behavior of bouncing ping-
pong and golf balls
Free fall of Teflon and metal spheres in a sugar
solution
Water drainage from a tank
A bottle of perfume
A layer of oil on a large area of flat land
Evaporation of water in a graduate cylinder
Liquid chemical spill on a sea

3. PROCESS MODELING II
Process Modeling II complements Process
Modeling I. It uses the same systematic
method to develop mathematical models of


more complex processes. Topics covered in
the lectures include: dh
* Use of dimensionless and normalized dt f\
variables in models d2H
* Functionalform of solutions to model
equations and design of experiments Given the press
* Steady states and steady state multiplicity
* Asymptotic stability (AS) and AS analysis Figure 2. Sa
of a steady state
* Multi-time-scale processes, quasi-steady state (QSS) as-
sumption, QSS model, andfast model
* Spatially distributed dynamic processes
* Method of combination of variables
As in Process Modeling I, many carefully chosen process
examples from different disciplines are used to teach the topics
listed above. Table 2 presents a list of sample process examples
covered in Process Modeling II lectures.
Vol. 44, No. 1, Winter 2010


Example 2. Consider the tank shown in Figure E2. Develop a model to describe level of
water in the tank, h (m), as a function of upstream and downstream pressures (Pi and P2).
FI and F2 are volumetric flow rates of water (n3/s) through the two pipes connected to the


Spring Force = kH

Gas Pg
Pi Water
F, P h F2
Pl P2
A
Figure E2: Tank process of Example 2.
bottom of the tank. For a moment, let us assume that water flows into the tank through
the left pipe and leaves the tank through the right pipe. P (kg/m/s2) is the pressure at the
bottom of the tank. P, (kg/m/s2) is the gas pressure at the top of the tank. mt (kg) is the
mass of the tank top.

Assumptions
Constant tank cross-sectional area (A), No evaporation of water, Air does not dissolve in
water, Constant water density (pw), Constant temperature (T), Friction-free tank top

Process Model Development
I. Conservation Equations
-Mass balance on water in the tank
Fip, F2p, + 0 - 0= [m] (kg of waters)
-Mass balance on air in the tank d
0-0+0-0 - [m,] (kg of air/s)
where m is the mass of water in the .I , ma is the mass of air in the tank, and t is
time (s). The second equation implies that m, is constant.
-H is not constant in this example and denotes the position of the top. Momentum
balance along H for the top:
d [ dH1
pgA - mg - kH = (kgm/2)
II. Constitutive Equations
m I (tank has a constant cross-sectional area); F = fl(P - P); F2 f2(P P2)
P = pagh+ P,; Ideal gas law (assumption): A(H - h)Pg '., I
III. Constraints
h, H > 0, t > to (reference time), P, P2, P2 > 0 (absolute pressures)
IV. Conditions
The model has one 1st-order and one 2nd-order ordinary differential equations; three
conditions are needed. For example, h(to) = ho, H(to) = Ho, dH, 0.
Process Model


maRT maRT
S , pgh - f2 h w gh P2 , h(to) he

h A - Ttg - kH, H(to) Ho, dH , 0, h, U H, P, P1,P2>0, t

ures Pi(t) and P2(t), the model can predict h(t).


mple Process Modeling I lecture notes (size reduced)-Part II.

Each week a process modeling project is assigned to
students. The titles of several sample projects assigned
in Process Modeling II are listed in Table 3. In most of
these projects, students are provided with data (measure-
ments) or are taken to a lab to conduct experiments and
collect (make) data (measurements). They use the data
(measurements) to perform model-parameter estimation
and evaluate the quality of their model predictions. The
process modeling project reports should include all of the












TABLE 4
Building Blocks of a Process
Modeling II Project Report
1. Describe the process modeling
problem accurately
2. Understand the problem
a. Collect information
b. Do bounding calculations
c. Draw a picture of the entire process
d. Identify the process variables
3. Derive process model equations
a. Draw schematic of systems and
subsystems
b. Write down notations
c. Choose assumptions
d. Write down relevant axiomatic laws
e. Write down relevant constitutive
equations
f. Assemble and simplify model
equations
g. Write down constraints, and
boundary and initial conditions
4. Solve the model equations
a. Select a solution method (analyti-
cal or numerical)
b. Implement the solution method
c. Evaluate the adequacy/reliability
of the numerical solution
5. Perform parameter estimation to
calculate the model parameters
6. Present model predictions (plots
and/or tables)
7. Evaluate the accuracy of the model
predictions
8. Discuss the results


TABLE 5
Student self-assessment of Process Modeling I course-objectives collected over 14
quarters, Fall Quarter 00-01 to Spring Quarter 06-07 (5=expert, 4=good, 3=fair,
2=poor, l=no experience); Number of students responded = 160; Number of students
not responded=251.

Average Standard
Deviation
Identifying fundamental phenomena Entering this course 2.46 0.92
governing a given process Leaving this course 4.05 0.59

Writing relevant balance equations Entering this course 3.13 0.84
Leaving this course 4.23 0.60
Developing mathematical models for Entering this course 2.23 0.86
chemical processes Leaving this course 4.10 0.65

Solving mathematical model equations Entering this course 2.67 0.91
governing a process Leaving this course 3.96 0.70



TABLE 6
Student self-assessment of Process Modeling II course-objectives collected over 12
quarters, Winter Quarter 00-01 to Summer Quarter 05-06 (5=expert, 4=good, 3=fair,
2=poor, l=no experience); Number of students responded = 155; Number of students
not responded=221.

Average Standard
Deviation
Developing a mathematical model for a Entering this course 3.09 0.69
process that you had not seen before Leaving this course 4.04 0.61
Leaving this course 4.04 0.61

Evaluating the accuracy/adequacy of a Entering this course 2.84 0.81
process model that you developed Leaving this course 3.98 0.73

Writing a technical report on a process Entering this course 3.39 0.84
modeling project that you carried out Leaving this course 4.10 0.69
Leaving this course 4.10 0.69

Identifying and presenting efficiently Entering this course 3.03 0.80
the main results of a process modeling Leaving this course 4.08 0.67
project


components listed in Table 4. Presentation quality of the
report is also very important. The report should show clearly
and concisely a) the work performed to develop the model, b)
the prediction capabilities of the model, and c) an evaluation
of accuracy/reliability of the model predictions. The final
grade is based on the project scores (weight of 45%), two
exam scores (weight of 25% each), and the 5% performance
(points collected in the class). As in Process Modeling I, the
students are provided with typed weekly lecture notes with
a one-week delay, as no appropriate textbook is currently
available for this course, either. Several textbooks-l13] are
recommended as references.


4. COURSE ASSESSMENT AND STUDENTS'
COMMENTS

At the end of the week in which each course ended, an e-mail

70


was sent to each enrolled student to ask the student to com-
plete an online evaluation form. The survey was anonymous.
Each of the online evaluation forms included questions on
the course objectives. For Process Modeling I, the course
objectives were:
* Identifying fundamentalphenomena governing a given
process
* i;1 i,, relevant balance equations
* Developing mathematical models for chemical processes
* Solving mathematical model equations governing a process
and for Process Modeling II:
* Developing a mathematical modelfor a process that you
had not seen before
* F.i, in,. ,,, the accuracy/adequacy of process model
that you developed
* Ti; in,, a technical report on a process modeling project
that you carried out

Chemical Engineering Education












TABLE 7
Sample comments made about Process Modeling I and II by students after graduation
"How do you blend science, experiment, casual observation, and common sense to predict the behavior of a system? How do you use both the
art and science of engineering to solve a problem? Process Modeling I and II approached these questions and in doing so served as an integral
piece of the chemical engineering curriculum. In particular, Process Modeling benefited students by: 1) offering a general systematic problem-
solving methodology; 2) depicting purpose for mathematics in engineering; and 3) being taught through flexible, hands-on approach."
"The only courses taken prior to Process Modeling I were Material and Energy balances. While both courses were obviously invaluable to a
chemical engineer, they focused on simple approaches to a narrow scope of problems. Process Modeling I offered a systematic, general problem
solving approach to modeling any system.
"Such an approach allowed young chemical engineering students to model systems involving momentum, mass, and heat transfer without
formally beginning the transport sequence. Reaction kinetics and reactor design are explored years before students take the formal course. An
inductive approach to looking at constitutive equations is used- through which students gain a qualitative rather than scientific grasp on the
concepts."
"Process Modeling I was a very good introduction to modeling techniques as well as methods to solving algebraic and differential equations.
It incorporated aspects and examples found in other chemical engineering courses such as Material and Energy Balances and the transport
courses. The course approached problems in a very systematic way, making it easier to define, outline, and set up problems."
i . .,, .n students are generally required to take multivariable calculus and differential equation coursesprior to beginning their engineer-
ing studies. Eventually students start questioning the usefulness of mathematics in their studies. Process modeling shows students how math-
ematics is nothing more than a body of knowledge that engineers tap into to help perform their work.
"In Process Modeling I and II, few problems are analytically solved. Rather, students build quick dynamic models (which are solved numerically
in homework assignments). This approach keeps the focus of the course on modeling rather than solving."
"The professor's teaching style complements the nature of the course. Rather than typical professor-to-students teaching, an interactive hands-
on learning approach is encouraged. A significant portion of the class is devoted to solving in-class problems (without the professor's immedi-
ate guidance). Most chemical engineering courses would immediately become more valuable by adopting such an approach. Simple concepts
become needlessly difficult to students because the basics were overlooked."
"Process Modeling II expanded on the previous course and went further in depth. It, too, took a systematic approach to solving problems. It
highlighted the importance of dimensionless variables and developed methods for solving partial-differential equations. A significant amount of
time was spent on the analysis of steady states and stability."
"Process Modeling I and I provided a solidframeworkfor modeling any process as well as preparing the student for higher-level chemical
engineering and math courses. Both courses helped me in analyzing problems by knowing how to determine the fundamental issues, simplify and
lump systems, evaluate relative rates of change, and apply known laws and concepts to come to a conclusion."
"Process Modeling I was the class I liked the most during my undergraduate study at Drexel. It helped me to defeat my fear when I encounter a
problem that I haven't seen before. It taught me all the steps on how to attack a problem that looks complicated. In this class, I learned that noth-
ing is too intricate in science; all the formulas and equations come from basic simple principles.
"Process Modeling II taught me the importance and the power of different math software such as Mathematica, Maple, and many others. It
taught me how to look at things in the big picture."
"The modeling courses at Drexel provided a core foundation that allowed subsequent courses to build upon. Being able to successfully model
a process is integral to the understanding and solving of many systems in the chemical engineering discipline. Compared to other chemical
engineering courses at Drexel, Ifound Process Modeling I and II to be two of thefew courses that tie together many of the other coursework
including reaction kinetics and all of the transport classes.
"While studying with other students for Ph.D. qualifying exams it became apparent that other universities do not stress the importance of model-
ing processes, but instead focus more on a quantitative approach. Without fully understanding the purpose of modeling it is difficult to attack
unfamiliar problems. Ifeel that having taken these classes has absolutely raised my confidence level and understanding of problems I haven't
seen prior."


* Identifying and efficiently ;' .....,, the main results of a
process modeling project
The assessment results collected over seven and six years
for Process Modeling I and Process Modeling II, respectively,
are given in Tables 5 and 6. The second column from the left
in each of the two tables lists the average of the "entering" and
"leaving" scores given by students for each course objective.
The difference between entering and leaving average scores
for each course objective is a measure of the impact of the
course on the students in terms of the specific course objec-
tive. Overall, the assessment results confirm the effectiveness
of the courses in transferring knowledge from the instructors
to the students.

Vol. 44, No. 1, Winter 2010


Sample comments made by students who already took
Process Modeling I and II and graduated, are listed in Table 7.
The students wrote and sent these comments to the instructor
in response to a request by the instructor, as part of the end-of-
course instructional survey. Before including these comments
in this paper, the authors edited a few of the comments to take
care of a few spelling/grammatical errors therein. In addition
to the course-assessment data collected right after the courses
ended, comments received from students and graduates of
our chemical engineering program over the past decade have
confirmed that the course contents, the method of delivery,
and the "in-course" feedback given to students through evalu-
ation of the students' performance in assigned homework

71











and exams all successfully contributed to the realization of
the three course objectives. Notable among the comments
are that the courses instill systematic problem-solving skills
in the students and show clearly and logically the purpose
of mathematics in engineering, using a flexible, hands-on
teaching approach. The courses blend science, experiment,
casual observation, and common sense to enable the students
to predict the behavior of a system.

5. CONCLUDING REMARKS
A successful application of inductive learning in process
modeling was presented. Two process modeling courses that
use inquiry learning and problem-based learning, among
other types, were described. The courses have been very
popular among the students and graduates of our chemical
engineering program. Students returning from Drexel Co-op
have expressed their satisfaction with the techniques they
learned in these courses, as they used the techniques or saw
the application of the techniques directly in their co-op proj-
ects. The graduates who took industrial positions requiring
them to develop process models or went to graduate school
described how confident they felt when they encountered and
solved complex process modeling problems. Both groups have
pointed to three significant impacts of the courses: first, the
confidence that the courses built in them to attack and solve
new process modeling problems; second, the systematic and
universal nature of modeling techniques covered and learned
in the courses; and third, their ability to remember easily and
apply quickly the techniques they learned.

ACKNOWLEDGMENT
This work was supported in part by the Department of
Chemical and Biological Engineering at Drexel University
and the National Science Foundation (NSF) through the grant
CTS-9703278. Any opinions, findings, conclusions, or recom-


mendations expressed in this material are those of the authors
and do not necessarily reflect the views of NSF

REFERENCES
1. Foss, A.S., and G. Stephanopolous, "Leading Undergraduates Along
Structured Paths to the Building of Good Process Models," Proceed-
ings ofASEE, Session 3613 (1999)
2. High, K., and E. Maase, "Active Problem-Solving in a Graduate Course
on Modeling and Numerical Methods," Proceedings ofASEE, Paper
AC 2007-1423 (2007)
3. Nocito-Gobel, J., M. Collura, and S. Daniel, "A Multidisciplinary
Modeling Course as a Foundation for Study of an Engineering Dis-
cipline," Emerging Trends in Engineering Education, Session 1693,
Paper 2006-2372 (2006)
4. Pang, K.H., "Teaching Chemical Engineering with Physical Plant
Model at Cal Poly, Pomona," Proceedings of ASEE, Session 2213
(2001)
5 Rappin, N., M. Guzdial, M. Realff, and P Ludovice, "Experiments
in Learning Chemical Engineering Modeling Skills," Proceedings of
ASEE, Session 2213 (2001)
6. Rabb, R., and D. Chang, "Interdisciplinary Teaching Techniques and
Learning in Dynamic Modeling and Control," Proceedings ofASEE,
Session 501 (2008)
7. Layton, R.A., "Using Modeling and Simulation Projects to Meet
Learning Objectives in an Upper-Level Course in Systems Dynamics,"
Proceedings ofASEE, Session 2320 (2003)
8. Prince, M.J., and R.M. Felder, "Inductive Teaching and Learning Meth-
ods: Definitions, Comparisons, and Research Bases," J. of Engineering
Education, 95(2), 123 (2006)
9. Prince, M., "Does Active Learning Work?A Review of the Research,"
J. of Engineering Education, 93(3), 223 (2004)
10. Felder, R.M., and R.W Rousseau, Elementary Principles of Chemical
Processes, 3rd Ed., Wiley and Sons (2005)
11. Welty, J.R., C.E. Wicks, G.L. Rorrer, and R.E. Wilson, Fundamentals
of Momentum, Heat, and Mass Transfer, 5th Ed., Wiley and Sons
(2008)
12. Gerald, C.E, and PO. Wheatley, Applied NumericalAnalysis, 7th Ed.,
Addison-Wesley Publishing Co (2003)
13. Perry, R.H., D.W. Green, and J.O. Maloney, Perry's Chemical
Engineer's Handbook, 7th Ed., McGraw-Hill (1997); knovel.com> 1


Chemical Engineering Education











curriculum
-0


INTRODUCING DAE MODELS


IN UNDERGRADUATE AND GRADUATE

Chemical Engineering Curriculum




RAVI KUMAR [ANDELA
Clarkson University * Potsdam, NY
L.N. SRIDHAR
University of Puerto Rico at Mayaguez * Mayaguez, PR
RAGHUNATHAN RENGASWAMY
Texas Tech University * Lubbock, TX


Models play an important role in understanding
chemical engineering systems. While Ordinary
Differential Equation (ODE) models for physical
systems are well covered in the undergraduate curriculum, not
much attention is paid to DAE models of chemical engineering
systems. DAE models arise naturally in several chemical/
physical systems that admit quasi steady-state approxima-
tions. Common examples of these can be found in separation
and reaction systems. While these models can be thought of
as a natural extension to ODE models, there are several new
concepts that arise when one considers DAE systems. These
are the index of the system, consistent initialization, and so
on. Some of these concepts might seem abstract for chemical
engineering undergraduates. There is a need to teach these
ideas at an undergraduate level, however, as these types of
models are becoming commonplace.
In this article, we propose a simple reactive flash as an
excellent vehicle for introducing DAE systems to chemical
engineering undergraduates. The power of this example is that
it allows DAE systems to be introduced naturally in several
courses. While one obvious place to introduce this example is
after a discussion of ODE systems in a mathematical methods
course, this example can be equally easily incorporated into a
thermodynamics course. A concomitant solution approach that
is a straightforward extension of the Rachford-Rice procedure
can be taught for solving flash problems. The reactive flash
increases the complexity of the flash calculation by one level
through the inclusion of reactions to phase equilibrium. The rest


of the article is organized as follows: In the section 1, a simple
reactive flash model is discussed with all the assumptions in-
volved. The steady state solution of the reactive flash system
through a modified Rachford-Rice procedure is then proposed.
In section 2, we show how the more involved ideas, such as the
index of a DAE and consistent initialization, can be discussed
in the context of this chemical engineering problem.

Raghunathan Rengaswamy is a professor in
the Chemical Engineering Department at Texas
Tech University. He received his B.Tech from
IITMadras and a Ph.D. from Purdue University,
all in chemical engineering. He spent seven
years as a facultymember at Clarkson, NY, and
five years as a faculty member at IIT Bombay,
India, prior to joining Texas Tech. His research
interests are in the
areas of Fuel Cells
and Process Sys-
tems Engineering.
Dr. Sridhar is a professor at the University of
Puerto Rico at Mayaguez. He received his
Ph.D. from Clarkson University.
Ravi Kumar Man-
dela is a Ph.D.
student in the
Chemical and
Biomolecular En-
gineering Department at Clarkson University,
Potsdam, NY. He received his B.Tech form
Andhra University, India, and M.S. from the
University of Nevada, Reno, both in chemical
engineering. His research interests are in the
areas of Process Control and Nonlinear State
Estimation.


� Copyright ChE Division of ASEE 2010


Vol. 44, No. 1, Winter 2010










1. THE REACTIVE FLASH PROBLEM
Reactive Distillation (RD) is an intensification process
involving the combined operation of reaction and separa-
tion in a single processing unit. For a detailed review of the
various chemical systems that can benefit from the concept
of RD, refer to Sundmacher and Kienle.m1 In many of these
cases, combining reaction and separation can lead to increased
conversion, high selectivity, and reduced capital investment.
RD is useful when the chemical system involves reversible
reactions, azeotropes, and undesired product formations.
The idea of RD can be explained to undergraduates through
a simple reactive flash, which is an extension of the equilib-
rium flash problem. The reactive flash can be thought of as an
idealization of a single stage in an RD column. In this process,
the two streams exiting from the reactive flash are assumed to
be in phase equilibrium and the ratio of the reaction rate to the
mass transfer rate is given by a dimensionless number, called
the Damk6hler number. Additional assumptions of isothermal
and isobaric behavior further simplify the problem formula-
tion. A model for the reactive flash is discussed next.
1.1 Reactive Flash Model
The single-stage, isothermal, isobaric, reactive flash being
considered is shown in Figure 1. The following assumptions,
which simplify the model equations, are made while formulat-
ing the process model[21:
1. Liquid and vapor leaving the stage are assumed to
be in equilibrium
2. Liquid hold up is assumed to be constant
3. Isobaric and isothermal process


Figure 1. Schematic of a reactive flash.


4. Non-ideality of liquid phases
5. Reaction takes place in the liquid phase
The equations used to describe this dynamical system are


dx
H -t =Fz -Lx
dt
y = Klx

where K = p
P
x1=I
EYx =1
Ty, =1


Vy, + Hvix


L and V are liquid and vapor flow rates; x and y are liquid
and vapor mole fractions respectively; F is the feed flow rate;
H is the liquid hold up; the Damk6hler number,

Da- H/F
Da =
1/kre

is a dimensionless ratio of the characteristic liquid resi-
dence time to the characteristic reaction time; v1 is the
stoichiometric coefficient; E is the extent of the reaction;
,1 is the activity coefficient; K is the equilibrium constant;
and the index i runs from 1 to n, where n is the number of
components in the system. kref is the forward rate constant
evaluated at a reference temperature. Introducing the di-
mensionless variables,
L V
' F' F

and the dimensionless time, = t/(H/F), the dynamic state
equation becomes


dx
= z - Zx - Ox y1
dTr

y = Klx (2b)
x1 = 1(2c)
TY = 1 (2d)


Da
--vi (2a)
kf


A simple degrees-of-freedom analysis will show that there
are n differential equations and n+2 algebraic equations mak-
ing up a total of 2n+2 equations. The number of variables are
the n vapor mole fractions, n liquid mole fractions, and the 2
normalized flows, resulting in a total of 2n+2 variables.
1.2 Modified Rachford-Rice for the Steady State
Solution of a Reactive Flash Problem
The steady state version of the reactive flash problem can
be introduced as a natural extension to the flash problem.
The steady state solution to the reactive flash problem can be
derived as a modified Rachford-Rice procedureP as discussed
below, in cases where the equilibrium constants are either
constant or functions of liquid mole fractions.
Chemical Engineering Education


Zi

T, P, Da










Defined as
R1 = Hvl/F (3)


= ZR1 (4)

As we solve the steady state by setting the differential term
in equation (2) to zero, we get

(z - 01x - 0yl) R = 0 (5)

Since


(y, -X)= 0

and

z1 + R1
S 1 K1Ov

through substitution, we arrive at
z +R (K -I
(01 +K Ov)


Summation of Eq. (5) over all the components gives
01 =1+ -

Substituting for 01 from Eq. (9) into Eq. (8) gives
zl+R ( K -l)= (0 )=0
[1+ ' 1 - 0K 0

The derivative of this function with respect to 90 is


- z L (KR -1)2) 'O,) (11)


Using this derivative, the Newton-Raphson method can be
used to find the roots of this nonlinear equation.
Example 1: Steady State Simulation of MTBE Reactive Fash
Consider the example of a reactive flash where Methyl
Tertiary Butyl Ether (MTBE) is formed from isobutene and
methanol in the presence of an inert compound n-butane.
Isobutene + MFEOH - MTBE


(6) There are four components in the system. Consider an
isothermal, isobaric, reactive flash. Let us assume that the
Wilson equation is used for calculating the liquid phase ac-
tivity coefficients. The Wilson binary interaction parameters
and the Antoine coefficients for this system can be found in
(7) Chen, et al.[2 A solution to this isothermal, isobaric, reactive
flash problem can be found using the modified Rachford-Rice
procedure outlined in the last section. A pseudo-code can be
developed as given below.
(8) * Enter the input conditions
* Start with an initial guess for liquid compositions
* Outer loop checks for the convergence of liquid
mole fractions
(9) * Give the initial guess for 0
* Inner loop checks for the convergence of 0
* Update liquid mole fractions using Eq. (7) with R
(10) and K at previous liquid mole fractions and 0
* Terminate when the outer loop satisfies the conver-
gence criteria of liquid compositions
If we were to use the following process conditions: feed
composition (z) = [0.1569 (isobutene), 0.1555 (methanol), 0.1
(MTBE), 0.5876 (n-butane)], temperature = 370.4729 K, pres-
sure = 1114300 Pa, feed flow
Rate = 100 mol/h, holdup =
30 mol, then the solution for
0 is 0.4873.


0.9 0.95 1 1.05 1.1 1.15 1.2 1.25
Pressure (Pa) x106
Figure 2. Effect of pressure on the amount of vapor leaving the system.


This means that 48.73%
of the feed will leave the
unit as vapor and the re-
mainder will leave as liquid.
This example can be used to
further analyze the effect of
pressure and temperature on
the amount of vapor leaving
the system. Figure 2 depicts
the effect of pressure on the
vapor leaving the system,
and it can be seen that
with increasing pressure
the amount of vapor at the
outlet decreases.


Vol. 44, No. 1, Winter 2010










A similar analysis can be performed to study the effect
of temperature on the vapor flow rate. Figure 3 depicts this
relationship. In contrast to the pressure, with increasing tem-
perature, the vapor flow rate actually increases.

2. DAE ANALYSIS OF THE REACTIVE FLASH
PROBLEM
DAE systems involve both differential and algebraic con-
straints. The most general formulation for a DAE system is
S= f(x,z,X) (12)

0 = g(x,z,X)

Some of the examples of DAE systems can be found in pro-
cess control, chemical reaction engineering, separation process
modeling, network modeling, and constrained body dynamics.
DAE systems are characterized by the index of the system. The
index of a DAE system is defined as the number of differentia-
tions required to eliminate the algebraic terms, i.e., to convert
the DAE into an ordinary differential equation (ODE) system.
A simple example of an index 2 DAE system is

2 = +- X (t (13)
0 = 2 + (t)

Differentiating the algebraic equation once, we get

2 = -2(t) (14)

Differentiating the algebraic equation twice [differentiation
of Eq. (14)] yields


2 = Y ,+ (t)


2it


Putting these equations together we get an ODE as shown
in Eq. (16).

S,=y,+ X (t) (16)

I = -x (t) - (t)


Figure 3. Effect of
temperature on the
amount of vapor
leaving the system.


Since the algebraic equation had to be differentiated twice
to get to the ODE form, the above system is an index 2
system. There are differences in solving DAE systems for
steady-state and dynamic behavior. The steady-state solu-
tion is simply achieved by solving all the equations with
the differential part on the left-hand side (LHS) set to zero.
The dynamic solution for the DAE system is more involved,
however. One approach is to convert the DAE into an ODE
through multiple differentiations as discussed above and to use
a regular integration approach. This is not always advisable
because it can be shown that the solution error in the algebraic
constraint is proportional to the extent of index reduction.
The error increases linearly if the index is reduced by one,
quadratically if the index is reduced by two, and so on. It has
to be ensured that the solution satisfies the original algebraic
constraints at all times. The initial values have to satisfy the
original constraint equation and also the new constraints
generated through differentiation. This is called consistent
initialization of the DAE system. A more detailed discussion
on consistent initialization can be found in References 4-6.
Some of the numerical integration and iterative methods for
solving DAE systems are found in References 7 and 8.
These concepts can be taught effectively through the reac-
tive flash example. Notice that the reactive flash example is a
DAE system with n differential and (n+2) algebraic variables.
This system can be simplified into just n differential variables
and 2 algebraic variables system by direct substitution of
equation (2b) into (2a) and writing equation (2d) in terms of
the x variables. We chose to work with the original system
to bring out the various aspects of the DAE systems in this
article. Note also that this reduction is not always possible. In
the general form of the DAE model given in Eq. (12), it might
not be possible to write the algebraic equations as explicit
functions of the differential variables.
The first exercise for the students would be to calculate the
index of the reactive flash model, which is 2. This is because


Chemical Engineering Education


0.9

0.8 -

S0.7

-2 0.6-
0.
a-
S0.5-

2N 0.4-

o 0.3-
z
0.2-

0.66 368 370 372 374 376 378 380
Temperature (K)










the mole fraction summation equations have to be differentiated twice to get the differential (61 and 0) terms. For solving the
dynamic simulation problem, there are several possible approaches. The DAE system can be converted into an index 1 problem
by differentiating the summation of liquid mole fractions and vapor mole fractions equations. The resulting DAE is of the form
given in Eq. (17).
dx Da
=z -1x -0 y1 --v (17a)
dT- 1 kref
y = Klx (17b)
dx Da
d= 1- 0 - v + Da e = 0 (17c) (17)
dT kref
dy dx dK
d =l K x' = 0 (17d)
dTr dr dT
This is an index 1 DAE because one differentiation of Eq. (17b), (17c), and (17d) will lead to a regular ODE in all the (2n+2)
variables, which is the original dimension of the system. This index 1 DAE can now be directly solved using the MATLAB
odel5s routine. One has to ensure that consistent initialization of the variables is performed, however. Choosing the initial values
of the (2n+2) variables such that equations (17b) - (17d) are satisfied will not yield a correct solution to the original problem.
This is because when Eq. (2c) and (2d) are differentiated some information is lost. Eq. (17c) and (17d) will ensure that each
summation of mole fractions equals a constant but not necessarily one. For consistent initialization, Eq. (2c) and (2d) should be
added to Eq. (17b)-(17d), and this constrains the initial values. A generic criterion for when differentiating a subset of nonlinear
DAE equations will further constrain the initial values can be found in Reference 4. If the reactive flash example is used in a
postgraduate mathematical methods course, then the students can be asked to explore this aspect further. This model can also
be used to look for singularities such as Hopf, branch, and limit points as discussed in Reference 9 and is another avenue for
exploration in a postgraduate mathematical methods course.
For the reactive flash problem, the number of constraints to be satisfied by the initial values is equal to (n+4) and the number
of initial values to be chosen equals (2n+2). The degrees of freedom equal (n-2). The reactive flash problem can be converted
into ODE with the same number of variables (2n+2) or a reduced number of variables. In the following, the derivation of the
reactive flash model with only differential equations in x is discussed. Eq. (17b) is eliminated from the set of equations by incor-
porating this equation directly into Eq. (17a). 01 and 0 appear explicitly in Eq. (17c). Performing algebraic manipulations on
Eq. (17d) will also yield an equation that has 01 and 0, explicitly in it. Now these two equations can be solved simultaneously
to explicitly write 01 and 0, in terms of the x variables. The final expressions for 01 and 09 are

9= TZx, T vex K Z+ vex -+KKx+TKxx 1-2


- Txx+1- l x 1 rTK XX (18)



= TZx >T vex K Z vex - Txx 1 1-
S ITjZxDa Vx-IK Z Da Ix i j K2D
1 I kref J 1 1 kref 1 - IJ - kr1 f



K K X +Kxx - Txx 1 (19)


with the following definition
dK
T =- 1 (20)
lj Ox

The 01 and 0 expressions are substituted into Eq. (17a). Now Eq. (17a) with substitutions for the variables y, 0, and 0 is an
ODE system [shown in Eq. (21)] that represents the dynamics of the reactive flash problem. The calculation of analytical deriva-
tives is performed in the MAPLE environment. The equations generated in MAPLE are converted into a MATLAB function.
The initial values are still constrained for the index 1 system that we discussed because the latest manipulations were algebraic


Vol. 44, No. 1, Winter 2010














The idea of

Reactive

Distillation can

be explained to

undergradu-

ates through a

simple reactive

flash, which is

an extension of

the equilibrium

flash problem.

The reactive

flash can be

thought of as an

idealization of a

single stage in

an RD column.


with the same equations. Hence, for consistent initialization there are (n-2) degrees of freedom.
In other words, for the ODE in the variables x, only (n-2) components can be specified and the
remaining two mole fractions have to be solved to satisfy the constraints. This reflects the fact
that the ODE resulted from a DAE and hence the initial states are constrained, unlike a pure
ODE. Once consistent initialization is performed, the solution to the reactive flash problem can
be obtained using any standard ODE solver, such as the ODE solver in MATLAB. The mole
fractions that are computed will automatically solve the summation constraints.


dx
.. .=z -Z 0(X...x x
dT


Da
kref


Example 2: Dynamic Simulation of TAME Reactive Flash
Consider a reactive flash where Tertiary Amyl Methyl Ether (TAME) is synthesized by an
acid catalyzed equilibrium reaction of isoamylenes and methanol. The reaction considered is
2M1B + 2M2B + 2MEOH - 2TAME
The reaction kinetics for TAME synthesis have been studied by various authors. Hwang, et al.,E101
used a concentration-based expression for combined etherification reactions from two isoam-
ylenes. Later, rigorous kinetic studies were undertaken by Oost, et al.,"11 Christian, et al.,[121
and Sundmacher, et al.'13 These authors reported activity-based kinetic models for lumped as
well as separate etherification reactions. In this example, the activity coefficients are calculated
using a Wilson model. The Wilson binary interaction parameters and the Antoine coefficients
were taken from Chen, et al.[14] Light gasoline fraction (C5-cut) from the fluidized catalytic unit
is the source of isoamylenes. There are three isomers of amylene: 2 Methyl-1-Butene (2M1B)
and 2-Methyl-2-Butene (2M2B), which are reactive, and 2-Methyl-3-Butene (2M3B), which
is nonreactive. The reactive iso-amylenes are diluted with n-pentane as an inert solvent. The
rate model used is taken from Chen, et al.[141
There are five components in the system. Consider an isothermal, isobaric, reactive flash
with the following conditions. In the data given below, the component feed compositions are
provided in the order MeOH, 2M1B, 2M2B, TAME, and n-pentane.
Feed composition (z) = [0.2647,0.0463,0.2846,0.0000,0.4044]
Temperature= 335 K
Pressure=2.55 bar
Da =0.462


S 5 10 15 20 . 0 5 10 15 20
Dimensionless time (Tau) Dimensionless time (Tau)
Figure 4. Plots of first component and third component vs. dimensionless time (T).


Chemical Engineering Education


S(xI...x )K (x ...x X)x











For the dynamic simulation using the ODE, we need to
choose consistent initial estimates. Since this is a five-compo-
nent problem, the degrees of freedom equals 3 (5-2). Hence,
we choose the three mole fractions in the liquid phase ( x (0))
as [0.1828, 0.0303, 0.2944] and we get the other two mole
fractions as [0.1131, 0.3794] for consistent initialization.
These mole fraction values lead to y,(0) = [0.2561, 0.0359,
0.2789, 0.0174, 0.4117]. Notice that the mole fractions sum
to one. The values for 01 and 0 are 0.2720 and 0.6849,
respectively. It is important to note that consistent initializa-
tion leads to the mole fractions summing to one, but does not
automatically take care of the constraint that the mole fractions
have to be non-negative. With this initialization, a dynamic
simulation can be performed. The plots of the first component
and the third component compositions against dimensionless
time (T) are shown in Figure 4. The numerical simulation
results are shown in Table 1. In the first five columns of
Table 1, the mole fraction values of the liquid at the outlet
for the first 10 dimensionless time values are reported; the
second-to-last column reports the mole fraction summation
for the liquid, and the last column reports the mole fraction


TABLE 1
The mole fractions of five components in liquid phase and the
summation of mole fractions in liquid phase and vapor phase
for the first 10 instances.
Dimensionless xi x2 x3 X4 X5 Xx Xy
Time(z)
0.2 0.1821 0.0290 0.2930 0.1132 0.3827 1.0000 1.000
0.4 0.1815 0.0280 0.2920 0.1132 0.3853 1.0000 1.000
0.6 0.1809 0.0273 0.2913 0.1132 0.3872 1.0000 1.000
0.8 0.1804 0.0269 0.2908 0.1132 0.3887 1.0000 1.000
1.0 0.1798 0.0266 0.2905 0.1132 0.3899 1.0000 1.000
1.2 0.1793 0.0263 0.2902 0.1133 0.3909 1.0001 1.000
1.4 0.1789 0.0262 0.2901 0.1133 0.3917 1.0001 1.000
1.6 0.1785 0.0261 0.2900 0.1133 0.3923 1.0001 1.000
1.8 0.1781 0.0260 0.2899 0.1133 0.3929 1.0001 1.000
2.0 0.1777 0.0259 0.2899 0.1132 0.3933 1.0001 1.000

0.181 0.293

0.18- 0.2925

S0.179 0.292
0 0
E 0.178- E 0.2915
8 8
p 0.177 e 0.291

o 0.176 - o 0.2905

S0.175 - 0.29

" 0.174- " 0.2895

0.173-- 0.289

0.172 0.2880
0172 m 10 20 30 00.28 1 0
Dimensionless time (Tau) Dimens


summation for the vapor. Notice that solving a consistently
initialized ODE incorporates the mole fraction summation
qualities at all values of the dimensionless time. The same
simulation is run with a different consistent initial estimate:
x(0) = [0.1796, 0.0298, 0.2892, 0.1138, 0.3876]. The plots
of the first and the third component compositions against
dimensionless time (T) using this initial value are shown in
Figure 5. It is clearly evident from Figures 4 and 5 that the
same steady state is reached, even though the initial values
for these simulations are different.
The dynamic solution starting with anon-steady state condi-
tion can be summarized as follows


Vol. 44, No. 1, Winter 2010


* Identify the index of the DAE system
* Reduce the given DAE system to index 1 or ODE
form
* Find the degrees of freedom, which is the differ-
ence between the total number of variables and the
total number of algebraic equations
* Perform consistent initialization by specify-
ing the number of initial values equal to the
degrees of freedom and the other values that
- are obtained from the solution to the alge-
braic constraints
* Once initial values are specified, solve the
ODE or index 1 DAE using an appropriate
solver

o 3. CONCLUSIONS
i0
0 In this paper, the introduction of DAE systems in
0 the undergraduate curriculum is proposed through
0 an interesting chemical engineering example, the
30
0 equilibrium reactive flash. This is a reasonably
0 simple example that is easy to understand both
0 physically and mathematically. It is shown that this
- simple example can be used to teach the concepts
of index and consistent
initialization that are rel-
evant in solutions to DAE
systems. This example can
be introduced in an under-
graduate thermodynamics
or mathematical methods
class. Further, other options
for deeper investigation
into DAE systems, starting
with the same example,


Figure 5. Plots of first
component and third
component vs. dimen-
i 20 30 (
;ionless time (Tau) sionless time (r).












are also pointed out if this material were to be included in a
postgraduate mathematical methods course.


AVAILABILITY OF MATERIAL FOR USE IN
CLASSROOM

The MATLAB codes used in this article can be obtained by
contacting Dr. Rengaswamy at raghu.rengasamy@ttu.edu.


REFERENCES
1. Sundamacher, K., and A. Kienle, Reactive Distillation: Status and
Future Directions, John Wiley and Sons, NJ (2003)
2. Chen, E, R.S. Huss, M.E Doherty, and M.E Malone, "Simulation of
Kinetic Effects in Reactive Distillation," Computers and Chem. Eng.,
24, 2457 (2000)
3. Ruiz, G., L.N. Sridhar, and R. Rengaswamy "The Isothermal Isobaric
Reactive Flash Problem," Industrial and Engineering ( ,....... , Re-
search, 45, 6548 (2006)
4. Pantelides, C.C., "The Consistent Initialization of Differential-Alge-
braic Systems," Siam J. Sci. Stat. Comput., 9(2), 213 (1998)
5. Vieira, R.C., and E.C. Biscaia, "Direct Methods for Consistent Initial-
ization of DAE Systems," Computers and Chem. Eng., 25(9), 1299
(2001)
6. Neumann, J., and C.C. Pantelides, "Consistency of General Point


Conditions for DAE systems," Computers and ( ..... i I 30(1), 125
(2005)
7. Pennestri, E., and L. Vita, "Strategies for Numerical Integration of
DAE Systems in Multibody Dynamics," Computer Applications in
Eng. Ed., 12(12), 106 (2004)
8. Miekkala. U., "Dynamic Iteration Methods Applied to Linear DAE
systems," J. Computational and Applied Methods, 25(2), 133 (1989)
9. Ruiz, G., M. Diaz, and L.N. Sridhar, "Singularities in Reactive
Separation Processes," in press, Industrial and Engineering( ......
Research (2009)
10. Hwang, W.S., and J.C. Wu "Kinetics and Thermodynamics of Synthesis
of Tertiary Amyl Ether Catalyzed by Ion Exchange Resin," J. Chin.
Chem. Soc. 41, 181-186 (1994)
11. Oost, C., and U. Hoffmann, "The Synthesis of Tertiary Amyl Ether
(TAME): Microkinetics of the Reaction," Chem. Eng. Science, 51,
329-340 (1996)
12. Christian, J., K. Sundmacher, and U. Hoffman, "Residue Curve Maps
for heterogeneously Catalyzed Reactive Distillation and Fuel Ethers
MTBE and TAME,"( I,.... i Sci., 52, 993-1005 (1997)
13. Sundmacher, K., G. Uhde, and U. Hoffmann, "Multiple Reactions in
Catalytic Distillation Processes for the Production of Fuel Oxygenates
MTBE and TAME: Analysis by Rigorous Model and Experimental
Validation," Chem. Eng. Science, 54, 2839, 2847 (1999)
14. Chen, E, R.S. Huss, M.E Doherty and M.E Malone, !,II.1I -" ..I
States in Reactive Distillation: Kinetic Effects," Computers and Chem.
Eng., 26, 81 (2002) 1


Chemical Engineering Education











In1 laboratory


MICROFLUIDICS IN THE


UNDERGRADUATE LABORATORY:

Device Fabrication and an Experiment

to Mimic Intravascular Gas Embolism




ERIN L. JABLONSKI, BRANDON M. VOGEL, DANIEL P. CAVANAGH
Bucknell University * Lewisburg, PA, 17837
KATHRYN L. BEERS
National Institute of Standards and Technology * Gaithersburg, MD, 20879


Microfluidic technology is rapidly finding utility in
a wide range of applications from chemical syn-
thesis and separations to genetics and biochemical
assays.[12] The flow of nanoliter quantities of fluids through
micrometer-scale channels offers a unique environment for
improved flow control, heat transfer, and fluid mixing.[3] The
prohibitive expense of manufacturing microfluidic devices,
however, has created a significant barrier to the use of micro-
fluidics for academic instruction. Traditional manufacturing
methods involve aggressive chemical etching or sensitive
photolithographic techniques that require clean room facili-
ties. Some fabrication techniques lessen the need for these
clean room facilities, but still require fabricating an expensive
negative master to reproduce channels in an elastomeric
material.[4] Having a safe, inexpensive, and versatile method
to prepare masters for microfluidic devices in a typical lab
environment[5,6] creates an opportunity to develop labora-
tory experiments that reinforce basic ideas in subjects from
chemistry and biology to chemical engineering and materials
science. Besides bubble-flow analysis, microfluidic devices
can be used to study myriad fluid dynamics phenomena in
different channel geometries.
The experiment detailed here involves the break-up of air
bubbles in an aqueous flow. Droplet and bubble break-up have
been studied extensively in the literature experimentally and
numerous models have been proposed that describe droplet
break-up in microchannels. An excellent review of droplet
dynamics is provided by Cristini and Tan.'7 Jousse, et al.,


provide an analysis of both experimental and theoretical work
involving the dynamics of the flow of droplets.[8] Specifically,
they characterize the flow of droplets from a single parent
channel into two daughter channels. Leshansky and Pismen
detail the behavior of droplets incident on a T-junction and

Erin L. Jablonski is an assistant professor in the Department of Chemical
Engineering at Bucknell University. Her interests in engineering education
include effective integration of classroom and laboratory activities through
project-based design. Her research interests are in the characterization of
surface chemical properties and diffusion in polymeric materials.
Brandon M. Vogel is an assistant professor of chemical engineering and
is a Jane W. Griffith Fellow at Bucknell University. He received his B.S. in
chemistry and B.ChE. from the University of Minnesota, Twin Cities, and
his M.S. and Ph.D. in chemical engineering from Iowa State University.
He teaches biomaterials, bioprocess engineering, senior design, and
statistics. His research interests include the synthesis of new materials to
detect, target, and treat disease.
Daniel P. Cavanagh is chair and associate professor in the Department
of Biomedical Engineering at Bucknell University. He currently holds the
Richard C. and Gertrude B. Emmitt Memorial Chair in Biomedical Engi-
neering and was awarded the Christian R. and Mary F. Lindback Award
for Distinguished Teaching at Bucknell University. His research interests
focus on the use of microfluidic technology to investigate the mechanics
of intravascular gas embolism.
Kathryn L. Beers is the deputy division chief and Sustainable Polymers
Group leader in the Polymers Division of the National Institute of Standards
and Technology. Her current interests include the use of microfluidics and
microreactors for polymerizations based on renewable feedstocks and
using green transformation routes. In the past, her research focused on
measurements of polymers at interfaces, and polymerizations in confined
spaces, as well as developing microfluidic technology for complex fluid
measurements.


� Copyright ChE Division of ASEE 2010


Vol. 44, No. 1, Winter 2010










describe the contribution of the capillary instability to droplet
break-up in their system.[9] Droplet traffic at a T-junction has
been well characterized by Engl, et al., who present a method
to determine how the "feedback" from droplets beyond a
junction point influences the behavior of subsequent droplets
upstream.[10] Several researchers have investigated droplet
break-up at junctions of arbitrary angle, both experimentally
and theoretically, for a broad range of flow conditions.1112]

OVERVIEW OF THE EXPERIMENT
This experiment introduces students to the design and
fabrication of a microfluidic device. In fabricating the de-
vice, students use processing steps common to lithography;
a photomask and ultraviolet (UV) exposure source are used
to pattern a negative-tone photoresist and contact lithogra-
phy is then performed using a mold. The strengths of the
device-fabrication method are ease and low cost of produc-
ing masters that are traditionally difficult and prohibitively
expensive to make.
The flow experiment presented allows students to inves-
tigate the behavior of bubbles in liquid flows as a model of
intravascular gas embolism, or the presence of gas bubbles
in the bloodstream. The microfluidic environment facilitates
the investigation of bubble dynamics as a function of bubble
size and bulk liquid flow rate at biologically relevant length
scales. Through using microbubbles to model the behavior
of gas emboli in the small model vasculature, students gain
an appreciation of the dynamics of fluid behavior at smaller
scales. Without prior preparation by a laboratory instructor,
designing and fabricating the microfluidic device requires
three successive laboratory sessions. The laboratory instruc-
tor, however, can reduce this to one meeting with sufficient
preparation-particularly of the release and frame layers. To
expedite the process, the instructor can prepare master slides
that students can use to pattern polydimethylsiloxane (PDMS),
and have patterned PDMS ready to be sealed against glass, so
that students will carry out all aspects of the fabrication but
not have to wait for PDMS to degas or cure before moving on
to the next step. For the experiment modeling gas bubbles in
the bloodstream, one laboratory session should be enough to
observe bubble dynamics with varying flow conditions.

MATERIALS AND EQUIPMENT'
. Device Fabrication
After mask design, the microfluidic device is created by
preparing a negative of the channel in a thiolene-based resin
using masks such as those in Figure 1. The negative is then
reproduced in PDMS and sealed against a glass slide. Small-
gauge Teflon (or Stainless steel) tubing is inserted into the
PDMS at the circular channel ends for inlets and outlets.
Equipment necessary for the fabrication procedure includes
a radio frequency (RF) plasma etcher (with oxygen or air)
(REFLEX Analytical Corporation) and an ultraviolet (UV)


Figure 1. Typical photomasks used for microfluidic
device masters.
flood exposure source (365 nm, UVA) (Spectronics).
The materials needed for device fabrication include glass
slides (2 in. X 3 in.), a polystyrene container (15 cm X
20 cm X 2 cm) (or large glass Petri dish 20 cm diameter),
thiolene optical adhesive (NOA-81 resin, Norland Products
Inc., Cranbury, NJ), and PDMS prepolymer (Sylgard 184,
Dow-Coming).
The microfluidic mask can be designed using a software pack-
age such as Microsoft Publisher, Canvas, or Adobe Illustrator.
The PDMS release (for fabricating the master mold) and
frame (supports the mold during device fabrication) layers
were prepared in advance. Aluminum foil was used to protect
the benchtop and to prevent spreading the PDMS solution to
surrounding areas in the laboratory. From the Sylgard 184 kit,
the silicone elastomer base and curing agent were mixed in
a 10:1 ratio in a disposable beaker. This PDMS solution was
vigorously mixed with a disposable utensil for approximately
3-5 min or until the solution was opaque with bubbles. For
the release layer, the PDMS solution was poured into a con-
tainer approximately 15 cm X 20 cm X 2 cm (or into a glass
Petri dish with 20 cm diameter). The release layer stays in its
container and can be used multiple times; it is convenient to
have several release layers for quickly fabricating multiple
devices. For the frame, the PDMS solution was poured into
a 10 cm X 15 cm X 0.5 cm polystyrene container. The frame
is removed from its container, so this container is disposable.
The PDMS solution was de-gassed in a vacuum desiccator by
pressurizing and depressurizing the unit until all the bubbles
were removed. This process takes about one hour, depend-
ing on the quantity of PDMS solution. When the PDMS
solution was devoid of bubbles, the containers were placed
into an oven for at least 3 hr at 70 �C. The oven shelf must
be level to ensure even curing of the release layer and frame.

t Equipment and instruments or materials are identified in the paper in
order to adequately specify the experimental details. Such identifica-
tion does not imply endorsement by the authors, nor does it imply the
materials are necessarily the best available for the purpose.


Chemical Engineering Education



























Figure 2. PDMS frame layer on top of a PDMS
release layer.


The frame slab was carefully removed from the polystyrene
container and a razor blade was used to cut a rectangular
hole in the center of the slab. The inner dimensions of the
frame are slightly smaller than the glass slide used to create
the microfluidic device. The frame was placed in the center
of the release layer, as shown in Figure 2.
Photomasks were created with the desired device designs
using Microsoft Publisher and printed on a transparency with
a 1200-dpi (or better) resolution laser printer (HP 8000). The
photomask may contain any pattern provided the dimensions
are suitable for the glass slide. The optical adhesive used to
make the master is a negative-tone photoresist. Therefore, the
mask background is black, while the lines that will become
the microfluidic channels are white, as shown in Figure 1. UV
light polymerizes optical adhesive in only the white part of the
mask. In the photomask, small circles were added to the end of
the lines where the entry and exit points are located to facilitate
connecting the channels to the syringe pumps (optional). To
ensure complete opacity of the photomask, two copies of the
photomask were printed on a transparency and then taped
with the channels of both copies carefully aligned.
To make the master (the mold used to imprint a PDMS slab
to make the microfluidic device), a commercially available
optical adhesive, NOA-81, was used. NOA-81 is stored in the
refrigerator and must be allowed to come to room tempera-
ture before use. The glass slide that became the substrate for
the master was cleaned with absolute ethanol and dried with
compressed air, then placed in the RF plasma cleaner (oxygen
or air plasma) for 5 min at full power. To fabricate the master,
the NOA-81 solution was poured into the frame on the re-
lease layer until the frame was filled and nearly overflowing.
Any air bubbles were allowed to dissipate or were carefully
popped with a razor blade. The plasma-cleaned glass slide was
allowed to cool in the plasma cleaner for a few minutes, and
then carefully laid over the NOA-81 to avoid trapping any air
bubbles underneath the slide. The photomask was positioned


Figure 3. Mask added to glass slide on top of the PDMS
frame containing optical adhesive.

on top of the glass slide, as shown in Figure 3. This assembly
was placed under a flood UV source roughly 15 cm from the
light source and cured for 3 min to 5 min. Getting the appropri-
ate feature height requires some trial and error to optimize the
exposure time and distance from the UV lamp. The resulting
feature height was measured with a micrometer.

After the UV exposure, the photomask was lifted off and
the glass slide (master) was carefully removed from the frame.
The master has raised crosslinked areas where the photomask
is transparent and uncrosslinked adhesive on other portions.
Using compressed air (house line is suitable), as much of the
uncured adhesive as possible was blown away. The master
was washed with absolute ethanol to remove more of the
uncured adhesive. A razor blade was used to remove any ex-
cess unreacted material from the edges. The master was then
carefully washed with a small volume (10 mL) of acetone
and then immediately washed again with copious amounts of
absolute ethanol. Acetone can cause delamination if left on
the master for too long. It is important to remove the excess
adhesive to create a device with features that replicate the
photomask pattern. After removing the excess adhesive, the
master was subjected to a second UV exposure (post-cure)
(with lamp 15 cm away) for approximately 20 min. The
master was then baked at 50 C in an oven overnight to ad-
here the NOA-81 to the glass slide (Figure 4, next page). A
master that is properly cleaned and fully cured will last for
many PDMS reproductions.

The master was used to mold a slab of PDMS by placing the
master with the raised mold facing up into an appropriately sized
"boat" made from aluminum foil. The "boat" was then filled with
PDMS solution and the PDMS was cured in the oven.

The patterned PDMS slab still on the master was removed
gently from the foil and a razor blade was used to trim the
excess PDMS that exceeded the dimensions of the glass slide.
The patterned PDMS slab was then carefully peeled away


Vol. 44, No. 1, Winter 2010










from the master. It is important the patterned PDMS slab not
be cracked when being removed from the master. A needle (or
punch) was used to puncture holes through the PDMS where
the circular entry ports were patterned. A new glass slide that
had been cleaned with ethanol and dried with compressed air
and the patterned PDMS slab (pattern side up) were placed
into the oxygen plasma cleaner for five minutes on the lowest
setting. (Determination of the suitable time/power of plasma
exposure for proper device sealing may require trial and er-
ror.*) After plasma cleaning, the glass slide and patterned
PDMS slab were pressed together to form the microfluidic
device. To improve bonding between the PDMS slab and glass
slide, they were placed in an oven at 70 �C for about 10 min.
Hollow metal or flexible capillary Teflon tubing was inserted
to create ports of entry and exit. Figure 5 is an image of the
completed device.
II. Gas Emboli in Model Blood Flow Experiment
The major components of the experimental setup are: two
multisyringe infusion/withdrawal syringe pumps (one for
bulk fluid, one for air infusion) (Cole-Parmer), binocular low-


Figure 4. (above) Completed master of optical adhesive
on glass.
Figure 5. (below) Completed microfluidic devices
showing inlet and outlet ports using hollow metal
tubing ports and large flexible tubing connectors.


power compound microscope with high-speed digital camera
linked to computer with image processing software (Cole-Par-
mer), a high-speed imaging system (Roper Scientific, Motion
Corder SR-500), and the microfluidic device.
In preparation for running the gas emboli experiment, a 60
mL syringe was loaded with approximately 50 mL of water.
Next, a 20 mL syringe was filled with air. A 60 cm piece of
luer-lock tubing was used to connect the 60 mL syringe to one
port on the three-way luer-lock stopcock. Using a second 60
cm piece of luer-lock tubing, the 20 mL syringe was connected
to an open port on the stopcock. A third piece of luer-lock
tubing was used to connect the outflow port of the stopcock
to the inlet channel on the microfluidic device. The stopcock
was set to have all three pathways open.
With the microfluidic device securely placed on the stage
of the microscope, the device was aligned on the stage for
optimal viewing of both the parent tube and the bifurcation
point. An image acquisition rate of 500 frames/s (fps) was
used.t The bulk liquid syringe pump was set at a flow rate
less than or equal to 3 mL/min. The flow was maintained long
enough to flush any residual air from the liquid flow tubes
and the microfluidic device. Once the liquid path was clear of
bubbles, the air syringe pump was started at a flow rate less
than that of the liquid pump. In general, if the bubbles in the
system are very long or are passing the bifurcation at a very
high frequency, the flow rate may be decreased to produce
smaller and fewer bubbles. Several air bubbles were allowed
to traverse the microfluidic device to verify the correct op-
eration of the system. On verification of correct operation,
the air syringe pump was turned off and the liquid pumping
was used to flush any remaining air from the system. The
liquid pump was then turned off in preparation for starting
the experiment.
For the experimental microfluidic device described here,

* If a radio frequency plasma cleaner is unavailable, an alternative
and less expensive method for bonding PDMS to glass is detailed in
Reference 13.
f The necessary camera speed fr for capturing bubble bifurcation
is dependent on the velocity through the channels. Depending on the
camera available, the flow rates should be adjusted accordingly.


Chemical Engineering Education










the parent channel dimensions are 280 gm high and approxi-
mately 500 gm wide, splitting into two daughter channels
each having a width of nearly 250 gm. The cross-section of
the channels has a semicircular shape. The dimensions of the
photomask control the channel widths but the UV exposure
time fixes the channel height (longer exposure times give taller
features but compromise lateral resolution). These channel
sizes approximate the size of an artery or vein and are easy
to obtain with a photomask printed on a laserjet printer with
a minimum resolution of 1200 dots per square inch (dpi).

EXPERIMENTAL METHODS AND RESULTS
The bulk fluid syringe pump infuses the bulk liquid (water or
blood mimic) through 1.59 mm ID luer tubing to a three-way
stopcock. A second syringe pump infuses air into another port
on the stopcock. Bubble-laden flow exits from the stopcock
by 1.59 mm ID tubing that is attached to the port inserted in
the microfluidic device at the entrance to the parent channel.
The behavior of the bubbles as they traverse the bifurcation
is recorded with the digital imaging system. After flowing
through the device, the fluid with bubbles leaves through exit
ports into a waste container. The bulk liquid syringe pump
flow rate is varied between 2.0 and 3.0 mL/min as this range is
representative of flows found in similarly sized blood vessels
(Table 1). The bulk flow rate of 3 mL/min was found to be
the highest flow rate allowable to still record quality images
with the digital imaging system at its maximum rate of 500
fps. While it is possible to obtain images at a rate as low as
100 fps, it is difficult to observe bubble-splitting behavior
at this rate. The experimental bubbles were infused into the
bulk flow by the second syringe pump that was set at a flow
rate lower than the bulk flow syringe pump. Because of the
dynamic nature of bubble formation upon infusion into the
bulk flow, creating successive bubbles of identical size was
difficult. Therefore, the typical approach used in these experi-
ments was to record 8-10 seconds of images that were later
examined using image processing software* to measure and
group bubbles by size.
To perform the required image analysis, images recorded by
the camera mounted on the microscope were downloaded to
a computer. Besides permitting the transfer of images to the

TABLE 1
Properties of Blood Flow in Vessels. Reynolds number
values calculated for blood flows in blood vessels. The
standard uncertainty of the measured values is � 5%.
Vessel Inner Diameter Velocity Re
(mm) (cm/s)
Arteries 0.15 to 15 10 to 40 500
Arterioles 0.01 to 0.14 0.1 tolO 0.7
Capillaries 0.008 < 0.1 0.002
Venules 0.01 to 0.14 < 0.3 0.01
Veins 0.15 tol5 0.3 to 5 150

Vol. 44, No. 1, Winter 2010


computer, this imaging setup also enables the user to view
the same images through the optical lens of the microscope
during the experiments. To test the repeatability of the experi-
ments, each combination of bubble size and bulk liquid flow
rate was carried out four times.
With the system primed and flushed of air as instructed
above, the image acquisition system was activated to record
continuously and store 8 seconds of images. The bulk liquid
syringe pump was started and liquid pumping continued until
there were no bubbles in the flow path. With a clear flow path
and readied image acquisition system, the air syringe pump
was started to infuse air into the flow path. In general, it may
be necessary to adjust the flow rate of air to produce a steady
stream of bubbles with the desired size. With a stream of
bubbles flowing through the microfluidic device, the image
acquisition system was started to record 8-10 seconds of im-
ages at 500 fps.
The image files were examined to find sequences of im-
ages that contain bubbles traversing the bifurcation. The
axial length of the bubble in the parent tube was measured
to classify the bubbles-size categories. The behavior of the
bubble was then examined as it traversed the bifurcation to
further classify bubbles based upon behavior at the branching
point. Bubbles are grouped by size and behavior to examine
the correlation between the two. To examine the effects of
bulk liquid flow rate, this experimental process can easily be
repeated for different bulk liquid flow rates. After running a
series of experiments examining various bubble sizes and flow
rates, the user can create a table that indicates the observed
behavior of bubbles of specific sizes at specific flow rates.
This table can be used to determine the effect of bulk liquid
flow rate and bubble size on bubble behaviors at bifurcations
in channels of known size.
Upon completing the investigation, many possibilities exist
for extending the experiments for further investigations. These
additional possibilities may also be used by an instructor
to provide for experimental variations between laboratory
groups. Possible modifications include varying the bulk fluid
including water, oil, or xantham gum solutions as blood ana-
logs. An independent experiment to determine the viscosity
of the bulk fluid should be carried out. Different microfluidic
device designs can also be used. Finally, two immiscible
liquids could be used to examine liquid droplet behavior in
surrounding liquid streams. In general, this experiment pro-
vides for a wide range of experimental possibilities.

DISCUSSION
The relevant parameters that best characterize the behavior
of liquid flows in pipes or channels are average velocity and
Reynolds number (Re). The average velocity and Re of actual

* The authors recommend ImageJ (Reference 15). Matlab can also be
used for image analysis.











blood flow in blood vessels are dependent on the diameter of
the vessel as shown in Table 1 where representative values
are listed.[141 These values are calculated using the definition
of Re for flow in circular pipes, Re = pVD/g, where D is pipe
diameter, p is the density of the fluid [g/cm3], V is the aver-
age fluid velocity [cm/s], and gis the fluid viscosity [g/cm-s].
Because one of the goals of this experiment is for students to
characterize the fluid flow by calculating Re, the noncircular
cross-sectional profile of the channels must be considered.
This is accomplished by replacing the pipe diameter, D, in the
equation above with the hydraulic diameter (DH) resulting in
the following equation, Re = pVDH/ g, with DH = 4A/Pw where
A and Pw are the cross-sectional area and wetted perimeter of
the channel, respectively. Therefore, Re of the fluid flows in
the parent channels in this experiment can be calculated given
Pwt = 1.0 g/mL and waie = 0.01 cP. For flow rates ranging
from 2 mL/min to 3 mL/min, the Reynolds number is 30 <
Re < 50. Assuming the definition of laminar flow in a circular
pipe (Re < 2000) can be applied to this semi-circular channel
system, the flows in this experiment are laminar.
Bubbles introduced into the fluid flow move with the fluid
because of the pressure gradient produced within the channel
by the syringe pump. When a bubble meets a division (Figure
6a, 6b), the leading edge of the bubble stops at the apex and


the trailing edge continues to move. If the trailing liquid-gas
boundary has enough momentum to reach the junction, a
pinch point forms and the bubble splits (Figure 6c and 6d).
Conversely, if the trailing liquid-gas boundary does not have
enough momentum, the bubble does not break but instead
flows to one side of the division.
The observed behaviors of bubbles traversing the bifurca-
tion as a function of bubble size and flow rate are presented in


TABLE 2
Behavior of bubbles traversing bifurcations as a func-
tion of Reynolds number (Re) and bubble size (0). A =
no bubble division; B = bubble division; L and R indi-
cate into which channel bubbles traveled; B* = bubble
division with daughter bubble lodged in daughter
channel.
Reynolds number (Re) in the parent channel
Re/X 33 41 50
1.5-1.99 A (2L:2R) A(4L) A (4L)
2.0-2.49 A(3L;1R) A (2L:2R) A (4L)
2.5-2.99 B* B* B
3.0-3.49 B B B
3.5+ B B B


Figure 6. Images of bubble (the dark shape) flow behavior at a bifurcation in a microfluidic device, a) The leading edge
of the bubble hits the bifurcation, b) The bubble splits as the trailing edge flows toward the bifurcation. c) The trailing
edge of the bubble hits the bifurcation. d) The bubble splits into two separate bubbles as they continue to flow down the
daughter channels of the device.


Chemical Engineering Education


a














C


"Stj


A/-.. .. ...











Table 2. In general, two distinct behaviors are found. The first,
indicated as type A, is when the entire bubble travels down
either the left (L) or right (R) daughter tube without dividing.
With each type A behavior in Table 2, we also indicate into
which daughter channel (L or R) each of the bubbles traveled
in the four experimental trials. The second type, indicated
as type B, occurs when the bubble divides at the bifurca-
tion producing a daughter bubble in each daughter channel.
Also, for two combinations of Re and X, we observed a few
instances where after dividing, one of the daughter bubbles
became lodged in a daughter channel. These behaviors are
indicated as B*.
Overall, bubbles are more likely to break if they are rela-
tively large for the characteristic length of the channel and
are in a liquid flow with a higher Re. Larger bubbles have a
"sausage" shape in the channel; they form this shape to lower
the interfacial surface area between the gas and the fluid. The
small size of the channel inhibits the bubbles from forming
a perfect sphere (the lowest free-energy shape). Because the
"sausage" shape is not the preferred shape for the bubble, it
is unstable and will break at the division to lower its energy
interfaciall surface area) if given enough momentum. Alter-
natively, a bubble may remain intact if there is not enough
momentum to destabilize it at the division or if one daughter
channel offers lower resistance to flow (that is, is larger than
the other daughter channel). The bubbles that are smaller than
the width of the channel are able to preserve their lowest en-
ergy shape (the sphere) and therefore they do not need to break
at the apex to lower their energy. Students can experimentally
discover many of these phenomena using devices with several
different channel configurations. It is important to point out
that observations from this simple microbubble experiment
may give insight to potential paths to reduce emboli.

CONCLUSION
We present a protocol for fabricating microfluidic devices
and results from an example experiment modeling intravas-
cular gas emboli that occur in the bloodstream. Microfluidic
devices are well suited for this investigation because they
contain channels with cross-sectional dimensions in the ~500
pm range depending on the fabrication procedure. The advan-
tages to these devices are their low cost, small sample size,
and quick analysis. The designs of the channels are limitless
and allow for a wide range of experiments including mixing


chemical solutions, filtering of solutions, liquid chromatogra-
phy and microchemical reactors. With this fabrication proto-
col, undergraduate students will be able to perform laboratory
exercises involving microfluidic technology.

ACKNOWLEDGMENTS
The authors thank Kristin McDonnell, Tracey Perry, Lisa
Schultz, Lauren Shafer, and Amy Ukena (Bucknell Univer-
sity) for their work on the experimental protocol.

REFERENCES
1. Auroux, PA., D. Iossifidis, D.R. Reyes, and A. Manz, "Micro Total
Analysis Systems 2. Analytical Standard Operations andApplications,"
Analytical ( h ....-.. , 74(12), 2637 (2002)
2. Vilkner, T., D. Janasek, A. Manz, "Micro Total Analysis Systems.
Recent Developments," Analytical ( -....... i, 76(12), 3373 (2004)
3. Reyes, D.R., D. Iossifidis, PA. Auroux, and A. Manz, "Micro Total
Analysis Systems 1. Introduction, Theory, and Technology," Analytical
( ', .". ", 74(12), 2623 (2002)
4. Xia, Y.N., and G.M. Whitesides, "Soft Lithography," Angewandte
Chemie-International Edition, 37(5), 551 (1998)
5. Cabral, J.T., S.D. Hudson, C. Harrison, and J.E Douglas, "Frontal
Polymerization for Microfluidic Applications," Langmuir, 20(23),
10020(2004)
6. Harrison, C., J. Cabral, C.M. Stafford, A. Karim, and E.J. Amis, "A
Rapid Prototyping Technique for the Fabrication of Solvent-Resistant
Structures," J. Micromechanics and Microengineering, 14(1), 153
(2004)
7. Cristini, V., and Y.-C. Tan, "Theory and Numerical Simulation of
Droplet Dynamics in Complex Flows-a Review," Lab on a Chip,
4(4), 257 (2004)
8. Jousse, E, R. Farr, D.R. Link, M.J. Fuerstman, and P Garstecki,
"Bifurcation of Droplet Flows within Capillaries," Physical Review
E, 74(3), 036311-1 (2006)
9. Leshansky A.M., and L.M. Pismen, "Breakup of Drops in a Microflu-
idic T Junction," Physics ofFluids, 21(2), 023303-1 (2009)
10. Engl, W., M. Roche, A. Colin, and P Panizza, "Droplet Traffic at a
Simple Junction at Low Capillary Numbers," Physical Review Letters,
95(20), 208304-1 (2005)
11. Menetrier-Deremble, L., and P Tabeling, "Droplet Breakup in Mi-
crofluidic Junctions of Arbitrary Angles," Physical Review E, 74(3),
035303-1(2006)
12. Link, D.R., S.L. Anna, D.A. Weitz, and H.A. Stone, "Geometrically
Mediated Breakup of Drops in Microfluidic Devices," Physical Review
Letters, 92(5), 054503-1 (2004)
13. Haubert, K., T. Drier, D. Beebe, "PDMS Bonding by means of a Por-
table, Low-cost Corona System," Lab on a Chip, 6(12), 1548 (2006)
14. Muth, C.M., and E.S. Shank, "Gas Embolism,"New England Journal
ofMedicine, 342(7), 476 (2000)
15. Collins, T.J., "ImageJ for Microscopy," BioTechniques, 43(1), S25
(2007) 1


Vol. 44, No. 1, Winter 2010







UNITEDSTATES Statement of Ownership, Management, and Circulation
POSTAL SERVICE, (All Periodicals Publications Except Requester Publications)
1. Publication Ti e 2. Publication Number 3. Filing Date

ChjCr c re - l0 Ie-I Ic. o Sp. ' 4, 20a
4. Issue Frequency 5 Number J ss-s oi': a.nsir. 6 J-r.a Sbsc;, ',x Pr nLe


7. Complete Mat l c. ;,: o ,, N*orp.- I".iS i ,r i:'.I"p "g (S'c - I *-':0 P11 " F-- " s
Ck'e-fv .eW.-4 LA. L40to L Cketv. er,.. 1>f4. _-A __AsL_
L vt-r; .i opF Fb ri e , (Eaasv.ke A C40'a Co.ntZ L. Te.et I.-i nLfja. %l.7=.
_ _ ____ ___ _ 135'Z- 3 '2- r
8. Complete Marling Address of Headqugrters or General Business Offr.e l. Publiher (Not pIrteI
C.crL� . s -b'~ V S r car C e- Pr E�tr r Ev cd'kCc�.f--
k "l\ .Po,-- lrcd ca L.^ ^.^--,-b,-.. 2-00 e
c i ' " i'..-- : - , ,:- ;. etc ".! ." r " , -n - j . ' - -r ;"_b..phe , E,: .,r arJ Mr . 'aa i,',g E.- " ! 3 ,.2 ;!'. .e - :
Publisher (Name and complete ; ad -

LOaSk .,t ,--, ct. 2.oo 0k
Editor (Name and complete making address) . , r r
-r-try. - 4tr sv l _.em. r g. �+. , fU--v . o rc>r^<
,oL rsvd lto Fl- 32-lt - (Loo
Managing Editor (Name and complete mailing address)
L , n H--ade1 cheK . g. e+, a- L-tc,, 0 t or -r4
G.(LnsAlty *F.L 32Z.(ll- Oc'S
10. Owner (Do not leave bank. If the pubicafion is owned ty a corporation, give the name and address of the : " .; immediately followed by rhe
names and addresses of all stockholders owning or holding 1 percent or more of the total amount of stock, if not owned by a corporation, give the
names and addresses of tne inrdviduai owners. if owned oy a partnership o other unincorporated irm, give ts name ard address s we# as those of
coach individual owner. If the publication is published by a nonprofit organization, give its name cnd address.)
Full Name Complete Maiiing Address


(fRcpa 4 ir:& <(L\ a- \d e- r6ed 4o0 ,: __+






1. Known Bondholders, Mortgagees, and Other Securrty Holders Owning or
Huldinj 1 Percent or More of Total Amount of Bo ds. Mortgages, or
Other Securities. If none, check box i None
Full Name Complete Mailfng Address










The nurpose., unction, 3rd' n onrofit stati'-, f h:l s organization and the exempt status for federal! icorne i~ x pirpTse'.
iP Has Not Changed Durring Preceding 12 ,Monrhs
Sl Has Changed During Precedng 12 vMo1,i:hs Pb..i,:sher mus: submitn exolanaiion of change with :ts statnem
88 Chemical Engineering Education






Visit
us
on the
Web
at




University of Florida Home Page
© 2004 - 2011 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Powered by SobekCM