Chemical engineering education

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

Material Information

Title:
Chemical engineering education
Alternate Title:
CEE
Abbreviated Title:
Chem. eng. educ.
Physical Description:
v. : ill. ; 22-28 cm.
Language:
English
Creator:
American Society for Engineering Education -- Chemical Engineering Division
Publisher:
Chemical Engineering Division, American Society for Engineering Education
Publication Date:
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 applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 01151209
lccn - 70013732
issn - 0009-2479
sobekcm - AA00000383_00176
Classification:
lcc - TP165 .C18
ddc - 660/.2/071
System ID:
AA00000383:00176

Full Text













chemical engineering education
















"Finger Kits:" An Interactive Demonstration of Biomaterials and Engineering for Elementary School
c Students ip. 125'
Canavan, Stanton, Lopez, Grubin, Graham

-o Random Thoughts: How To Write Anything (p. 139)
Felder. Brent
C
w A Lab Experiment to Introduce Gas/Liquid Solubility (p. 147)
c Fonseca. Almeida. Fachada
C
w Geothermal Cogeneration: Iceland's Nesjavellir Power Plant (p. 132i
Rosen
4-'
a Incorporating Risk Assessment and Inherently Safer Design Practices into ChE Education Ip. 1411
o Seay, Eden
C
3 Mixing Hot and Cold Water Streams at a T-junction (p. 154)
5a Sharp. Zhang. Xu, Ryan.Wanke. Afacan
F w

0C C

C

C C


.c
C C

u U





I,,












Author Guidelines for the

LABORATORY

Feature

The laboratory experience in chemical engineering education has long been an integral part
of our curricula. CEE encourages the submission of manuscripts describing innovations in the
laboratory ranging from large-scale unitoperations experimentsto demonstrationsappropriate
for the classroom. The following guidelines are offered to assist authors in the preparation of
manuscripts that are informative to our readership. These are only suggestions, based on the
comments of previous reviewers; authors should use their own judgment in presenting their
experiences. A set of general guidelines and advice to the author can be found at ourWeb site:
.

c Manuscripts should describe the results of original and laboratory-tested ideas.
The ideas should be broadly applicable and described in sufficient detail to
allow and motivate others to adapt the ideas to their own curricula. It is noted
that the readership of CEE is largely faculty and instructors. Manuscripts must
contain an abstract and often include an Introduction, Laboratory Description,
Data Analysis, Summary of Experiences, Conclusions, and References.
An Introduction should establish the context of the laboratory experi-
ence (e.g., relation to curriculum, review of literature), state the learning
objectives, and describe the rationale and approach.
The Laboratory Description section should describe the experiment in
sufficient detail to allow the reader to judge the scope of effort required
to implement a similar experiment on his or her campus. Schematic dia-
grams or photos, cost information, and references to previous publica-
tions and Web sites, etc., are usually of benefit. Issues related to safety
should be addressed as well as any special operating procedures.
SIf appropriate, a Data Analysis section should be included that concisely
describes the method of data analysis. Recognizing that the audience
is primarily faculty, the description of the underlying theory should be
referenced or brief.The purpose of this section is to communicate to the
reader specific student-learning opportunities (e.g., treatment of reac-
tion-rate data in a temperature range that includes two mechanisms).
*The purpose of the Summary of Experiences section is to convey the
results of laboratory or classroom testing. The section can enumerate,
for example, best practices, pitfalls, student survey results, or anecdotal
material.
A concise statement of the Conclusions (as opposed to a summary) of
your experiences should be the last section of the paper prior to listing
References.













EDITORIAL AND BUSINESS ADDRESS:
( h, nr, ,al I. rnima t e,' 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
James O. Wilkes, U. Michigan

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


-PUBLICATIONS BOARD
CHAIRMAN
John P. O'Connell
University of Virginia

VICE CHAIRMAN *
C. Stewart Slater
Rowan University

MEMBERS
KristiAnseth
University of Colorado
Jennifer Curtis
University of Florida
Rob Davis
University of Colorado
Pablo Debenedetti
Princeton University
Dianne Dorland
Rowan
Thomas F. Edgar
University of Texas at Austin
Stephanie Farrell
Rowan University
Richard M. Felder
North Carolina State University
H. Scott Fogler
University of Michigan
Jim Henry
University of Tennessee, Chattanooga
Jason Keith
Michigan Technological University
Steve LeBlanc
University of Toledo
Ron Miller
Colorado School of Mines
Susan Montgomery
University of Michigan
Lorenzo Saliceti
University of Puerto Rico
Stan Sandler
University of Delaware
Donald R. Woods
McMaster University


Vol. 42, No. 3, Summer 2008


Chemical Engineering Education
Volume 42 Number 3 Summer 2008




> DEPARTMENT
118 Chemical Engineering at Tennessee Technological University
Joseph J. Biernacki, with faculty and staff

> RANDOM THOUGHTS
139 How To Write Anything
Richard M. Felder, Rebecca Brent

> OUTREACH
125 "Finger Kits:"An Interactive Demonstration of Biomaterials and
Engineering for Elementary School Students
Heather E. Canavan, Michael Stanton, Kaori Lopez,
Catherine Grubin, and Daniel J. Graham


> CURRICULUM
141 Incorporating Risk Assessment and Inherently Safer Design Practices
into Chemical Engineering Education
Jeffrey R. Seay and Mario R. Eden

> LABORATORY
154 Mixing Hot and Cold Water Streams at a T-junction
David Sharp, Mingqian Zhang, Zhenghe Xu, Jim Ryan, Sieghard Wanke,
andArtin Afacan

147 A Lab Experiment to Introduce Gas/Liquid Solubility
I.M.A. Fonseca, J.PB. Almeida, and H.C. Fachada

> CLASS AND HOME PROBLEMS
132 Geothermal Cogeneration: Iceland's Nesjavellir Power Plant
Edward M. Rosen


CHEMICAL ENGINEERING EDUCATION (ISSN 0009-2479) is published quarterly by the Chemical Engineering
Division,American SocietyforEngineeringEducation, and is edited at the University ofFlorida. Correspondence regarding
editorial matter, circulation, and changes ofaddress should be sent to CEE, Chemical Engineering Department, University
of Florida, Gainesville, FL 32611-6005. Copyright 0 2008 by the Chemical Engineering Division, American Society for
Engineering Education. The statements and opinions expressed in this periodical are those of the writers and not necessarily
those of the ChE Division, ASEE, which body assumes no responsibilityfor them. Defective copies replaced if notified within
120 days ofpublication. Writefor information on subscription costs andfor 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).

117










[ll =1 department


ChE at


Tennessee Technological University


JOSEPH J. BIERNACKI,
WITH FACULTY AND STAFF
Tennessee Technological University (Tennessee Tech)
began its life as Dixie College in 1909, with a few
small but elegant Georgian buildings on 7th Street
and Dixie Avenue in Cookeville, Tennessee. The tiny, then-
private cl lk.g. evolved into Tennessee's only technologi-
cal university with a strong engineering, science, business,
and education emphasis. By the end of the 1940s the seed
for what would eventually become chemical engineering
was planted within the Department of Chemistry, but the
sapling soon withered and became dormant for another two
and a half decades. In 1966, a young man named John C.
McGee, a Ph.D. from North Carolina State University, was
hired. Kindled by the presence of a growing chemical indus-
try in Tennessee, McGee and a colleague from the already
established Mechanical Engineering Department officially
became the Chemical Engineering Department and McGee
served as its first departmental chairperson. Not long after,
W.D. (Denny) Holland (Ph.D., Georgia Tech) was hired,
followed by David W. Yarbrough (Ph.D., Georgia Tech) and
Clayton P. Kerr (Ph.D., Louisiana State University). These
four men, young and energetic, would build the department
from grass fields and empty rooms, lay the foundation for a
strong unit-operations tradition, construct a lasting laboratory
infrastructure, cultivate the master's program, embrace the
college-level Ph.D. when introduced in 1971, pioneer compu-
tational techniques that were budding ideas at the time-and
become respected and dedicated educators, researchers, and
friends for the next three-plus decades, thus sowing the seeds
that would grow into the present-day Department of Chemi-
cal Engineering.
While Professors McGee, Holland, Yarbrough, and Kerr
would, from time to time, be joined by other faculty, they
alone would remain for a life's career at Tennessee Tech.
Copyright ChE Division ofASEE 2008


The clock tower at Derryberry Hall.


These historical notes are brief yet important. The present
faculty acknowledges and owes much of the department's
ongoing success to the foundation that Professors Emeriti
McGee, Holland, Yarbrough, and Kerr established. In 1999,
both McGee and Holland retired, followed by Yarbrough and
Kerr in 2001 and 2002, respectively. Since then, the depart-
ment's faculty has been re-created, ushering in a new wave
of excitement and productivity. Much as the department's
founders laid the cornerstones and built a tradition and legacy,
the present faculty has initiated a renaissance: introducing
new research thrusts, modernizing both the undergraduate
and graduate curricula in content and pedagogy, and stepping
forward in service to regional outreach and their respective
professional communities. The remainder of this article deals
with the present, and to some extent, a vision for the future.

Dixie College,formally Dixie University, wasfounded by the Church
of Christ in 1909.
Chemical Engineering Education



































Yearbook photos (ca. 1976) of (clockwise from upper left)
early faculty W. Denny Holland, John C. McGee,
Clayton P. Kerr, and David W. Yarbrough, superimposed
on a picture of Prescott Hall, home of the Department of
Chemical Engineering since 1971.

PLAYING LIKE A TEAM-INSIGHTS FROM
OUR VISITING SPEAKERS
Each academic term the department hosts what has become
a model research seminar series within the university, bring-
ing in eight to 10 regional and national speakers. If you have
been one of our speakers, you know that this will not be a
day of rest for you. A full day of interaction with the students
and faculty will be carefully planned and integrated with your
seminar and, in the end, you will likely know who we are and
we will know something about you. Consistently and almost
universally, at the end of the day our guests tell us that the
single most striking characteristic of our department is how
the faculty clearly demonstrates collegiality, teamwork, and
a sense of scholarly community.
This is a posture that we cultivate and strive to perfect.
Pedro Arce, the department's chairperson, is a well-known
educator, having developed several strategies for teaching
and learning including "The Coach Model.1 2]" It is no
coincidence that we view the department as a team. Arce
mentors the junior faculty, gives them opportunity to grow
as team members, listens to his players, is sensitive to the
environment of the game and, in the end, lets his team play.
Formally, one might call our management structure distrib-
uted and unoriginal.t Faculty empowerment is the factor that
changes the equation.
Our faculty members represent the department and carry
their own authority as well as the confidence of the chairper-


son and each other. Each faculty member is empowered to
"play his/her position" and to "pass or shoot" when he or she
sees the opportunity.
A team has many elements; not all are on the field. Our
team has two additional individuals without whom the game
could not be played: Rebecca (Becky) Asher, our depart-
mental office coordinator, and Perry Melton, our laboratory
and machine shop technician. While Asher is holding the
department together by responding to the many requests of
the faculty and day-to-day student needs, Melton is keeping
the unit operations running and working with students to build
equipment that they design, as well as helping the faculty and
graduate students with research labs.

EDUCATIONAL OBJECTIVES-INSIGHTS
FROM OUR CONSTITUENTS
As part of our recent Southern Association of Colleges
and Schools (SACS) accreditation, Tennessee Tech has a
newly established Quality Enhancement Program (QEP) and,
accordingly, a QEP Committee.* This committee surveyed
the constituency of the university-our students, alumni,
employers of our students, and the faculty-and found that
unilaterally, "what really counts" are critical thinking skills
and the ability to solve "real-world problems.[3]" When it
came time to review our departmental Program Educational
Objectives (PEOs) it was simple: We would integrate critical
thinking and real-world problem solving in some way, and we
would write PEOs that were timeless.
The result is our present statement of PEOs that are our driv-
ing force and motivation:
Within roughly five to seven years our graduate population
will collectively exhibit the following traits:
be critical thinkers
be real-world problem solvers
have continued theirformal education
be working at the frontiers in chemical engineering
In addition to real-world problem solving and critical think-
ing, we have chosen to explicitly call for the continuation of
formal education and working at the frontiers in chemical
engineering. These two objectives complete the characteristics
of our program: an environment that empowers students to take
responsibility for their own learning (lifelong learning) wherein
research (the frontier) is highly integrated with, and pushes the
boundaries of, undergraduate education-making it compatible
and forming a continuum with graduate studies.


f Technically, the department uses a transformational-based manage-
rial structure with a strong team-based component.

Several ChE faculty members have received seed grants from the
University-wide effort to integrate critical thinking and real-world
problem solving activities across campus.


Vol. 42, No. 3, Summer 2008











THE FACULTY-INSIGHTS FROM OUR
AMBIDEXTROUS SCHOLARS
If you ask any one of us what characterizes the departmental
faculty best, we will tell you it is balance. We strive to create
an effective balance between research and education so that
students are exposed to an environment that maximizes their
learning, and we do it across the department as a pervasive
way of being. Such balance between excellence in teaching
and research is taken seriously, and our faculty members
demonstrate this characteristic by being active and visible in
both arenas-endeavoring to integrate research and educa-
tion in unique ways and to create new paradigms for student
achievement. Professor Donald Visco, for example, has re-
ceived both a Presidential Early Career Award for Scientists
and Engineers (PECASE) for his research on solving inverse
design problems, and the American Society for Engineering
Education ChE Division's
Ray W. Fahien Award for his
vision and contribution to Va
chemical engineering educa- E


tion.
So, what are the character-
istics that lead to a balanced
faculty member? The an-
swer, apparently, is that many
boundary conditions can lead
to similar outcomes.
Chairperson and Profes-
sor Pedro Arce was born in
Argentina and received his
undergraduate education in
his homeland's practice-
oriented engineering system
at the Universidad Nacional
del Litoral (Santa Fe). He
started his transformation
as a member of the presti-
gious National Council of
Research (CONICET) at
one of the leading research
and development institutes
(INTEC) of Argentina before
coming to the United States
in 1983. Arce's transforma-
tion was completed by the
great research and education
scholars at Purdue Univer-
sity, fusing-as his mentors
have-the desire to achieve
the perfect balance between
the two ideologies.
Professor Biernacki is the
undergraduate product of
120


Case Western Reserve University's research-driven program
of the late '70s and the more applied Doctor of Engineering
(DRE) program at Cleveland State University. He has 15 years
of industrial experience, yet retains a fundamental approach
to his research and says that "Teaching is a performance, and
I simply love the audience, the stage, and the script."
Assistant Professor Ileana Carpen, who is finishing up
her third year at Tennessee Tech, represents a great milestone
for the department. With a B.S. from Stanford, a Ph.D. from
Caltech, and a post-doctoral appointment at the University of
Twente, this faculty member demonstrates that it is possible
to recruit the finest academically trained individuals into our
program because it places an equal emphasis on both educa-
tion and scholarly research.
Assistant Professor Holly Stretz is the rarest of all-a high
school teacher turned Ph.D. chemical engineer. Stretz has a

TABLE 1
rious Awards and Honors of the Faculty
educational Scholarship and Related Service


2008 Outstanding Teaching Award, ASEE Southeastern Section
2008, 2007 Outstanding Faculty Award for Teaching, Tennessee Tech
2008, 2007, 2006 Tennessee Tech College of Engineering Brown-Henderson Award, for outstanding
Engineering faculty
2008,1' 2006[12] ASEE, Southeast Section, Thomas C. Evans Award, for best paper
2007 Outstanding Campus Representative, Zone 2, ASEE
2007 Outstanding Campus Representative ASEE Southeastern Section
2006[12] ASEE Corcoran Award, for best paper
2006 Quality Enhancement Program Award for Innovative Instruction, Tennessee Tech
2006[13] Annals of Research in Engineering Education, invited feature article
2006 Ray E. Fahien Award, ASEE
2004 Outstanding Campus Representative, 2nd Place, ASEE Southeastern Section
2001 ASEE Membership Award, ASEE Southeastern Section
Research Scholarship and Related Service
2007 Distinguished Faculty Fellow, Tennessee Tech
2007 American Concrete Institute (ACI), Fellow
2007 ACerS (American Ceramic Society) Profiles in Excellence
2007 Oronzio De Nora Postdoctoral Fellowship, The Electrochemical Society
2006 Invited Visiting Professor, University of Wollongong, Australia
2006 Oronzio De Nora Young Author Award, International Society of Electrochemistry
2006, 2005, 2002 Leighton E. Sissom Innovation and Creativity Award, Tennessee Tech
2006, 2005 College of Engineering Dean's Advisory Board Award for Excellence
2005 New Faculty Research Award (1st Place), ASEE Southeastern Section
2004 Presidential Early Career Award for Scientists and Engineers, DOE
2004 NNSA DOE-DP Early Career Scientist and Engineer Award, DOE
2003 New Faculty Research Award (2nd Place), ASEE Southeastern Section
2003 Outstanding Faculty Award for Professional Service, Tennessee Tech
2002[14] Sigma Xi Research Award, Tennessee Tech
2002, [15] 2000[16] Kinslow Engineering Research Award, for best paper, Tennessee Tech

Chemical Engineering Education










The faculty and staff
of the Department of
Chemical Engineer-
ing, from left to right,
front row:
Ileana Carpen,
Rebecca Asher, and
Holly Stretz;
back row: Don Visco,
Vijay Boovaragavan,
Perry Melton,
Venkat Subramanian,
Pedro Arce,
Joe Biernacki, and
Mario Oyanader


B.S. degree in chemistry from Texas A&M, worked in the
polymers and semiconductor industries for five years, taught
high school in the public school system, and studied in the
world-renowned research laboratories of Don Paul[4] at the
University of Texas, Austin.
Associate Professor Venkat Subramanian came to the
United States with an undergraduate degree from India's
distinguished Central Electrochemical Research Institute, to
join Ralph White's group at the University of South Carolina.
Subramanian merges his passion for applied mathematics with
electrochemistry and likely has one of the most productive
research groups on campus. He is presently digesting his work
into a text for undergraduate and graduate students.
Professor Visco, mentioned previously, studied at the Uni-
versity at Buffalo, State University of New York. He tutored
as an undergraduate student, an experience that he reports
would ultimately shape his career. Visco had an extended
industrial internship, served in the U.S. Navy, and is now
recognized as one of the nation's finest young researchers in
the field of inverse design, yet found time to be the depart-
ment's undergraduate program coordinator for the past four
years-weaving the fabric of our curriculum and creating
new and exciting opportunities in undergraduate chemical
and biomolecular engineering.
The department is also home to two other faculty members.
Adjunct Assistant Professor Mario Oyanader (originally
from Chile) is both an outstanding researcher and rising
young educator. Trained inArce's own group at Florida State
University (where Arce taught prior to becoming our chair),
Oyanader has the characteristics of both a scholar and an
educator who is "all about critical thinking." He is leading
the renaissance effort in process design and helping to re-cast
the Unit Operations Laboratory role within the new integrated
curriculum. Research Assistant Professor Vijayasekaran
(Vijay) Boovaragavan joined Subramanian's group as a


Vol. 42, No. 3, Summer 2008


post-doctoral researcher and was recently hired to his current
position. He received two international competitive distinc-
tions for his research while at Tennessee Tech and presently
co-teaches our Operations course.
The ChE Department at Tennessee Tech offers another
answer to the old question, "Research or education?" Schol-
arship in both is achievable, although a balance must be ac-
cepted. Collectively, the Tennessee Tech chemical engineering
faculty have earned 35 awards and distinctions since 2000 (see
Table 1). Many of these are top Tennessee Tech honors, oth-
ers are national recognition, and yet others are international
distinctions -the sum of which paint a picture of balance,
combining elements of research scholarship and excellence
in education. This does not happen by chance; emphasis must
be placed on maintaining a rational balance. Furthermore, our
experience is that an undergraduate program's excellence is
enhanced by strengthening the graduate program. The adage
"one cannot have a strong graduate program if too much at-
tention is paid to undergraduate education" is simply contrary
to the Tennessee Tech experience.

STUDENT-CENTERED LEARNING- INSIGHTS
FROM HOW OUR STUDENTS LEARN
It is well accepted that students learn best by doing,Ys] and
while there are a range of learning styles, most education
researchers would agree that active participation is key to
retention and ultimate internalization of learned informa-
tion.[6] Furthermore, most would also agree that educators
must relinquish their tradition of teacher-centered control
and place control in the hands of the students who must be
empowered to learn, ] i.e., we must provide student-centered
learning environments.[8] These pedagogical principles guide
the many changes that we are presently implementing across
the curriculum.
Active learning is one of several guiding principles being

121



















The 2005 National Cham-
pion AIChE Chem-E-Car
Team from Tennessee Tech,
from left to right:
Jonathan Phillips,
Braxton Sluder,
Jason Miller,
Regan Chandler,
Robert Phillips,
Jennifer Pascal,
and Haley Hunter.


AM& a.S .at .ia. Ato


Itu w au w k .&..


Integrated Curriculum
UOL
-'.- -- Packed Columns

.Extraction Station
-. I C apsltone LaD
SReactors Le el- Epenence

ILL Frontiers\ Tank and Pipe
Toolkit Stations Network
Bioreactor Capsto
SI / Term Project
Electrochemistry Heat Tran
HeatTransfer
Station
Micro-fluidics

Figure 1. The integrated lab-lecture (ILL) concept.

advocated at Tennessee Tech on a universitywide basis. The
Department of Chemical Engineering, however, has adopted
active and collaborative (team-based) learning at large, and
all of us are using some form of these approaches in our
classes. The most ambitious of these efforts is growing out of
our laboratory-and-lecture integration initiative. Our depart-
mental founders established and passed on what we refer to
as a "laboratory-oriented tradition." Four laboratory courses,
totaling five credit hours, were required up until 2006 when
the one-hour sophomore lab, the two-hour junior lab, and
the one-hour first-semester senior lab were integrated with
lecture courses across the curriculum. The remaining one-
hour, second-semester senior lab was preserved for what is
now called Capstone Lab. This bold move was met with some
skepticism when first introduced; however, the extraordinary
effort of the faculty to embrace this initiative has shaped and
defined this concept in unexpected ways.
Motivated by the observation that students are unable to
independently synthesize theory, computation practices, and
the "real world," Biernacki proposed to adopt the lab-inte-
gration concept across the entire ChE curriculum in 2005.)
122


Having pioneered small-scale labs and demonstrations as a
regular part of some of their courses already (e.g., courses on
momentum and heat transfer, reaction engineering, thermody-
namics, and process controls), the faculty team agreed that it
could be done on a larger, curriculumwide scale. Thanks to the
energetic leadership of Undergraduate Program Coordinator
(2003-2008) Visco, official integrated labs are now part of
six courses: Introduction to Mass Balances, all three transfer
science courses, Chemical Kinetics, and Solution Thermody-
namics/Separations. Newly named Curriculum Coordinator
Stretz is working diligently with Oyanader and other faculty
to re-create the role of the unit operations laboratory (UOL)
within the new curriculum. Figure 1 illustrates the concept.
Here, Level 1 experiences are related to our integrated labs.
These integrated activities draw from a variety of physi-
cal resources including a growing lab tool-kit, the existing
UOL, and New Frontiers chemical engineering stations. By
the time students reach the Capstone experience (Level 2),
they have numerous examples of how theory, computation,
and observation (experimentation) work together, thus they
are prepared to transition to a more independent, open-ended
capstone experience.
One final note is that the integrated lab and lecture is not
intended to be a course with a lab section (i.e., a lecture with
a lab). The single most important outcome of the lab-lecture
integration is to break down the barriers between theory,
computational practices, and the real-world. InArce's view,[71
"Traditional lectures give way to integrated environments with
a seamless transition from class to lab where the student is
in the learning driver's seat." The approach effectively uses
these three key elements of engineering education (classroom,
simulation, and lab activities) to create a continuum that is
seamlessly woven together, each providing input for valida-
tion of the other, and all working together. Certainly, some
individuals are gifted experimentalists and others, theoreti-
cians, yet the broader general education of the undergradu-
Chemical Engineering Education










ate must include a continuum that explores the relationship
between concepts (theory), calculations (the computer), and
the behavior of real things (the real world).

THE STUDENT CONTINUUM, RESEARCH
AND EDUCATION--INSIGHTS FROM
LESSONS IN SCALING
Arce was recently a recipient of the 2008 ASEE-SE
Section's Thomas C. Evans Award for best paper published
in engineering education during 2007. Arce is the only three-
time winner of this prestigious award, this time sharing the
honor with two co-authors, departmental adjunct faculty
member Mario Oyanader, and Steven Whitaker of University
of California at Davis, for their paper entitled, "The Catalytic
Pellet: ARich Prototype for Up-Scaling.[9]" In this paper, Arce,
Oyanader, and Whitaker explain that the traditional chemical
engineering curriculum and, for that matter, the traditional
engineering curriculum, attempts to teach design-a study
in scaling, if you will-in the final year of the program, thus
creating a sort of "step function" in design content of the
traditional curriculum. This approach effectively expects that
students will synthesize everything they have learned in the
past three years during two semesters, thereby transform-
ing them into "engineers." This approach, the paper argues,
has a long history but is fundamentally flawed. Many have
advocated "design across the curriculum," and similar pro-
grams, with some success. The paper's concept is to exploit
distributed laboratory courses and use real-world activities
(e.g., the experimental prototype,7 10]), thereby introducing
concepts of scale and effectively "scaling up" the knowledge
of students as they move through the curriculum-instead of
trying to accomplish the same all in one year. The approach
introduces a progressive type of curriculum that will require
new didactic materials (e.g., textbooks, simulations) and a
new vertical integration of the curriculum that, for Transport
Phenomena, is already in place at Tennessee Tech.
We take the concept even further in the department and,
in fact, view the student body, both graduate students and
undergraduates, as a continuum in the lifelong-learning
process. Barriers among student groups-juniors, seniors,
master's degree-seeking graduate students, Ph.D.-seeking
students, etc. -are considered obstacles to learning, growth,
and scholarly productivity. While there are a number of effec-
tive tools that can be used to unify the student body, we feel
that research is by far the most productive and learning-rich
vehicle. Research not only promotes critical thinking and
facilitates learning development, it also promotes the idea
of a community of learners among undergraduates, gradu-
ates, postdoctoral students, and faculty. Although not every
undergraduate will engage (formally) in research, we feel that
research activities should be a ubiquitous "fluid" that perme-
ates every classroom/lab and catalyzes as many elements of
the curriculum as possible, facilitating student learning and


critical thinking. To this end we have initiated a number of
integrating elements to our curriculum, and we advocate that
students engage in them along with university-led research
activities.
While there are numerous aspects of the program that em-
phasize integration, several merit special mention in addition
to the lab-lecture integration:
1) the Distinction in the Major (DITM) option, aformal-
ized and intensive undergraduate research track that
leads to a written and oral thesis defense;
2) the Research Seminar Series,1"' briefly mentioned
above;
3) the Chemical Engineering Graduate Research As-
sociation (CEGRA), a student-governed -,. ,, ,i. i .i..
that serves the needs of the graduate student body.

We feel these aspects are critical to the departmental suc-
cess. In combination with pedagogy that empowers students
to take charge of their own learning, we hope these aspects
will create a culture of scholarship emphasizing critical
thinking, problem solving, lifelong learning, and extending
the frontiers of knowledge. Our faculty, although relatively
small compared to others in the Southeast, is able to offer
Tennessee Tech's students a cross-section of frontier areas in
which to work. Arce and Oyanader are interested in electric
field-based processing, e.g., electrokinetic hydrodynamics,
corona discharge processing. Biernacki's main research focus
is experimental reaction kinetics, most recently emphasizing
portland cement-based materials. Carpen's interests focus
on complex fluids (the rheology of colloidal suspensions
and polymer composites) and on biomedical systems (tumor
growth and tissue engineering). Stretz's efforts focus on
the experimental characterization of nano-particle polymer
composite behavior, and nano-particle ordering and high
temperature behavior of similar materials. Subramanian's
computational research brings together electrochemistry, com-
plex transport modeling, applied mathematics, and computer
programming for the development of ultra-efficient algorithms
for real-time batteries and fuel cell performance prediction,
system optimization, and control. Subramanian is joined by
Boovaragavan, who is presently funded by a prestigious in-
ternational grant award from the The Electrochemical Society.
Finally, Visco's work involves a spectrum of computational
thermochemistry and molecular design initiatives as well as
laboratory research on phase equilibrium. Collectively the
faculty are or have been funded by numerous government and
private-sector organizations working closely with Tennessee
Tech's three state-funded research centers, and have many
ongoing or prior research collaborations. Such collabora-
tions enhance opportunities for our students and faculty to
work with leading scholars amid some of the finest research

The conceptparallels the "High Performance Learning Environment
(HiPeLE)" developed by Arce and collaborators, 2004, CEE.


Vol. 42, No. 3, Summer 2008











infrastructure in the world.
The outcomes are remarkable. In the last five years our
undergraduates have amassed top honors in numerous com-
petitions:
Our Chem-E-Car team has consistently been in the
top tier and has pioneered first-of-a-kind competition
vehicles, having established the current national record
and first place at the national competition in 2005 and
second at the AIChE Southeast Regional i I. t,,1, this
year (2008) with one of the first Bio-Cars to be entered
in the competition.

Collectively, our students have received top paper and
poster awards at regional and national competitions.

In 2006, Jennifer Pascal (one of our first DITM gradu-
ates and a current Ph.D. student) received the AIChE
Othmer Award, and Hope Sedrick was selected as one
of 10 students to receive a NIST SURF internship.

Recent graduate students have performed similarly well.
Ph.D. student Vinten Diwakar received The Electro-
chemical Society's Industrial Electrolysis and Electro-
chemical Engineering Division Student Achievement
award in 2006for his research on battery and fuel cell
modeling, (Subramanian, his Ph.D. advisor, was also a
recipient of this; .- ;,,. .-i award as a student).

Ph.D. student Pravin Kanan was selected for the BASF
International Summer Course in Bohn, Germany, in
S... -,,,,t* *- of his research on polystyrene foam thermal
decomposition.

Master's student John M. Richardson received an NSF
Graduate Research Fellowship in 2002 for his research
on nano-pore structure of hydrated portland cement, a
first for Tennessee Tech.

In addition, our B.S. and M.S. graduates are being
hired by leading companies. Recent Ph.D. graduate Dr.
Baburao was highly sought by design companies and
Dr. Swaminathan is currently a post-doctoral research
associate at the Technical University of Denmark.

The list of similar awards and achievements is long, but
these illustrate the excitement and success among our students
and, we believe, the result that is achievable with a program
that focuses on scholarship in both education and research
through integration rather than separation.

ABOUT TOMORROW-INSIGHTS FROM OUR
VISION OF THE FUTURE
It seems appropriate to dream and to speculate just a bit at
this point in the department's story. We recently revised our
vision statement as well as our PEOs, and after considerable
debate over two words-"will be"-we were convinced by
our Board of Advisors to phrase our vision statement in the
present tense and to use the word "is." Thus, our official Vi-
sion now reads as follows:


The Department of Chemical Engineering is a ... % i,-.., -
leader in chemical engineering education ;lar.,... .11 I., I.',
in teaching, research, and service.
This statement "is" a vision of the future for us-simple,
yet a bold supposition that we believe "will be." The path
from here to there for us is clear: (1) continue to respect and
build upon the foundations of our legacy; (2) develop and
grow a faculty that "plays like a team;" (3) have clearly stated
educational objectives that are simple and timeless; (4) main-
tain a balance between research scholarship and education,
and strive to excel in both; (5) fervently maintain a student-
centered learning environment; and (6) integrate the student
body, integrate research with education, and do not let "size
effects" influence the scale of our productivity.

REFERENCES
1. Arce, PE., "Colloquial and Coach Approach Environments in Engi-
neering Education: Identification and Role of POK's," Proceeding of
the 8th Latin American Congress on Heat/Mass Transfer, LATCYM
2001, 551-554, Puerto Veracruz, Mexico, February 2001
2. Arce-Trigatti, M.P, and PE. Arce, "Parallel Between Active Learn-
ing and Coaching Team Sport Techniques: Analysis and Selected
Examples," Annual Conference Proceedings, American Society for
Engineering Education (CD-ROM), June 2000
3. ,
slide #15 of Tennessee Technological University Quality Enhancement
Plan
4. Koros, W.J., "Don Paul of the University of Texas at Austin," Chem.
Eng. Ed., 35(2) 86 (2001)
5. Cardellini, L., "An Interview with Richard M. Felder, "J. Science Ed.,
3(2), 62 (2002).
6. Johnson, D.W, R.T. Johnson, and K.A. Smith, Active Learning: Co-
operation in the College Classroom, Interaction Books, Edina, MN
(1991)
7. Arce, PE., and L.B. Schreiber, "High-Performance Learning Environ-
ments, Chem. Eng. Ed., 38(4) 286 (2004)
8. Creighton, L., "Kicking Old Habits," ASEE Prism, p. 33, April
(2001)
9. Arce, P M.A. Oyanader, and S. Whitaker, "The Catalytic Pellet: A
Rich Prototype for Up-Scaling," Chem. Eng. Ed., 41(3), 187 (2007)
and Plenary Lecture at the ASEE SE Section Annual Meeting, the
University of Memphis, April 6, 2008.
10. Arce, P, J.J. Biernacki, and P Melton, "The Experimental Prototype:
Critical Thinking and Real-World Problem Solving in Engineering
Education," presented at the AIChE Annual Meeting, San Francisco,
November 2006
11. Tennessee Tech Department of Chemical Engineering, Research
Seminar Series, htm>
12. Biernacki, J.J., "A Course-Level Strategy for Continuous Improve-
ment," Chem. Eng. Ed., 39(3) 186 (2005) and Plenary Lecture at the
ASEE SE Section Annual Meeting, the University of Alabama, April
2006.
13.
14. Visco, D.P Jr., R.S. Pophale, M.D. Rintoul, and J.L. Faulon, "Devel-
oping a Methodology for an Inverse Quantitative Structure-Activity
Relationship Using the Signature Molecular Descriptor, "J. Molecular
Graphics and Moidellin,, 20, 429 (2002)
15. Biernacki, J.J., P Stutzman, and P.J. Williams, "Kinetics of the Reac-
tion Between Fly Ash and Calcium Hydroxide, "ACI Mat. J., 98(4),
340 (2001)
16. Visco, D.P Jr., and D.A. Kofke, "Modeling the Monte Carlo Simulation
of Associating Fluids, "J. Chem. Phys., 110, 5493 (1999) 1


Chemical Engineering Education











outreach
-0


"FINGER KITS:"

An Interactive Demonstration of Biomaterials and Engineering

for Elementary School Students



HEATHER E. CANAVAN, MICHAEL STANTON*
University of New Mexico Albuquerque, NM
KAORI LOPEZ
Albuquerque Public Schools Albuquerque, NM
CATHERINE GRUBIN
University of Washington Seattle, WA
DANIEL J. GRAHAM
Asemblon Corporation Seattle, WA
F rom personal computers to cell phones, there is no Heather E. Canavan is an assistant professor of chemical engineering at
dispute that technology developed by scientists and the University of New Mexico. She has a B.A. in biology from UC Santa Bar-
engineers greatly influences the lives of U.S. residents. bara, and an M.S. and Ph.D. in physical chemistry from George Washington
University. Her research expertise is in the area of biomaterials, specifically
Despite the increasingly important influence that science mammalian cell interactions with "smart" polymers. She is the founding
and engineering have on our lives, however, the numbers of director of the UNM Biomaterials Engineering Outreach Program and a
member of the Center for Biomedical Engineering (CBME) at UNM.
undergraduate and graduate degrees awarded in these areaMichael Stanton is a doctoralstudent at the University of New Mexico in the
to U.S. students have decreased steadily for the past several College of Education, Organizational Learning and Instructional Technol-
decades.[1 In addition, the numbers of women and minorities ogy. He has a B.A. in secondary education and M.A. in special education
from UNM. He is a consultant to K-12 and post-secondary institutions,
in this field remain low compared to our nation's demograph- providing expertise through his research in design and implementation
ics (e.g., New Mexico is now a minority-majority state). [2] of systems that transform and redesign classrooms, systems, and orga-
St r m e h e nizations. He serves as a graduate assistant managing and facilitating
For these reasons, many efforts have emerged with the goal all outreach activities to schools for the UNM Biomaterials Engineering
of attracting students into engineering and science disciplines, Outreach Program and a member of the Center for Biomedical Engineer-
including outreach efforts such as those sponsored by the ing (CBME) at UNM.
National Science Foundation. Catherine E. Grubin is the lead teacher in the University of Washington's
National Science undation. Engineered Biomaterials program (UWEB). She has a B.S. from The
The authors of the present work include researchers from Evergreen State College and a Ph.D. in immunology from the University
of Washington. She has been involved in science education since 2000,
both the University of New Mexico (UNM) and the University working on K-5 curricula with The Chicago Science Group, middle school
of Washington (UW). The technical expertise of the authors science with the Youth Take Heart program, and 7-12 with UWEB and the
Seattle Biomedical Research Center. She is currently pursuing her M.A.
is in the field of biomaterials, or the interaction of man-made in teaching at Seattle Pacific University.
materials with biological systems. The term "biomateri- Kaori A. Lopez is the coordinator and instructor of an elementary school
als," therefore, encompasses a number of research interests Model United Nations program in Penasco, N.M., that integrates social
studies and science in an authentic setting. She has her K-8 teaching
including microbially induced corrosion of ship hulls, the certification, bilingual endorsement and endorsement in gifted education.
development of DNA microarrays, and the optimization of She has focused her instruction on project-based, integrated curriculum
materials used for biological implants. While it is unlikely with fifth-graders.
Daniel J. Graham is a founder and principal scientist of Asemblon Inc.,
students in the fifth grade (the target audience of the fol- in Redmond, Wash. He has a B.A. in chemical engineering from Brigham
lowing demonstration) are familiar with DNA microarrays, Young University, and a Ph.D. in bioengineering from the University of
Washington. His research expertise is in the area of surface modification
and characterization. Dan has been involved with K-12 outreach for more
*Also affiliated with Albuquerque Public Schools than 10 years, and is one of the inventors of the Finger Kit project.
Copyright ChE Division of ASEE 2008
Vol. 42, No. 3, Summer 2008 125











many have already been exposed to implanted materials, as
they or members of their families may have contact lenses,
use a glucose monitor for their diabetes, or have hip or other
implants. Therefore, giving students a project that emphasizes
biomaterials taps into something with which they already
may be familiar and that is becoming increasingly important
in the lives of millions of people around the world. As noted
by Tobias,[3] tying science to societal issues via cooperative
and interactive learning styles may increase participation by
women and other under-represented minorities in science
and engineering. Furthermore, as more people are going to
be affected by biomaterials, there will be more opportunities
for exciting, rewarding jobs in the field of biomaterials. It
has been estimated that by the end of this decade, there will
have been more than a 30% increase in bioengineering-related
employment positions.[4]
To capture part of this excitement, we present a real-world
problem to the student: Someone has an injured finger joint,
and the students in the class need to design an implant to
replace it. After presenting the problem, we discuss how
the students could go about making a replacement finger
joint. In order to do this, the students need to understand
what comprises a finger. Next, the students have to consider
what materials are available that match the properties of the
components in a finger. As this is an engineering project, we
ask students to develop design goals for the finger (e.g., that
it is flexible, bends only in one direction, is able to pick up
an object).
At UNM, we made a number of adaptations to the original
design. For instance, to better communicate with the bilingual
and monolingual (Spanish-only speaking) students within the
Albuquerque Public School (APS)
district, we translated the text used


in our brief presentation into Span-
ish and recruited Spanish-speaking
volunteers. Furthermore, our par-
ticipating volunteers are typically
a combination of undergraduates,
graduate students, postdocs and
faculty to show representatives of
women and minorities that have
gone on to successful engineering
careers. This hands-on activity en-
gages the students' creativity while
also teaching them a basic under-
standing about what biomaterials
are and how one would go about
designing and building them.

BEFORE THE VISIT
Prior to any outreach events, the
following should be addressed: 1)
provisions of key ideas and vocabu-
lary necessary to understand the les-


son, 2) assembly of kits, and 3) training of volunteers.
Vocabulary
To achieve the first task, our outreach coordinator (co-author
Stanton), when scheduling outreach visits, provides teachers
and principals with vocabulary words and concepts necessary
to understand the lesson. Table 1 presents the key vocabulary
that the outreach director (co-author Canavan) and elementary
school teacher (co-author L6pez) discussed prior to the first
UNM visit to L6pez's classroom. These vocabulary words are
also defined early in the interactive talk portion of the visit
to reinforce these concepts before more difficult matter is
discussed. In addition, each of the visits is scheduled such that
it follows instruction on the human body. Table 2 identifies
the parts of the body especially important when considering
restoring the function of the finger joint. While this is not a
comprehensive list (e.g., fingernail is not listed), these are the
body parts that make primary contributions to the function of
the joint, which is the focus of the lesson. For example, it is
the contraction and expansion of the muscles that lead to joint
movement, and bone which provides structural support.
Assembly of Kits-Materials
Table 3 lists the contents to be included in each kit. No
additional supplies are required to perform this activity. For
each of the suggested materials (e.g., chalk), a potential use
in the finger kit design is listed (e.g., bone). Therefore, each
of the materials listed in Table 3 should approximate those
of the human body listed in Table 2. It is important to note
that we provide this list to aid volunteers using this demon-
stration, but we do not provide it to the students themselves.
Also, we often see students find creative uses for the materials
provided that we had not initially envisioned (e.g., the use

rl A 1 1


IAJ L t 1I
Vocabulary given to 5th grade teachers prior to demonstration to prepare for visit.
Vocabulary word Definition
Biology The study of living organisms.
Engineer (noun) A person who designs, builds, or maintains engines or machines.
Implant A tissue or an artificial object in a person's body introduced by surgery.
Material The substance of which a thing is made, such as wood, glass, or metal.
Prosthetic An artificial body part, such as a leg or a heart. May be internal or external.


TABLE 2
Parts of the body taught in 5th grade module prior to visit by UNM.
Additional body parts may be added by students (e.g., fingernail).
Part of the Finger Function
Arteries and veins Blood flow
Bone Structural support, mechanical strength
Muscles Contract and extend to move joints
Nerves Sensation of heat/cold, movement of body via attachment to muscles
Skin External surface of body
Tendons Attach muscles to bone

Chemical Engineering Education











of a paperclip as a fingernail to make the finger aesthetically
more pleasing, instead of as a structural element to hold the
design together). Although most of the parts can be re-used
(e.g., pipe cleaners, straw, etc.) and the kits recycled through
many events, at UNM each kit is used only once, and the
students retain possession of their designs. Therefore, to
maximize the number of students we can interact with over
a year, the materials used in the finger kits are low-cost items
that can be purchased in bulk (e.g., Popsicle sticks vs. tongue
depressors). We estimated that, when the contents of the kits
are purchased in bulk (e.g., to make 400 kits), the cost of the


Figure 1. Examples of finished finger designs. Note
that the resulting design will reflect the design consid-
erations agreed to by the students at the beginning of
class, as well as their own individual preferences. For
example, design a) uses a Popsicle stick as a finger-
nail, while design d) uses it as a brace to
prevent backward movement.


kits fall to ~$1 student.
Training of volunteers
It is advisable to train all volunteers prior to the outreach
"season" so they understand what to expect from the events.
In particular, the volunteers should understand how the ma-
terials in the kit can be used in designs. Such an event will
also yield a number of diverse designs (as illustrated in Figure
1), demonstrating to the volunteers that they may see many
different ways the materials will be used by the elementary
school students. Also, the volunteers should be briefed about
what to expect from a visit to an elementary school, including
any necessary information about the school's dress code. If
possible, the teachers participating in outreach visits should be
invited to speak with the volunteers about the general level of
knowledge and understanding, as well as modes of learning,
that young students demonstrate.

DURING THE VISIT
The outreach visit contains several elements: 1) a brief pre-
sentation outlining the topic and project (~15-20 minutes); 2)
discussion and formulation of the design and test parameters
(~2-5 minutes); 3) fabrication of the designs by the students
(~30 minutes); and 4) evaluation of the design according to
the parameters previously outlined ( 10 minutes). In addition,
a fifth step (evaluation of the efficacy of the visits) may be
conducted after the visits to allow for any modifications or
improvements to be made as needed.
Introductory Talk
At the beginning of the visit, the lead volunteer will give
a brief talk51 to introduce the range of topics in bioengineer-


TABLE 3
Contents of the Finger Kit. Note that all materials are meant to be commonly commercially available and of low cost.
Substitutions, deletions, and additions to the kit may be made to accommodate
the preferences of the demonstrators, as well as the availability of material.
Qty Item Potential Use in Finger Design
1 Zip-close sandwich bag None (used to contain other parts listed below)
3 Pieces of clear, flexible tubing: pieces ~1.5" long and wide "Skin"that holds all parts together
enough for dowel pieces/chalk to slip into
1 Tongue depressor/Popsicle stick "Brace" to prevent fingers from bending backwards
4 Toothpicks "Brace" to prevent fingers from bending backwards/sideways
1 Pipe cleaner Actuator (muscle/tendon) or body part (blood vessel/nerve)
1 Piece of copper wire 10" long (~24 gauge) Actuator (muscle/tendon) or body part (blood vessel/nerve)
(muscle/tendon) or body part (blood vessel/nerve)
1 Flexible straw Actuator (muscle/tendon) or body part (blood vessel/nerve)
4 Rubber bands Actuator (muscle/tendon), body part (blood vessel/nerve), or to
hold design together (tendons)
3 Wooden dowels, -1.5"long x 0.25" diameter Bones
3 Half-pieces of chalk Bones
2 Small paper clips To hold design together (tendons) or body part (fingernail)
2 Large paper clips To hold design together (tendons) or body part (fingernail)

Vol. 42, No. 3, Summer 2008 12










ing, including examples of biochemical engineering (e.g.,
pharmaceuticals and dialysis), biomechanics (e.g., grafting
procedures, prosthetics, and implants), and biomaterials (e.g.,
contact lenses). These subjects, which were first introduced
to the students using the vocabulary list (see Table 1), are
used to lay the groundwork for the design students will be
performing. Therefore, it is imperative they understand what
the terms mean. It is important to remember that most fifth-
grade students are (at best) unfamiliar with even the most
basic language traditionally used in bioengineering, and often
hold erroneous beliefs (such that the term "engineer" solely
applies to people who repair car engines).
We have found that the best way to engage the students'
attention and get these concepts across is to make the talk
highly interactive, with the lead volunteer asking the students
questions and listening to their responses. For instance, the
students become quite animated when asked: "Do you know
anyone who has contact lenses or hearing aids?" Such ques-
tions solidify ideas in the students' minds, as they are able to
make connections between the new material and something
they are already familiar with. In fact, according to elementary
school teacher and co-author L6pez, the most important factor
for a successful demonstration is to pay attention to cues from
the students to determine if they are understanding the mate-
rial, and therefore connecting bioengineering to their lives.
Later in the talk, the materials used in biomaterials are dis-
cussed. In particular, there is significant emphasis placed on
how the properties of a material are used to match a specific
function. For instance, metal implants are often used in bone
replacements (due to their mechanical strength), whereas
flexible rubber would not be a suitable replacement. Also,
the idea that a design may be perfected over time (such as the
early use of ear trumpets prior to the invention of a hearing
aid) can show students that rarely is a design "perfect" from
its first prototype.
Finally, we use the talk as a chance to educate the students
about the career path that bioengineers take, from their cur-
rent position to college and graduate school. Many of our
participating volunteers are from the local community (20%
of UNM volunteers attended APS as students, and 78% of
APS participants attended APS as students). As importantly,
many volunteers are members of groups traditionally under-
represented in science and engineering (45% are women, and
50% identify themselves as Latino, Hispanic, or Chicano) and
are Spanish speakers (35%).
Formulate Design and Test Parameters
At the conclusion of the talk, the students are told they will
not have to wait until college or graduate school to start their
research career, and that they will be bioengineerss for a day."
The students are told they will design a replacement finger
and will test their designs according to parameters they agree
upon. This leads to a discussion to elicit design goals from


the students themselves. For instance, the volunteers may
ask, "What do fingers do?" [They bend.] "Could you play the
piano/push a doorbell if your fingers 'bent' in all directions?"
[No, your finger would just bend backwards if you pressed
on the keys/buzzer.] "Can you be more specific about how
they should bend?" [In one direction only.] "What could you
do to allow them to bend, but in just one direction?" [This is
where students tend to start equating the materials to what
actually makes up their fingers.] We have found that asking
the students to come up with their own rubric for a good
design is far more engaging than providing these parameters
directly. In addition, this method requires higher-order think-
ing skills and often causes a great deal of excitement among
the students. Figure 2 shows a page from a student's notebook
that lists the design considerations students agreed a finger
should have, as well as the parts of the body that should be
included (mitiriols [sic]).
Fabrication of the Designs
Once the design parameters have been agreed upon, the
volunteers distribute the kits to the students, and fabrication
of the designs begins. During this portion of the visit, the
volunteers and teacher should circulate among the students.
While we discourage the volunteers from telling the students
how to make the design, they can provide guidance, remind-
ing students about the parameters (e.g., if the finger bends,
but does so in many directions, it hasn't been optimized). If
students are stumped, the volunteers can help provide prompts
to get the students working on their designs. For instance,


Figure 2. Page from a student's notebook at Longfellow El-
ementary. The student has listed the design considerations
the students agreed a finger should do, as well as the parts
of the body that should be included (materials). She has
also illustrated her own finger for the design.


Chemical Engineering Education











the volunteers can ask: "What body parts are in a finger?
[Bone.] Is there anything in this kit that reminds you of that
part? [Chalk or wooden dowel.] This also gives the students
a chance to interact with their potential role models and ask
them questions about themselves while they work on the de-
signs. Furthermore, this gives the volunteers a chance to stress
the creative aspects of science and engineering. In addition,
the students often begin to observe the designs used by others
to overcome a particular challenge (e.g., the use of toothpicks


Figure 3. Examples of the different modes of learning during a
to fifth-grade students at Longfellow Elementary School (Albu
que, NM). UNM volunteers circulate throughout the classroom
guidance and providing translation into Spanish, where necess,
students often teach each other lessons learned from their desi
classic example of peer learning (b); a student consults the har
skeleton to determine the parts of the finger (c); the students' t
reminds the fifth-grade bioengineers that one
criterion of their design is that it must bend (d).


as "braces" to prevent their design from bending backward)
and incorporate it into their own. Such peer mentoring is a
natural occurrence in this environment. Figure 3 illustrates
each of these modes of learning during a visit from the UNM
Biomaterials Engineering Outreach Program to Longfellow
Elementary School.
Evaluation of the Design
At the end of the visit (or in a post-visit session with their
teacher), students evaluate their designs according
to the parameters to which they previously agreed.
Table 4 is an example of a rubric designed by the
students in co-author L6pez's fifth-grade class,
including the design parameters (or criterion) con-
sidered important by the students, as well as their
standards for the design. For example, the students
considered the ability to hold something up as an
important criterion, and a design that is capable of
holding up a pencil in the middle section of the finger
may be considered an advanced design. Often, the
students come up with new designs or improve-
ments not initially listed as criteria in the rubric. For
instance, some students will attempt to improve the
aesthetics of the design by including a paperclip as
a fingernail. The volunteers and teacher may then
remind students that many designs are improved
over time for aesthetic reasons (e.g., less noticeable
dental fillings and hearing aids).

I visit AFTER THE VISIT
quer- The design evaluation is considered the final stage
giving of the formal visit, and the students are allowed to
iry (a), keep their designs. As well, some teachers may want
gnS, a
d ofa to display the top designs or keep them for follow-
?acher up. In most cases, students from the classrooms we
visited sent "thank you" letters to the demonstrators,
relating what they learned, how much they enjoyed


TABLE 4
Example of a rubric used to evaluate the students' designs. Note that the criteria used will vary to reflect those that are pro-
posed and agreed upon by the students at the beginning of the demonstration.
Criterion Emergent (1) Nearing proficient (2) Proficient (3) Advanced (4)
Moves like a Doesn't move. Moves somewhat, but in Moves forward. Moves forward, not back-
finger. too many directions that wards. Does not move to
are not appropriate, side; only the whole finger
can move side to side. Has
two places where it can
bend.
Has components Has components that Has only two compo- Has bone, skin, muscles, Has bone, skin, muscle,
of a finger. don't function like a nents: possibly bone and and fingernail. Compo- fingernail, capillaries,
finger. Components are muscle or skin and bone. nents are generally in the nerves and other sensors.
not in the right place. Components may not be correct place. Components are in the cor-
in the right place. rect place.
Holds something Doesn't hold it up at all. Barely holds it up-falls Holds up pencil. Holds up pencil in the
up (e.g., pencil), quickly. middle section of finger.

Vol. 42, No. 3, Summer 2008 12











...the hands-on "finger kit" demonstration addresses a number of benchmarksfrom

the State of New Mexico61 as well as the Project 2061 Benchmarks for Science Lit-

eracy (from the American Association for the Advancement of Science, AAAS),

after which many states have modeled their standards."7


the visit, and their interest in science. The letters are always
appreciated by the volunteers, and help reinforce the value of
outreach activities to them. In addition, it is also a chance for
the outreach organizers to get valuable feedback. For instance,
due to the large number of students who mentioned in their
letters how much they wanted to keep their designs (to show
their parents or siblings, or to improve on the design), we at
UNM now make kits for each student, rather than attempting
to recycle them (as had been done previously).

EXPECTED IMPACT
As previously stated, one of the primary complaints that
elementary school teachers have about outreach projects is
that they are often considered stand-alone demonstrations with
little thought to how they will be integrated into the regular
curriculum. As it is described above, the hands-on "finger kit"
demonstration addresses a number of benchmarks from the
State of New Mexicod61 as well as the Project 2061 Bench-
marks for Science Literacy (from the American Association
for the Advancement of Science, AAAS), after which many
states have modeled their standards."7 The individual bench-
marks pertaining to the fifth grade are outlined below.
NM Benchmarks Addressed
Strand I/Standard I/Benchmarks I & II (Scientific Think-
ing and Practice): By observing and experimenting on their
model, and analyzing their product (using the rubric), the
students learn to understand the scientific method. In addition,
the students learn to communicate their findings during the
class discussion at the end of the event, and learn that their
conclusions are subject to peer review.
Strand II/Standard I/Benchmarks II & III (Physical Sci-
ence): By addressing how the muscles of the finger work and
discussing the energy source that makes fingers move, this
project addresses the state benchmark pertaining to forces and
motions. While studying the action of muscles and tendons
on the finger joint, the students learn that when a force acts
upon an object, it will move in a different direction.
Strand II/Standard II/Benchmark III (Life Science): By
addressing the purposes of the finger joint (in relation to
replacing that function using a prosthetic), the students learn
the properties, structures, and processes of living things, and
how cells and tissues are related to the behavior of an entire
organism.
Strand III/Standard I/Benchmark I (Science and Society):
With its emphasis on how machines have been engineered
130


to aid in human health (e.g., glucose monitors, hearing aids,
etc.), the introductory talk demonstrates to students how
technology has affected the lives of individuals. Emphasizing
how rudimentary prosthetics have evolved allows students to
understand how scientific discoveries, inventions, practices,
and knowledge are influenced by individuals and societies.
Furthermore, in an integrated elementary curriculum, the
progress of bioengineers could be connected to the effect on
social issues.
AAAS Benchmarks Addressed
Benchmark 1B (The Nature of Science/Scientific Inquiry):
Developing the rubric to evaluate their designs allows students
to learn that scientific investigations may take many differ-
ent forms, including observing what things are like or what
is happening somewhere, collecting specimens for analysis,
and doing experiments.
Benchmark 1C (The Nature of Science/The Scientific
Enterprise): Discussing their results reinforces to students
that clearly communicating their results is an essential part
of doing science. Because the volunteers participating in
this event are both men and women of many different ages
and backgrounds, it is reinforced that people who perform
scientific work come from all populations.
Benchmark 3A (The Nature of Technology/Technology and
Science): As they learn to design replacement finger joints,
students learn that technology extends the ability of people
to change the world, often in response to the need to meet
basic survival needs.
Benchmark 3B and C (The Nature of Technology/Design
Constraints and Systems and Issues in Technology): While
building their designs, the students rapidly grasp the need
to match properties of materials and engineering principles
while designing solutions to problems. Simultaneously, the
students learn about design tradeoffs. For instance, the finger
design with the best side-to-side stability may aesthetically
be the least pleasing.
Benchmark 6C (The Human Organism/Basic Functions):
As they learn how nerves stimulate the muscles in a joint to
contract, the students learn how the human body functions
as a system-with the brain giving signals to the body to
stimulate movement.
Benchmark 8F (The Designed World/Health Technology):
By learning about prosthetics and designing a replacement
body part, students learn that technology has made it possible


Chemical Engineering Education











to repair and sometimes replace some body parts.
Benchmark 11A (Common Themes/Systems): By learning
about how the materials in their kits (and the parts of the body)
work together as a system, students learn that the parts of a
system influence each other and may not work well if they
are broken, worn out, or misconnected. This is applicable to
both their own creation as well as the body part it is meant
to replace.
Benchmark 12C (Habits of Mind/Manipulation and Obser-
vation): By asking the students to relate the material properties
of the objects in their kits (e.g., chalk) to those in the human
finger (e.g., bone), the students learn to choose appropriate
common materials for making simple mechanical construc-
tions and repairing things.

CONCLUSIONS
This work describes one hands-on activity and demonstra-
tion developed at UW and further refined at UNM. The goal
of the project is to provide a hands-on experience with an
engineering project. While the project itself is goal-oriented,
it is also creative and open-ended, with many possible so-
lutions to the problem presented. In this way, the creativ-
ity involved in the project is emphasized, rather than only
relying on science and math ability. In our experience, the
demonstration works best when it is tailored to suit the needs
of the community, and we recommend that anyone adopt-
ing this outreach demonstration take the time to do so with
their own community. It is for these reasons that we at UNM
chose to focus our activities on one grade level (5th grade)
in one school system (APS). Furthermore, the specific needs
of the community (e.g., bilingual students in NM) should
be addressed (e.g., translation of the slides into Spanish).
Finally, we chose simple, low-cost materials to maximize


the number of students that can be reached with the activity.
Ultimately, we strove to develop a fun experience that will
get students excited about career opportunities in science and
engineering. After all, although not all students will ultimately
pursue science and engineering-related careers, we feel that a
general population more educated in the area of science and
engineering is also a valuable pursuit.

ACKNOWLEDGMENTS
This work was supported by NSF-Partnerships for Research
and Education in Materials (PREM) Program grant # DMR-
0611616 to the UNM Biomaterials Engineering Outreach
program, and NSF-Engineering Research Center (ERC) Pro-
gram grant # EEC-9529161 to the University of Washington
Engineered Biomaterials Outreach (UWEB) program. We also
thank the many volunteers who have participated in outreach
events in Washington and New Mexico, including the UNM
Student Chapter of the Biomedical Engineering Society. In
particular, we thank Danielle Garcia, Rosalba Rinc6n, and
Ulises Martinez for assistance translating our presentation
into Spanish.

REFERENCES
1. Mannix, M., "Getting It Right, "Prism, 10(7), 14(2001)
2. Smith, T.Y., "Science, Mathematics, Engineering, and Technology
Retention Database," Making Strides, 2(2) (2000)
3. Tobias, S., They're Not Dumb, They're Different: Stalking the Second
Tier, Tucson, AZ Research Corporation. 70 (1990)
4. i ... ..... I ,,I Outlook: 2000-2010, U.S. Dept. of Labor, Ed., Monthly
Labor Review (2000)
5. UNM Biomaterials Engineering Outreach Materials Online: W W W -C Ih.., u,,,11 ..... I 1 ,I I, I l1ll1l , 1,, I lh 11'i1. ,,III, l 1/
k8.htm>
6. New Mexico Public Education Department:

7. AAAS Benchmarks Online:



Vol. 42, No. 3, Summer 2008











M classs and home problems


GEOTHERMAL COGENERATION:

ICELAND'S NESJAVELLIR POWER PLANT


EDWARD M. ROSEN
EMR Technology Group Chesterfield, MO 63017
energy use in Iceland (population 283,000) is higher per
capital than in any other country in the world.11 Some
53.2% of the energy is geothermal, which supplies
electricity as well as heated water to swimming pools, fish
farms, snow melting, greenhouses, and space heating.
The Nesjavellir Power Plant is a major geothermal facility,
supplying both electricity and heated water to Reykjavik. The
purpose of this paper is to interest students in geothermal
energy, describe a simulation of this plant, and determine the
plant's suitability for classroom study.

PLANT DESCRIPTION
The plant (commissioned in 1998[2]) is located near one of
the largest high-temperature fields in Iceland.[3]
Iceland's high-temperature fields are so rich in gas and
minerals that the waters cannot be used directly in the distri-
bution system.[4] Its high pressure and thermal energy, how-
ever, makes it suitable for heating fresh water and generating
electricity.
Ballzus, et. al.,[2] provide a plant flow diagram (Figure 1)
on which stream flows and temperatures are indicated. Where
data is specified, the diagram is modified to include stream


names (e.g. {S 1}, {S2}). In addition, the heat exchangers are
labeled ({HX1}, {HX2}, {HX3}).
Steam mixed with water {S 1} is conveyed from boreholes
through collection pipes to the separation station, where the
water is separated from the steam. Excess steam and unused
water go into a steam exhaust outside the separation station.
From the separation station, steam and water proceed by
separate pipes to the power plant at a pressure of about 12
bara and a temperature of 190 C. The steam (after passing
through a mist eliminator) is conveyed to steam turbines,
where electricity is generated. Each turbine (two of them)
produce 30 MW of electricity (MWe).

Edward M. Rosen received his B.S. and M.S.
degrees in chemical engineering from the Illinois
Institute of Technology, and his Ph.D. in chemi-
cal engineering from the University of Illinois.
After retiring from the Monsanto Company, he
founded EMR Technology Group. He has served
as a trustee of the CACHE Corporation and as a
program evaluator for ABET. With E.J. Henley,
he is co-author of Material and Energy Balance
Computations (Wiley 1969).


Copyright ChE Division ofASEE 2008


Chemical Engineering Education


The object of this column is to enhance our readers' collections of interesting and novel prob-
lems in chemical engineering. Problems of the type that can be used to motivate the student by
presenting a particular principle in class, or in a new light, or that can be assigned as a novel home
problem, are requested, as well as those that are more traditional in nature and that elucidate dif-
ficult concepts. Manuscripts should not exceed 14 double-spaced pages and should be accompanied
by the originals of any figures or photographs. Please submit them to Professor James O. Wilkes
(e-mail: wilkes@umich.edu), Chemical Engineering Department, University of Michigan, Ann
Arbor, MI 48109-2136.





































































































Vol. 42, No. 3, Summer 2008




























































































Chemical Engineering Education











In the condenser {HX1} the steam exhaust from the tur-
bines is utilized to preheat cold water {S35}. This cold water
is then further heated in the heat exchanger {HX3} by the
separated geothermal fluid {S3}. (A second heat exchanger
{HX2} can be utilized to preheat a portion of the cold water
with the separated geothermal fluid from {HX3}. In this
simulation, however, {HX2} is not utilized). Since the min-
eral-rich geothermal fluid causes scaling that coats the heat
exchanger pipes, steel particles are allowed to circulate in the
stream, impacting against the pipes to remove any scaling as
it occurs.s5]
The cold water {S21} is saturated with dissolved oxygen
that corrodes steel after being heated. To rid of the oxygen,
the water is sent to a vacuum deaerator.J61 The main flow
{S 11} enters the central part of the deaerator. The water boils
vigorously (due to a vacuum) and sprays over filling material.
Steam and gas rise to the top. The steam is condensed through
the injection of cold water {S30} before the gas is ejected.
Finally, a very small quantity of steam containing acid gases
{S37} is mixed with the water to eliminate the last traces of
dissolved oxygen and lower the pH of the water in order to


TABLE 1
Composition of Geothermal Fluid Stream {S1}
Vapor Fraction 0.3527
Temperature (Deg C) 189.2
Pressure (kPa) 1235
Flow (kg/s) 326
Enthalpy i 1., 1500


Water (kg/s) 325.56
Hydrogen Sulfide (kg/s) 0.1495
Carbon Dioxide (kg/s) 0.2875
Oxygen (kg/s) 0
Sulfur (kg/s) 0


TABLE 2
Composition of Cold Water Stream {S21}
Vapor Fraction 0
Temperature (Deg C) 5
Pressure (kPa) 101.33
Flow (kg/s) 1129
Enthalpy li l., 21.1


Water (kg/s) 1128.985
Hydrogen Sulfide (kg/s) 0
Carbon Dioxide (kg/s) 9.889E4
Oxygen (kg/s) 0.0144416
Sulfur (kg/s) 0

Vol. 42, No. 3, Summer 2008


prevent precipitation in the distribution system. The following
reaction takes place. 71

2HS (g) + 02 (aq) = => 2H20 (aq) + 2S (s)

Small quantities of H2S ensure the dissolved oxygen that
could get into the storage tanks is eliminated The H2S also
gives the water the "good smell" for which the water from the
water supply system in Reykjavik is known today.

THE VMGSIM SYSTEM[8]
The VMGSim system is a modern interactive process
simulation system. One of the partners of VMG (Virtual
Materials Group) founded Hyprotech and another created
and wrote most of HYSIM. As a general policy, VMGSim
is provided to universities free of charge when used for aca-
demic purposes.
The system uses Microsoft Visio for the graphical input
engine. Amenuis provided that allows the user to drag streams
and unit operations onto a graphical screen to build a complex
system. The system uses the interactive calculation principles
of nonsequential unit operation calculations with partial data
flow. It is considered to be the fastest approach developed for
creating and evaluating process models. Equilibrium stream
calculations are carried out as pressure-enthalpy flashes.
The physical property system has been carefully crafted
and evaluated to allow the user to have confidence in it. A
simple click of the mouse will allow the user to evaluate dif-
ferent physical properties for his/her simulation. Similarly,
different units (SI, Field, etc) can be implemented with a
simple click of the mouse. Custom models can be created
using Excel (VBA).

TABLE 3
Pressure Specifications
Pump Pressure
Specified Rise Specified
(A P- kPa) Efficiency %
P1 82.33 85
P2 150 80
P3 150 75
P4 75 75
Heat Exchangers
Specified Drop Specified Drop
Tube (AP- kPa) Shell (AP-kPa)
HX1 30 1
HX2 30 20
HX3 30 20
Valves
Specified Drop
(A P- kPa)
V 1 (Mist Eliminator) 35
V2 to V9 68.94











TABLE 4
Expander {Exl}, Condenser {Hxl} and {Hx3}
Expander
APdrop kPa (specified) 1180
Energy MWe (specified) 60
Adiabatic Efficiency % (caculated) 82.45
Condenser (Hxl}
Tube side Temperature Rise (specified)[141 50
Heat Exchnger {Hx3}
Shell Side Temperature Rise (specified)E14] 96.9


POLYMATH Report
Nonlinear Equations

Calculated values of NLE variables
Variable Value f(x) Initial Guess
1 xl 2159.089 0 4338.9
2 X2 32.52249 0 32.52
3 x3 0.0052474 0 0.009
4 x4 2.552E+06 0 2.552E+06
5 x5 2.5524 2.193E-15 4.53
6 x6 5.431066 0 9.31
7 X7 2.55E+06 0 2.548E+06
8 x8 0.1275113 8.882E-16 0.13


Nonlinear equations
1 f(xl) = x4+x5+x6-TotS28 = 0
2 f(x3) = x5/(x4+x5+x6)-H2SinS28 = 0
3 f(x2) = x8/(x7+x8)-021nS26 = 0
4 f(x8) = C021nS4*x3-x6 = 0
5 f(x7) = WatinS4*x3/18.016+x7/18.016+x8/16-x4/18.16 = 0
6 f(x6) = H2SInS4*x3/34.06-x8/16-x5/34.06 = 0
7 f(x5) = 02inS24-x2-x8 = 0
8 f(x4) = WatinS24-x-x7 = 0


Explicit equations
1 WatinS24 = 2552384.23
2 H2SinS28 = 1E-6
3 TotS28 = 2552400.
4 O21nS26 = 50E-9
5 H2SinS4 = 538.14
6 021nS24 = 32.65
7 WatinS4 = 412914.84
8 CO21nS4 = 1035


Figure 3. Equations to determine deaerator performance.


Overall the system is very flexible and easy to learn and
use. It allows very rapid evaluation and optimization of dif-
ferent cases.

PLANT SIMULATION
VMGSim is used to simulate the plant and match the data
given in Figure 1. Steam Table is selected as the physical
property system. The components in the simulation are:
1. Water
2 Hydrogen Sulfide
3. Carbon Dioxide
4. Oxygen
5 Sulfur
Figure 2 (p. 134) is the VMGSim flow sheet
depiction of the process
There are two feed streams to the plant. In
the first, geothermal fluid {S1} contains water,
HS, and CO2. Table 1 (p. 135) gives the stream
composition based on the values of HS and CO,
(CO2 2500 ppm, HS 1300 ppm) in the high-
pressure steam {S2}.191 The VMGSim system
is used to determine (by iteration) the values
of the temperature and pressure of {S1} from
the composition of the high pressure steam, the
enthalpy and vapor fraction (=115/326) of {S1}
specified in Figure 1.


The cold water (Table 2, p. 135) at 1 atm and 5
C is assumed saturated with oxygen and carbon
dioxide. Values of Henry's Law constants (H) are
taken from Perry:E101
H(O2) = 29100 atm/mole fraction (air 20.94%
oxygen)
H(CO2) = 878 atm/mole fraction (air 0.0314%
carbon dioxide)

where partial pressure(atm) = H x (mole frac-
tion)

The pressure drops throughout the system are
generally not specified in Figure 1 (an exception
is the pressure drop across the tubine: 12 bara
- 0.2 bara). As a result, literature suggestionsE11
12] for pressure drops in the valves and heat ex-
changers and pressure rises in the pumps (arbi-
trary) are used as shown in Table 3 (p. 135). Exit
streams are assumed to be at about atmospheric
pressure. The mist eliminator is simulated as a
valve {V1}.
The steam turbine {Exl} is simulated by an
expander, and the electrical energy (\ We) is
specified as 60 MW. The condenser is simulated


Chemical Engineering Education


# xl = Mass Water in S25
# x2 = Mass Oxygen in S25
# x3 = Fraction of 84 to reactor
# x4 = Mass Water in S28
# x5 = Mass H2S in 828
# x6 = Mass C02 in S28
# x7 = Mass Water in S26
# x8 = Mass 02 in S26












as a heat exchanger ({Hxl}) and a separator that purges the
noncondensable gases (Table 4).

VMGSim does not have a model of a vacuum deaerator. 61
It is simulated, however, with a mixer {M4} and a Compo-
nent Splitter {CSP1} A determination is madeE131 to find the
amount of water and 02 that the deaerator is required to purge
in order to meet specifications on the exit water content of 02


{S26} and the H2S content of the water delivered to Reykjavik
{S28}. (All CO2 is assumed to go overhead)

Figure 3 is a set of eight equations (in eight unknowns),
but with just three manipulated variables to achieve three
specifications.

Values To Be Determined:
1. Fraction of high pressure steam that goes to the reac-


TABLE 5
Calculated Streams Compared to Reference Number 2
Iceland's Nesjavellir Co-Generation Power Plant
Stream Description Flow kg/s Temp Deg C Enthalpy kJ/kg Pressure kPa Frac
Ref Ref Ref Ref Vapor
[2] [2] [2] [2]
S1 Geothermal Fluid 326 326 189.2 1500 1500 1235 0.3527
S2 High Pressure Steam 115.13 115 189.2 2775.36 1235 1

S3 Geothermal Fluid 210.86 211 189.2 803.57 1235 0
S4 High Pressure Steam 115.13 188.2 188 2775.96 1200 1200 1
S5 Low Pressure Steam 114.53 115 60 2251.9 20 20 0.8519
S6 Condensate 112.62 56.3 60 235.74 19 0
S9 Warm Water 667 667 55 230.39 221.33 0
Sl1 Warm Water 667 667 86.4 88 361.96 122.89 0
S14 High Pressure Steam 0 0
S15 Geothermal Fluid 0 0
S20 Geothermal Fluid 323.48 326 79.9 81 334.65 101.33 0
S21 Cold Water 1129 1129 5 21.1 101.33 0
S28 Warm Water 709 709 81.7 83 342.19 118.45 0
S30 Cold Water 42 42 5 21.3 182.39 0
S35 Cold Water 1087 1087 5 21.29 251.33 0
S38 Warm Water 420 420 55 230.39 152.39 0
S54 Geothermal Fluid 219.86 92.3 92 387.41 1146.06 0

Turbine Output: Ref [2]= 60 MWe, Simulation = 60 MWe ; Thermal MWt: Ref [2] = 127 MWt, Simulation = 123.88 MWt
*Note: Numbers in bold are those specified in Figure 1

TABLE 6
Distribution of Noncondensable Gases
HS CO, 0,
In Out In Out In Out
Stream kg/h kg/h Stream kg/h kg/h Stream kg/h kg/h
S1 543.67 S1 51.99
S20 12.34 S1 1035.09 S25 32.52
S28 2.55 S21 3.56 S32 19.34
S39 528.51 S20 2.75 Reaction 0.13
Reaction 0.27 S25 2.24
S28 5.43
S38 1.32
S39 1026.91
Sum 543.67 543.67 Sum 1038.65 1038.65 Sum 51.99 51.99

Vol. 42, No. 3, Summer 2008 13;











tor in separator {SP4}
2. The amount of water purged in the deaerator {S25}
3. The amount 0, purged in the deaerator {S25}.

It is assumed all the CO2 will be purged.
Specifications
1. The ppb of O, in the liquid leaving the deaerator
{S26} 50 ppbE15
2. The ppm of H2S in the exit water {S28) 1 ppmE5'
3. The total flow of the exit water {S28} 709 kg/s
The results of the computation are used to enter the fractions
overhead into the Component Splitter Block in the VMGSim
simulation. The exit temperature of the deaerator (81.2 C ) is
determined by an enthalpy balance around the deaerator.
Thermal power (MWt) is calculated based on the flow of
heated water {S28} and its temperature above 40 C:
MWt = mass flow of heated water X heat capacity X (Out-
put Temperature 40)
= 123.88

Table 5 (p. 137) gives the results of the simulation. Numbers
in bold are those taken from Figure 1. Other values are results
of the simulation.

DISCUSSION OF THE SIMULATION
As shown in Table 5 the VMGSim simulation matches the
indicated conditionsE21 reasonably well. Two important fac-
tors, however, impact the comparison of the simulation and
the data of Figure 1.
1. The plant data of Figure 1 does not indicate any vent-
ing from the condenser {Sep2} or specify the amount
of high pressure steam in stream {S37}. The simula-
tion calculates both {S37} and {S39}.
2. The plant data of Figure 1 does not indicate any
venting from the deaerator. The deaerator vents both
water and noncondensable gases.
Small changes in the flow to the expander cause consider-
able changes in downstream streams {S5}, {S6}, and {S20}.
Similarly, small changes in the concentration of HS in the
heated water {S28} greatly affect the amount of water purged
in the deaerator.
The deaerator design is based on data suggested by an
authors5l other than Ballzus. [2
The distribution of the noncondensable gases was not ad-
dressed in Figure 1 but is discussed by Gislason.J141 Table 6
(p. 137) lists the distribution in this simulation. A comparison
with Gislason is difficult as he lumps the flows of HS and CO2
together and indicates different amounts of the noncondens-
able gases in the entering streams ( {S1} and {S21} ) than
used in this study. Also, Gislason does not account for 0,.


CONCLUSIONS


Study of Iceland's Nesjavellir Power Plant appears to be
well suited for classroom instruction and inclusion in under-
graduate energy courses.[15i Such a study illustrates both the
advantages of geothermal energy as well as indicating some
of its limitations in terms of the suitability and source of
geothermal fluids.

Carrying out a simulation draws attention to a variety of
energy tradeoff issues, material balance questions, physi-
cal property estimates, equipment design selection, water
chemistry, and environmental control. Interest in geothermal
energy generated by this study can be pursued by searching
(e.g., on the Internet) for other ways of using this source
of energy.[16]


ACKNOWLEDGMENT
The author would like to thank Gerald Jacobs of Virtual
Materials Group for making the VMGSim system available
to the author to conduct this study.


REFERENCES
1. Gunnlaugsson, E., A. Ragnarsson, and V. Stefannson, "Geothermal
Energy in Iceland," International Symposium in Izmir, Turkey 4-5-
(October 2001)
2. Ballzus, C., H. Frimannson, G. Gunnarsson, and I. Hrolfsson, "The
Geothermal Power Plant at Nesjavvellir, Iceland, "Proceedings World
Geothermal Congress 2000 Kyushu Tohoku, Japan, May 28 (June
10, 2000)
3. Reykjavik Energy, Nesjavellir Power Plant
4. "Geothermal Resources in Iceland," edu/is/reyk/resources.htm>
5. "How the Plant Works," works. htm>
6.
7. "Microscale Gas Chemistry: Experiments with Hydrogen Sulfide,"

8. Virtual Materials Group, Inc. Version 3.1.44 (January, 2008) www.virtualmaterials.com>
9. Gunnarsson, A., B.S. Steingrimsson, E. Gunnlaugsson, J. Magnusson,
and R. Maack, "Nesjavellir Geothermal Co-Generation Power Plant,"
Geothermics, 21(4), 559 (1992)
10. Perry, J.H., Chemical Engineers'Handbook, 3rd Ed., 673-674 (1950)
11. htm>
12.
13. <1.11 - I ".. 1li_software.com/>
14. Gislason, G., "Nesjavellir Co-Generation Plant, Iceland. Flow of
Geothermal Steam and Non-Condensable Gases, "Proceedings World
Geothermal Congress 2000 Kyushu Tohoku, Japan, May 28 (June
10, 2000)
15. Edgar, T., "A Course on Energy Technology and Policy," Chem. Eng.
Ed., 41(3), 195 (2007)
16. 1


Chemical Engineering Education











Random Thoughts...








HOW TO WRITE ANYTHING






RICHARD M. FIELDER
REBECCA BRENT

"I write when I'm inspired, and I see to it that I'm inspired at nine o'clock every morning."
(Peter De Vries)


Here's the situation. You're working on a big writing
project-a proposal, paper, book, dissertation, what-
ever-and in the last five weeks all you've managed
to get done is one measly paragraph. You're long past the
date when the project was supposed to be finished, and you
just looked at your to-do list and reminded yourself that this
is only one of several writing projects on your plate and you
haven't even started most of the others.
If you're frequently in that situation (and we've never met
a faculty member who isn't) we've got a remedy for you.
First, though, let's do some truth in advertising. Lots of books
and articles have been written about how to write clear and
persuasive papers, proposals, dissertations, lab reports, techni-
cal memos, love letters, and practically everything else you
might ever need to write. We're not going to talk about that
stuff: you're on your own when it comes to anything having
to do with writing quality. All we're going to try to do here is
help you get a complete draft in a reasonable period of time,
because that usually turns out to be the make-or-break step in
big writing projects. Unless you're a pathological perfectionist
(which can be a crippling obstacle to ever finishing an.i til ni.
once you've got a draft, there's an excellent chance that a
finished document suitable for public consumption won't be
far behind.
We have two suggestions for getting a major document writ-
ten in this lifetime: (1) commit to working on it regularly, and
(2) keep the creating and editing functions separate.*
* Dedicate short and frequent periods of time to your
major writing projects

We didn't invent either technique-you canfind variations ofboth in
many references on writing. A particularly good one is RobertBoice,
Professors as Writers, Stillwater, OK: New Forums Press, 1990.
Vol. 42, No. 3, Summer 2008


See if this little monologue sounds familiar. "I don't have
time to work on the proposal now-I've got to get Wednesday's
lecture ready and there's a ton of e-mail to answer and I've
got to pick the kids up after school tomorrow . BUT, as
soon as fall break (or Christmas, or summer, or my sabbati-
cal) comes, I'll get to it."
It's natural to give top priority to the tasks that can be done
quickly or are due soon, whether they're important (preparing
Wednesday's lecture) or not (answering most e-mails), and
so the longer-range projects keep getting put off as the weeks
and months and years go by. If a major project has a firm due
date, you panic when it approaches and quickly knock some-


Richard M. Felder is Hoechst Celanese
Professor Emeritus of Chemical Engineering
at North Carolina State University. He is co-
author of Elementary Principles of Chemical
Processes (Wiley, 2005) and numerous
articles on chemical process engineering
and engineering and science education,
and regularly presents workshops on ef-
fective college teaching at campuses and
conferences around the world. Many of his
publications can be seen at edu/felder-public>.
Rebecca Brent is an education consultant
specializing in faculty development for ef-
fective university teaching, classroom and
computer-based simulations in teacher
education, and K-12 staff development in
language arts and classroom management.
She codirects the ASEE National Effective
Teaching Institute and has published articles
on a variety of topics including writing in un-
dergraduate courses, cooperative learning,
public school reform, and effective university
teaching.


Copyright ChE Division of ASEE 2008










thing out well below the best you can do. If it's a proposal or
paper, subsequent rejection should not come as a surprise. If
there is no firm due date, the project simply never gets done:
the book you've been working on for the last 10 years never
gets into print, or your graduate students leave school with
their research completed but without their Ph.D.s because
they never finished their dissertations.
The strategy of waiting for large blocks of time to work on
major writing projects has two significant flaws. When you
finally get to a block, it's been so long since the last one that
it can take hours or days to build momentum again and you're
likely to run out of time before much gets written. Also, as
soon as the block arrives other things rush in to fill it, such as
your family, whom you've been neglecting for months and
who now legitimately think it's their turn.
A much more effective strategy is to make a commitment
to ,. .il. nl., devote short periods of time to major I iit,.
projects. Thirty minutes a day is plenty, or maybe an hour
three times a week. One approach is to designate a fixed time
period on specified days, preferably at a time of day when
you're at your peak, during which you close your door, ignore
your phone, and do nothing but work on the project. Alterna-
tively, you might take a few 10-15 minute breaks during the
day-times when you would ordinarily check your e-mail or
surf the Web or play Sudoku-and use them to work on the
project instead. Either way, when you start to write you'll
quickly remember where you left off last time and jump in
with little wasted motion. When you've put in your budgeted
time for the day, you can (and generally should) stop and go
back to the rest of your life.
These short writing interludes won't make much differ-
ence in how many fires you put out each day, but you'll be
astounded when you look back after a week or two and see
how much you've gotten done on the project-and when a
larger block of time opens up, you'll be able to use it effec-
tively with very little warm-up. You can then be confident of
finishing the project in a reasonable time . provided that
you also take our next suggestion.
* Do your creating and editing sequentially, not simul-
taneously
Here's another common scenario that might ring a bell.
You sit down to write ..,,,. d,.il, and come up with the first
sentence. You look at it, change some words, add a phrase,
rewrite it three or four times, put in a comma here, take one
out there . and beat on the sentence for five minutes and
finally get it where you want it. Then you draft the second
sentence, and thefirst one is instantly obsolete and you have
to rewrite it again ... and you work on those two sentences


until you're satisfied with them and go on to Sentence 3 and
repeat the process ... and an hour or two later you may have
a paragraph to showfor your c t -: i,
If that sounds like your process, it's little wonder that you
can't seem to get those large writing projects finished. When
you spend hours on every paragraph, the 25-page proposal
or 350-page dissertation can take forever, and you're likely
to become frustrated and quit before you're even close to a
first draft.
At this point you're ready for our second tip, which is to
keep the i. i,. and editing processes separate. The routine
we just described does the opposite: Even before you complete
a sentence you start criticizing and trying to fix it. Instead of
doing that, write whatever comes into your head, without
looking back. If you have trouble getting a session started,
write ,i,: il,, o. -random words, if necessary-and after a
minute or two things will start flowing. If you like working
from outlines, start with an outline; if the project is not huge
like a book or dissertation and you don't like outlines, just
plunge in. If you're not sure how to begin a project, start with
a middle section you can write easily and go back and fill in
the introduction later.
Throughout this process you will, of course, hear the usual
voice in your head telling you that what you're writing is pure
garbage-sloppy, confusing, trivial, etc. Ignore it! Write the
first paragraph, then the next, and keep going until you get
as much written as your budgeted time allows. Then, when
you come back to the project the next day (remember, you
committed to it), you can either continue writing or go back
and edit what you've already got-and then (and only then)
is the time to worry about grammar and syntax and style and
all that.
Here's what will almost certainly happen if you follow that
procedure. The first few sentences you write in a session may
indeed be garbage, but the rest will invariably be much bet-
ter than you thought while you were writing it. You'll crank
out a lot of material in a short time, and you'll find that it's
much easier and faster to edit it all at once rather than in tiny
increments. The bottom line is that you'll find yourself with a
completed manuscript in a small fraction of the time it would
take with one-sentence-at-a-time editing.
We're not suggesting that working a little on big projects
every day is easy. It isn't for most people, and days will
inevitably come when the pressure to work only on urgent
tasks is overwhelming. When it happens, just do what you
have to do without beating yourself up about it and resume
your commitment the next day. It may be tough but it's do-
able, and it works. 7


Chemical Engineering Education


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











curriculum
-0


INCORPORATING RISK ASSESSMENT AND


INHERENTLY SAFER DESIGN PRACTICES

into Chemical Engineering Education







JEFFREY R. SEAY
University of Kentucky Paducah, KY 42002
MARIO R. EDEN
Auburn University Auburn, AL 36849


Process safety is a fundamental component of sound
process design. Although the chemical industry has
demonstrated an excellent safety record over the
years,"11 the quantities and hazardous nature of many of the
substances typically handled by chemical manufacturers
make the potential for large-scale disasters a constant con-
cern. Because safety is so critical in industry, it is vital to
introduce the concept of safe process design practices during
undergraduate chemical engineering education. From famous
historic disasters such as Flixborough and Bhopal to recent
events such as the Texas City BP Refinery explosion in 2005,
the importance of process safety in chemical process design
is abundantly clear. An appreciation of this gained during a
chemical engineer's education can only enhance chemical
manufacturing safety in the future.
In industry, the concept of process safety is firmly rooted
in the concept of risk. From government regulatory require-
ments, such as those outlined by OSHA and the EPA,[2 4] to
industry initiatives such as Responsible Care, the requirement
of quantifying and managing risk is paramount. In addition
to working within economic and environmental constraints,
the process design engineer is also tasked with reducing the
risk of operating a chemical manufacturing process to an
acceptable level for employees, regulatory authorities, insur-
ance underwriters, and the community at large. Therefore, a


holistic approach to process safety as an integral component
of sound process design is critical.
In addition to the study of toxicological impacts and quan-
tifying release scenarios, an understanding of how risk is
quantified in the chemical process industries will allow future
process design engineers to mitigate those risks at the earliest

Jeffrey Seayis an assistant professorin the De-
partment of Chemical and Materials Engineer-
ing at the University of Kentucky. He recently
moved to academia after 12 years in industry.
He holds a B.S. (1996) and Ph.D. (2008) from
Auburn University, and an M.S. (2004) from
the University of South Alabama, all in chemi-
calengineering. In addition to his professional
experience as a process design engineer, he
is a trained PHA Team Leader with extensive
experience in risk assessment methodology
and layer-of-protection analysis.
Mario Eden is presently an assistant profes-
sorin the Department of Chemical Engineer-
ing at Auburn University. He received his
M.S. (1999) and Ph.D. (2003) degrees from
the Technical University of Denmark. Dr.
Eden's work seeks to advance the state of
the art in process systems engineering (PSE)
research and education through innovative
and novel systematic methodologies for
integrated process and product design.
Dr. Eden is a recipient of the NSF CAREER
award (2006).


Copyright ChE Division of ASEE 2008


Vol. 42, No. 3, Summer 2008











stage of conceptual process development-the stage where
an engineer has the greatest influence on the final process
design. This paper will present, by case-study example, how
the fundamental concepts of inherently safer process design
can be integrated into chemical engineering education.

RISK ASSESSMENT METHODOLOGY
Quantifying Risk
In order to begin understanding the benefits of inherently
safer process design, the chemical engineering student must
first understand risk. The concept of risk is often misunder-
stood by both the general public and students of chemical
engineering. It is important to separate the concept of risk
from the concept of hazard. While the concept of hazard re-
lates to the potential for adverse consequences, risk is rather
a combination of both the severity of the consequences of an
upset scenario and the likelihood of that scenario's initiating


cause. This is an important distinction. The potential hazard
associated with a substance or process is an inherent property
that cannot be changed. The risk associated with handling a
substance or operating a process can be high or low, depending
upon the safeguards included in the design. Thus, for chemical
engineers, the most important distinction between hazard and
risk is that risk can be reduced through process design.

In order to begin to discuss risk, the process design engi-
neer must first consider potential upset scenarios. In other
words, answer the question, "What is the worst thing that can
happen?" Answers to this question typically involve loss of
containment of a process chemical with causes ranging from
failure of control loops and operator errors to external events
such as fire, among many others. It is critical to note that the
answers to the aforementioned question must be considered
independently of the likelihood of the worst-case scenario oc-
curring. Again, it is the combination of both the severity and
the likelihood that deter-


TABLE 1
Example of a hazard scenario using "What If...?" methodology
WhatIf...? Initiating Cause Consequence Safeguards
1. There is High 1.1 Failure of the 1.1 Potential for pressure in 1. Conservation vent
Pressure in the pressure regulator tank to rise due to influx of sized to relieve over-
Cyclohexane on nitrogen supply nitrogen pad gas through failed pressure due to this
Storage Tank? line to Cyclohexane regulator. Potential to exceed scenario.
Storage Tank. design pressure of storage tank. 2. Pressure transmitter
Potential tank leak or rupture with high alarm set to
leading to spill of a flammable indicate high pressure
liquid. Potential fire should an in Cyclohexane Storage
ignition source be present. Po- Tank.
tential personnel injury should
exposure occur. 2.1 Potential
environmental release requir-
ing reporting and remediation.

TABLE 2
Inherently Safer Design Choices for Common Design Applications
Hazard Scenario Process Operation Potential Upset Case Inherently Safer
Design
Overpressure Filling a process vessel Overpressure by pump 1. Vessel design pres-
with a pump. deadhead due to overfill, sure greater than pump
deadhead pressure
2. Static head due to
vessel elevation plus
vessel design pres-
sure greater than pump
deadhead pressure.
Overpressure Operating a vessel un- Failure of inlet gas regula- 1.Vessel design pressure
der inert gas pressure. tor leading to overpressure. greater than inert gas
supply pressure.
Underpresure Emptying a process Blocked vent leading to 1.Vessel designed for
vessel with a pump. vessel collapse due to full vacuum
vacuum pulled during pump
out.
Underpressure Draining an elevated Blocked vent leading to 1.Vessel designed for
process vessel by vessel collapse due to full vacuum.
gravity. vacuum pulled during 2. Liquid drain lined
draining, sized to be self-venting.


mines the risk. In order
to ensure a complete
and consistent assess-
ment of potential upset
scenarios, a structured
approach must be ap-
plied. The need for such
an approach is the basis
for a Process Hazard
Analysis.
Process Hazard
Analysis
A Process Hazard
Analysis (PHA) is a
methodology for review-
ing and assessing the
potential hazards of a
chemical process by us-
ing a structured, facili-
tated, team brainstorm-
ing approach. A PHA is
typically facilitated by
a trained team leader
and attended by a wide
variety of plant person-
nel, including engineers,
managers, operators,
maintenance technicians
and safety, health, and
environmental (SHE)
personnel. Although
several techniques are
available for performing
PHAs,3s the goal of the
PHA is always the same


Chemical Engineering Education











-to identify the potential hazards of a process and determine
whether sufficient safeguards are in place to mitigate those
hazards.

CLASSROOM EXAMPLE OF APPLYING
PHA METHODOLOGY
The following is a simple example that can be used to illus-
trate the basic concepts of a PHA in the chemical engineering
classroom. Consider a low design pressure API storage tank
filled with cyclohexane. API type storage tanks are typically
designed for no more than 2.5 pounds of pressure and only a
few inches of water of vacuum. Therefore, careful control of
pressure is critical. Furthermore, assume that the storage tank
is equipped with a "pad/de-pad" vent system to control pres-
sure, and is located in a diked tank farm. Table 1 illustrates a
typical scenario that might be developed during a PHA using
the "What If... ?" methodology.
In Table 1, the listed safeguards would be effective means of
mitigating the personnel exposure and environmental impact
consequences identified for this scenario. In addition to the
cause illustrated, other causes of high pressure that might be
considered by a PHA Team include the following:
External fire in the area, leading to increased vapor
pressure in the storage tank.
Overfill via the supply pump, leading to overpressure
by deadhead pump pressure.


If the safeguards iden-
tified by the PHA team
are not deemed adequate,
recommendations are
made for the implemen-
tation of additional safe-
guards. This technique,
called Layer of Protec-
tion Analysis (LOPA), is
often employed by PHA
teams to quantitatively
assess the risk associated
with an upset scenario so
that appropriate layers of
protection can be applied
to adequately mitigate the
risk. [5
Hazard assessment and
layer of protection analy-
sis are complex subjects.
As such, a formal hazard
analysis is typically not
performed during the con-
ceptual phase of process
design. In most cases, the
PHA is performed during

Vol. 42, No. 3, Summer 2008


the engineering phases of a project. A basic understanding of
the fundamentals of risk assessment, however, is extremely
beneficial to the development of inherently safer designs
during the conceptual phase of process design. To make inher-
ently safer design choices during conceptual development of
a process, the design engineering student must be aware of
the types of hazard scenarios that may be identified for each
piece of equipment or system.
Inherently Safer Process Design
Inherently safe process design practices can generally be
grouped into five categories: 6 71
Intensification
Substitution
Attenuation
Limitation of effects
Simplification

Some examples of inherently safer design choices for typi-
cal process applications are included in Table 2.
Typically, however, these types of design choices are made
in later stages of engineering development. Although these
are important design considerations, it is very beneficial to
begin evaluating inherently safer design strategies at the earli-
est stages of process development, when the process design
engineer has the greatest opportunity to affect the safety


TABLE 3
Potential opportunities for making inherently safer design choices[61
Process Design Inherently Safe Potential Process Safety Impact
Choice Design Category
Reactor type Intensification Continuous reactors are typically smaller than batch reac-
tors for a given production volume.
Feed stocks Substitution Less hazardous raw materials may be available to make
the same products.
Process solvents Substitution Less hazardous and/or less volatile solvents may be
available.
Reaction mechanism Attenuation Endothermic reactions present less potential for runaway.

Operating conditions Attenuation Temperatures and pressure close to ambient are typically
less hazardous.
Process utilities Attenuation Low pressure utilities such as hot oil may be a safer
choice than high pressure steam.
Alternative technology Attenuation Use of alternative technology, for example pervaporation
instead of azeotropic distillation using a solvent entrainer.
Production rate Limitation of effects A continuous process making just what is required can be
safer that a batch process with a large hold-up volume.
Storage volume Limitation of effects Minimization of volume limits the potential effects of a
release.
Equipment layout Simplification Utilizing gravity flow minimizes the need for rotating
equipment.
Cooling by natural Simplification Utilizing natural convection simplifies the process and
convection eliminates the potential for process upsets due to loss of
utilities.
































Figure 1. Flow diagram of traditional solvent recovery pro


aspects of the process. Some examples of design choices that
are typically made at the onset of conceptual engineering are
illustrated in Table 3 (previous page).
Initially, inherently safer designs may seem to be more
expensive than applying traditional safeguards to processes.
When the total cost of the process is considered, however, the
inherently safer design is often more cost effective. Installing
and maintaining multiple independent layers of protection can
be quite expensive, but these costs are often ignored during
initial cost estimates. Conceptual phase cost estimates are
usually based on stand-alone major equipment costs that are
simply multiplied by factors to obtain the total installed cost.
These factors are intended to account for instrumentation
and controls, among other items needed for the complete
process installation. To apply the same factors to traditional
and inherently safer processes, however, can lead to an er-
roneous comparison and conclusion. Inherently
safer processes will typically require fewer safety
controls, which leads to lower installation and op-
erating costs. These factors should be considered
when evaluating processes during a hierarchical
approach to process design. Additional cost sav-
ings for inherently safer processes that are often
overlooked include insurance costs and costs
associated with regulatory compliance.
Case Study-Solvent Recovery
The following case study is presented as a
classroom engineering design problem to illus- teams:
trate the techniques of applying inherently safer 1 solvee Feed
2 Azedrope
design choices. 3 waste water
4 Solvent Rich I
Consider a chemical process using 1-propanol 5Water Rich P1
as a solvent. Currently, the waste solvent ends 6Recoere s
up as a waste-water stream for disposal. The task Figure

144


for the process design engineer is to develop a
process to recover the 1-propanol from the waste-
water stream. This separation is complicated by
the fact that water and 1-propanol form a mini-
mum-boiling azeotrope. Therefore, separation by
ordinary distillation is not possible.
Traditional Process
The traditional method employed for breaking
this azeotrope uses a third solvent, or entrainer.
For the water/1-propanol system, cyclohexane
works well for the separation. A sample flow
diagram of the azeotropic distillation process is
given in Figure 1.
In this process, the minimum-boiling azeotrope
is separated from the water in the Azeotrope Col-
Sunmn and is collected as an overhead product. The
azeotrope is then mixed with the cyclohexane in
cess. the Entrainer Vessel. The 1-propanol is soluble
in cyclohexane, while the water is not. The water
phase, with a small amount of 1-propanol, is then recycled
back to the Azeotrope Column, while the cyclohexane 1-pro-
panol mixture is fed to the Solvent Column, where 1-propanol
is recovered as a bottoms product and the cyclohexane--with
a small amount of 1-propanol-is recycled to the Entrainer
Vessel. This simple system is easily modeled using any pro-
cess simulation software package.
Potential Upset Scenarios
From a process design perspective, this process is certainly
acceptable. From the perspective of safety, however, some
significant concerns arise. In order to break the azeotrope, a
highly volatile solvent, cyclohexane, is introduced to the pro-
cess. A sample of some of the potential hazard scenarios that
might be generated during a PHA is illustrated in Table 4.
Some potential safeguards that might be used to mitigate
these hazards include safety relief valves, redundant in-


2. Flow diagram of inherently safer solvent recovery process.


Chemical Engineering Education











strumentation, and hardwired
interlocks independent from the
primary basic process control
system (BPCS). All of these
safeguards would be applied to
the process during later stages of
process design, as considering
inherently safer design choices
could make such safeguards
unnecessary.
An Inherently Safer
Process
An inherently safer approach
to this design problem will in-
clude technology to break the


azeotrope without introducing
additional, flammable solvents
to the process. One possible
solution is the use of a pervapo-
ration membrane. A pervaporation membrane separates two
liquids by partial vaporization through a nonporous mem-
brane, such as ceramic. The pervaporation membrane is able
to break azeotropes due to its ability to separate components
based on polarity differences between the molecules, rather
than relying on differences in vapor pressure, like distilla-
tion does.
Although the pervaporation technology could be used to
completely separate 1-propanol from the water in one step,
such a sharp split would most likely prove to be prohibitively
expensive. An optimum design using a combination of dis-
tillation and pervaporation can be achieved, as illustrated in
Figure 2.
In this design, the azeotrope is again separated from the
water as an overhead product in the Azeotrope Column, but in
this process, instead of using an entrainer, the Pervaporization
Unit is used to separate the liquids. Because this technology
is used in conjunction with distillation, a sharp split is not
needed. The water-rich phase leaving the Pervaporation Unit
is returned to the Azeotrope Column, and the 1-propanol is
recovered as a bottoms product from the Solvent Column,
with the azeotrope being collected overhead and returned to


overwhelm incinerator leading to possible
explosion. Potential personnel injury due
to exposure.


the Azeotrope column. This design is advantageous because
it can be optimized to minimize the impact of the cost of the
Pervaporization Unit.
Process Safety Improvements
The inherently safer design has the obvious advantage of
having eliminated the flammable solvent, cyclohexane, from
the process. Taking a wider view, not only is the cyclohexane
eliminated from the process itself, but also from storage ar-
eas, unloading areas, and waste treatment. In addition to the
benefit of eliminating a flammable solvent from the process,
the Pervaporation Unit minimizes the circulating flow of
material through the system. Therefore, the column and as-
sociated heat exchangers are smaller than with the traditional
process design. Based on the hazard scenarios identified for
the traditional process, illustrated in Table 4, the benefits of
the inherently safer process are illustrated in Table 5.
From this assessment, the benefits of the inherently safer
process are clear. The pervaporation process addresses three of
the five categories of inherently safer design choices: Attenu-
ation, Simplification, and Limitation of Effects. Attenuation
is due to the use of alternative technology, Simplification is
due to the elimination of the entrainment solvent from the


TABLE 5
Summary of Process Safety Implications of Design Choices for Case Study
Example.
Upset Scenario Traditional Process Inherently Safer Process
External Fire Large volume of flammable Flammable volume limited
liquid circulating in process. to recovered solvent only.
Overfill Cyclohexane entrainer more Minimal liquid hold up in
volatile than 1-propanol. Pervaporation Unit.
Overpressure Larger liquid hold-up leads to Volume limited to solvent
higher severity in the event of a distillation hold-up.
release.

Vol. 42, No. 3, Summer 2008


process, and Limitation of Effects is due
to the smaller equipment and chemical
inventories. Of course, 1-propanol is a
flammable liquid, so all of the upset sce-
narios listed in Table 4 would still need
to be considered, but by eliminating the
cyclohexane from the process, the overall
severity of the consequences would be
reduced. Since, as discussed previously,
risk is a combination of both severity and
likelihood, the overall risk of the inher-


TABLE 4
Potential Hazard Scenarios for Traditional Azeotropic Distillation Process.
What If...? Initiating Cause Consequence
1.There is higher pres- 1.1 External fire in the 1.1 Potential increased temperature and
sure in the Entrainment process area. pressure leading to possible vessel leak or
Vessel? rupture. Potential release of flammable
material to the atmosphere. Potential
personnel injury due to exposure.
1.2 Pressure regulator 1.2 Potential for vessel pressure to increase
for inert gas pad fails up to the inert gas supply pressure. Poten-
open. tial vessel leak or rupture leading to release
of flammable material to the atmosphere.
Potential personnel injury due to exposure.
2. There is higher level 2.1 Vessel level trans- 2.1 Potential to overfill vessel with cyclo-
in the Entrainer Vessel? mitter fails and indicates hexane. Potential to flood vent line with
lower than actual liquid leading to flammable liquid reaching
volume, the vent gas incinerator. Potential to











ently safer design would be reduced. Although a more in-depth
study would be required before making the choice of which
solvent recovery process is preferred, it should be clear that
these decisions must be made in the early stages of conceptual
process development in order to benefit the process.

CONCLUSIONS
One of the responsibilities of every chemical engineer is to
ensure that the excellent safety record enjoyed by the chemical
process industry is maintained. Therefore it is important to
begin introducing the fundamentals of process safety during
undergraduate chemical engineering education. The purpose
of this work has been to underscore, by case-study example,
the natural relationship between inherently safe process design
and conceptual process development, and describe how it can
be integrated into undergraduate process design education.
As has been illustrated by this case study, taking a holistic
approach to process safety education can serve to reinforce
the benefits of beginning to consider the safety implications of
the decisions made during conceptual process development.
By reinforcing the benefits of making inherently safe design
choices during conceptual process development, students


of process engineering will be better prepared for the chal-
lenges of meeting the high standards of safety set by today's
chemical industry.

ACKNOWLEDGMENTS
The authors would like to acknowledge Felicia Foster and
Robert D'Alessandro of Evonik Degussa Corporation for
providing valuable insight and guidance on the industrial
applications of process safety.

REFERENCES
1. Sanders, R., Chemical Process Safety -Learningfrom Case Histories,
3rd Ed., Elsevier, Inc., (2005)
2. Nelson, D., Managing Chemical Safety, Government Institutes,
(2003)
3. Environmental Protection Agency, "Process HazardAnalysis, "40 CFR
68.67 (2005)
4. Occupational Safety and Health Administration, i ..- .. I ~ i Man-
agement of Highly Hazardous Chemicals," 29 CFR 1910.119 (2005)
5. Center for Chemical Process Safety, Layer of Protection Analysis
Simplified Process Risk Assessment, AIChE (2001)
6. Kletz, T., Process Plants: A Handbook for Inherently Safety Design,
Taylor and Francis (1998)
7. Center for Chemical Process Safety, Guidelinesfor Engineering Design
for Process Safety, AIChE, 1993. 1


Chemical Engineering Education











[L] = laboratory










A LAB EXPERIMENT TO INTRODUCE

GAS/LIQUID SOLUBILITY








I.M.A. FONSECA AND J.P.B. ALMEIDA
Universidade de Coimbra 3030-290 Coimbra, Portugal
H.C. FACHADA
Institute Politicnico de Coimbra 3030-199 Coimbra, Portugal
T he concept of the solubility of a gas in a liquid is
familiar in many everyday ways. When we drink a Isabel M.A. Fonseca is an associate profes-
sor in the chemical engineering department,
bottle of carbonated beverage, we understand it con- Universidade de Coimbra, Portugal. She re-
tains dissolved carbon dioxide, and we know animals can live ceived her M.S. in 1985 and her Ph.D. in 1991
From Universidade de Coimbra. Her academic
in oceans and rivers because oxygen dissolves in water and research has been developed in the experi-
animal life depends on the dissolution of oxygen in blood. For mental thermodynamic domain, particularly
chemical engineers, this everyday phenomenon is important VLE at low temperatures, determination of
thermodynamic and transport properties, and
from both practical and theoretical points of view.[1, 2] From treatment of experimental data.
a practical point of view, this concept is used in the design tre t of el
of absorption columns where a gaseous mixture is separated
by contact with suitable solvents that dissolve its components Horhcio C. Fachada is a professor adjunto
in the electrical engineering department of
differently. Further, knowledge of gas solubility in water is Instituto Superior de Engenharia de Coimbra,
important in the processes that control environmental distri- Portugal. He received his M.S. in 1991 from
Universidade de Coimbra. His research
bution of contaminants, such as halogenated hydrocarbons. interests are in the areas of control, mobile
From the theoretical point of view, the solubility of gases robotics, and vision navigation.
in liquids is an excellent tool to investigate solute-solvent
intermolecular forces in the liquid state, since solute-solute
interactions are almost negligible.
In this paper we present a simplified version recently devel- Jos6 P.B. Almeida received his B.S. in
oped by Fonseca, et al,[3] for experimental determination of chemistry in 2005 from the chemical depart-
ment of Universidade de Coimbra, Portugal.
the solubility of a gas in a liquid. This experiment is imple- Presently he is developing a research work
mented in the chemical engineering department of Coimbra, in the area of gas/liquid solubility of hydro-
fluorocarbons (HFCs) supported by FCT,
and is a part of one of the third-year laboratory courses in Fundagao para a Cidncia e Tecnologia
phase equilibria domain. The experiments are carried out by (Portugal), and FEDER.
groups of a maximum of three students during approximately
four hours. The full report of the experiment is completed
by the group at home and must be presented to the teacher Copyight ChEDision of ASEE 2008

Vol. 42, No. 3, Summer 2008 14










the following week and discussed one week later. The report
consists of the following sections: introduction, experiment
objectives, background of the experiment, description of the
apparatus, experimental procedure, and a discussion/results
section. The analysis of uncertainties must also be done using
the propagation law of errors, and the relative contribution of
each measured variable uncertain in the final result- i.e., the
solubility-evaluated. The students must also include in the
report a section of conclusions, where they can also present
comments and recommendations for improvements.

THEORY
When a gas is in contact with a liquid it tends to dissolve
in the liquid and the liquid evaporates until equilibrium is
reached. Consider a binary mixture of components 1 and 2
at temperature T and pressure P, represented schematically in
Figure 1. The component 2 is near or above its critical tem-
perature, which means this component is a gas at temperature
T. (A pure fluid above its critical temperature is called a gas,
and a vapor below To). When the liquid-phase mole fraction,
x2, is small and the equilibrium vapor-phase mole fraction,
y2, is large (near unity), it is conventional to call species 2
a "dissolved gas," and to label the physical situation one of
"gas solubility."[4] Gas/liquid solubility is a particular case of
vapor/liquid equilibrium; therefore, the classical treatment of
this subject is similar in some important aspects.
Let's consider species 1 the major component of the liquid
phase (the solvent) and species 2 the dissolved gas (the solute).
The liquid phase mole fraction, x2, represents the solubility
of gaseous solute 2 in liquid solvent 1. For component 2, the
thermodynamic condition of phase equilibrium states: 5]
fG fL
f =f (1)

where f and fL are the fugacities of component 2 in the
vapor and liquid phases, respectively. f~ is given by the
expression:

fG= .y2P, (2)
where (2 represents the fugacity coefficient of component 2
in the vapor phase and P is the total pressure. fL is repre-
sented by:


P0____
G Y1,Y2


L
XI.XI


Figure 1. Schematic representation
of gas/liquid solubility.


f= 2X2f (3)
where 72 is the activity coefficient of component 2 in the liquid
and fo the standard-state fugacity of the same component,
which is usually assumed to be the fugacity of the pure liquid
at the temperature and pressure of equilibrium (f0 = f,L
The substitution of Eqs. (2) and (3) in Eq. (1) gives:

Y22P = x2fL. (4)
The above equation can be simplified if we consider the
following assumptions: [5]
(i) 2 = 1 (y = 1), i.e., the liquid phase is an ideal solution;
(ii) (2 = 1 (p2 1), i.e., the vapor phase is an ideal mixture;

(iii) f2,L = P2 (P2 is the pure vapor pressure of compo-
nent 2 at temperature T) i.e., the effect of pressure on the
fugacity of the pure liquid phase is negligible (Poynting
correction =1) at moderate pressures; and p*2 = 1, (p*2 is the
fugacity coefficient of pure component 2) which is valid if
P"2 is low at temperature T.
Therefore Eq. (4) reduces to Raoult's law:
P2 = 2P = x2P2, (5)
where P2 represents the partial pressure of component 2.
The solubility x2, as given by Eq. (5), is called the "ideal"
solubility of the gas. This expression states that the solubility
is independent of the solvent and for a given gas, at a constant
partial pressure, the solubility always decreases with rising
temperature, which is not always true. Because of these dis-
advantages, the ideal solubility expression usually gives no
more than a rough estimate of gas solubility.
The solubility of gas in a liquid is often proportional to its
fugacity in the vapor phase. This situation is described by a
more realistic expression, Henry's law:

L H2,1X2, (6)
where H2,1 is a constant that for a given solute and solvent
depends only on temperature.
Again the assumptions leading to Eq. (6) can be readily
recognized by comparing it with Eq. (4). The more important
thing is to consider the constancy of the activity coefficients
(independent of the composition), which means that Henry's
law is only valid at infinite dilution. This can be stressed writ-
ing Eq. (6) in more convenient form:
fL
H 2,1 = lim (7)
x2- O X2

which is equivalent to

y2P
H 2,1 = lim (8)
x20 X2
taking into account Eqs. (1) and (2).


Chemical Engineering Education


T










EXPERIMENTAL SECTION APPARATUS
The determination of the solubility of gas in a liquid is made
using a volumetric method.J61 The principle of this procedure
is to bring a measured volume of liquid into contact with a
known volume of gas at a given temperature and pressure.
After equilibrium has been attained, the change in the gas
volume yields the amount of gas dissolved in the liquid
and hence the solubility. The solubility apparatus is shown
schematically in Figure 2. The apparatus is housed in a water
thermostat where the temperature is maintained constant using
a temperature controller, TC. The temperature is measured
with a precision thermometer graduated in 0.01 C. The level
of the thermostat bath can be adjusted using an elevator, E,
in order to immerse the whole vacuum line.
The main features of the apparatus are: EQ, equilibrium
vessel, (-50 cm3) where the dissolution of the gas takes place;
GB, gas burette, which consists in a piston-cylinder arrange-
ment; PT, pressure transducer, for total pressure readings; LA,
linear actuator, which moves the piston in the gas burette; and
PC, pressure controller.

PROCEDURE
The experimental procedure begins with the evacuation of
the whole apparatus. After this two fundamental steps in any
gas solubility measurement need to be performed:
(i) degassing of the solvent
(ii) dissolution of the gas
To accomplish step (i), the equilibrium vessel is removed
from the line, lowering the level of the thermostat bath, and
filled with a known amount (~ 6 g) of solvent. After this,
it is again connected to the line and the position of the
thermostat bath re-established. The stopcocks V2 and
V4 are opened to degas the solvent in EQ during about
10 min. The magnetic stirrer must be on. This degassing
procedure should be repeated two or three times until the
measured pressure equals the solvent vapor pressure at
the equilibrium temperature. Then EQ must be removed
from the line in order to be weighed. After this procedure,
EQ is connected again to the vacuum line.
Step (ii) begins with slowly opening stopcock V 1, with
V2 opened (V3 and V4 must be closed) to admit the gas
to the equilibrium vessel. The total pressure is adjusted
to ca 1 atm and, after this, stopcock V1 is closed and V3
is opened. This pressure acts as a reference value for the
pressure controller, which commands the linear actuator,
LA. As the gas dissolves the pressure decreases and this is
detected by PT. The linear actuator, LA, drives the piston
down the cylinder to maintain the pressure constant at the F
reference value. The number of encoder pulses is counted pe
and displayed, and a conversion is made to determine the P
volume of gas displaced from the precision-bore tube that G


comprises that cylinder of the burette. This volume represents
the volume of the gas dissolved. The detailed experimental
procedure is described in Appendix A.

ADDITIONAL INFORMATION FOR
INSTRUCTORS
In this experiment it is important to use gases, such as CO,,
N2O or CH3F, that are highly soluble in water. Methyl fluoride
has the disadvantage of being very expensive. Less soluble
gases (0,, N,, etc.) should not be used in this experiment since
the time of dissolution increases substantially. For these kinds
of gases an equilibrium vessel with a greater volume should
be used. 31 Other solvents such as primary alcohols (e.g.,
methanol, ethanol, propan- -ol, butan- -ol) can also be used.
They must have purities greater than 99.8 percent.
The accuracy of the method can be improved using lecture
bottles of gases, which have higher purity (>99.5%) than
other commercial gases. A pressure reducer should be used
connected to the lecture bottle. The lecture bottles are quite ap-
propriate to classroom experiments and must be used in these
experiments. An adequate pressure reducer must be coupled
to the lecture bottle. In the present experiments we have used
a pressure reducer HSB 280 5 (from PRAXAIR), but
other reducers can be adapted depending on the mark of the
lecture bottle. This allows a safe use of the lecture bottles.
The linear actuator, LA, consists of a permanent magnetic
DC motor, which drives a worm screw coupled to an optical
encoder. The worm screw moves the piston inside the cylinder
of the gas burette and the displacement is proportional to the
number of encoder pulses (np). The proportionality constant


igure 2. Solubility apparatus: TB, thermostated bath; TC, tem-
rature controller; PT pressure transducer; LA, linear actuator;
C, pressure controller; EQ, equilibrium vessel with connector;
,B, gas burette; V1...V4, high vacuum Teflon stopcocks; AGIT
magnetic stirrer; E, elevator.


Vol. 42, No. 3, Summer 2008











The students must also include in the report a section

of conclusions, where they can also present comments

and recommendations for improvements.


indicated in the manual of the linear actuator should be given
to the students. In the present apparatus, the displacement is
obtained from: Ah/(mm) = 2.20189 x 104 n In this experi-
ment we have used a pressure transducer from Honeywell,
model PPT0015AWN2VA-A.
The thermostat bath can be built using Perspex or glass (71
X 26 X 38 cm) where a thermostatic control unit is immersed.
The whole setup costs about 1,800 euros.

CALCULATIONS
Solubility Calculation
To simplify the treatment of the raw data we consider some
assumptions: neglect the volume change of the liquid sample
during saturation and ideal solution behavior.
The raw data obtained from experiment are: np, the number
of encoder pulses from the linear actuator, the equilibrium
temperature, T, the reference pressure, P, and mi, the mass of
solvent. The number of encoder pulses is converted in Ah, the
displacement of the piston, using a conversion factor indicated
in the user manual of the pressure controller. The change of
the volume of gas in the gas burette due to gas dissolution is
obtained from:
AV = 7r2 h, (9)
where r represents the internal radius of the gas burette. The
quantity of the gas absorbed in the liquid, n2 (in moles), can
be obtained from AV using the equation,
PVGmix /(RT) = 1+ BmRT / P, (10)
where VGix represents the molar volume in the vapor phase
and Bmx the second virial coefficient of the binary mixture,
given by the expression,
B.ix = yB11 +yB22 B 2yy2B12 (11)
where B,, B22, and B12 are the second virial coefficients of
pure components 1 and 2, and the second cross coefficient,
respectively.
Substituting, in Eq. (10), VGmx = AV / nG, where nG repre
sents the total number of moles contained in the V volume,
one obtains,
AV/nG = (RT / P)+ B.,x (12)
which multiplied by (1/y,) gives,
n2 =yPV /(RT + BP). (13)


The value of the solubility, x,, is then obtained from:
x,=n,/(n,+n,), (14)
where n1 and n2 represent the amount of solvent and solute
in moles in the liquid phase, respectively. The n1 is obtained
directly from n1 = m1 / M1. Since we need to know y2 to obtain
n, from Eq. (13), this calculation requires an iterative proce-
dure. The calculations begin with estimates of the vapour and
liquid phases obtained from Raoult and Dalton laws. In the
following iteractions, these compositions are improved using
Eqs. (13) and (14) and the following expression:

Y2 = 1-(1- ) Px (15)
P (p,
which results from the thermodynamic condition of phase
equilibrium written for the solvent.
The p, is the fugacity coefficient of solvent in the vapor
phase, which is given by,

P =exp (P/RT) Bl +y (2B12-B22-B l)j}, (16)

and p represents the fugacity coefficient of pure solvent in
saturation conditions obtained from,

S= exp BP / (RT)) (17)
where P is the vapor pressure of pure component 1 at the
equilibrium temperature, which can be obtained using a vapor
pressure equation.


290 294 298 302
T/K
Figure 3. Solubility of carbon dioxide (*) and nitrous ox-
ide (A) in water. The curves were obtained from Eq.(18).


Chemical Engineering Education


8e-4

7e-4

6e-4

5e-4

4e-4











The calculation ends when convergence is obtained between
two consecutive x2 values. The determination of Henry's
constant, H21, is then straightforward from Eq. (8). All the
solubilities found in this work were corrected to 1 atm partial
pressure using Henry's law, since the literature values are
referred to this pressure.


TABLE 1
Solubility of CO2 and N20 in water, expressed as mole fraction,
partial pressure P, = 101325 Pa. H,, is the Henry coefficie


ERROR Al

The analy
of any scien
that all meas
subject to sc



Sx at a
nt.


Solute T/K x/10 X2h/10 a (%)( H2,1 /(MPa)
CO, 290.27 7.70 7.61 1.1 131.7
291.49 7.36 7.34 0.3 137.7
292.11 7.22 7.21 0.2 140.4
293.39 6.90 6.94 0.6 146.8
294.58 6.65 6.71 0.9 152.3
295.15 6.66 6.60 0.9 152.1
296.19 6.36 6.42 0.8 159.2
297.19 6.19 6.24 0.9 163.8
298.39 5.98 6.04 1.0 169.4
299.37 5.89 5.89 0.0 172.0
300.15 5.80 5.77 0.5 174.8
301.10 5.61 5.63 0.3 180.5
302.13 5.43 5.48 1.1 186.7
302.93 5.35 5.38 0.5 189.5
NO 290.36 5.49 5.53 0.6 184.4
291.36 5.38 5.36 0.4 188.4
292.36 5.17 5.19 0.4 196.1
293.30 5.03 5.04 0.2 201.5
294.20 4.93 4.90 0.5 205.5
295.40 4.71 4.73 0.3 214.9
296.15 4.66 4.63 0.8 217.2
297.25 4.49 4.48 0.3 225.6
298.13 4.38 4.37 0.2 231.5
299.07 4.28 4.25 0.7 236.6
300.16 4.13 4.12 0.2 245.2
301.15 3.97 4.01 1.0 255.0
302.15 3.84 3.90 1.7 263.9
303.13 3.81 3.80 0.0 266.3

N -N
a ((%) =-- -O- O \ 100, where x represents the solubility value found in
X21t
this work and x2it is obtained from Reference 8.

TABLE 2
Parameters in the equation Rln x2 = A + B/T + ClnT.
Solute A(J K1 mol1) B/(J mol1) C/(J K1 mol1) AAD(%)
CO, -2785.2 138287.6 396.617 0.5
NO 890.1 -23763.0 -153.526 0.5

Vol. 42, No. 3, Summer 2008


ANALYSIS

sis of uncertainties or "errors" is a vital part
tific experiment. The instructor should stress
urements, however careful and scientific, are
)me uncertainties. In this experiment the de-
termination of the uncertainty of the final
result, the mole fraction solubility, provides
use of the propagation law of errors.71 This
law gives the relative magnitude of the
uncertainties of the measured variables,
i.e., the student will be able to tell which of
the experimental errors affect the solubility
value more.

Appendix B presents an entrance form
distributed to each group of students in order
to guide them in treating the raw data and
analyzing the uncertainties of the results. The
data was obtained by one of these groups to
determine the solubility of CO, in water, at
the temperature 298.39 K.


RESULTS AND DISCUSSION

The experimental solubility data and the
values reported in the literature for the sys-
tems CO HO and NO /HO are shown in
Table 1. The accuracy of the experimental
method is found to be about 1 percent.

We have also determined the ideal solubili-
ties of both gases at P = 101325 Pa using
Eq. (5). The vapor pressures of the pure
components were obtained from the Wagner
equation. [9 The value determined for CO2 is
x, = 1.57x102 (T = 298.39 K) and for NO is
x, = 1.91x102 (T = 298.13 K). These values
are quite different from the experimental
ones, since Raoult's law gives only a rough
estimate of the solubility of a gas independent
of the solvent.

The second virial coefficients were ob-
tained from polynomial functions fitted to
the experimental second virial coefficients
taken from the Dymond and Smith compila-
tion.1101

The dependence of the solubility on tem-
perature has been represented by:

Rlnx =A +B/T+ ClnT, (18)

with the parameters fitted to the data by a
least-squares method. The optimized param-
eters of Eq. (18) and the average absolute
151











deviation of x,, defined as,

AAD= (1 / li- I \ (exp)- x, (calc) / x (exp) x 100 (19)

where M is the number of experimental points, are listed in
Table 2 (previous page).
In Figure 3 (p. 150) we have plotted the experimental
solubility data and the fitted Eq. (18) for each system. The
agreement is good.

CONCLUSIONS
The apparatus used here to measure the solubility of gas
in a liquid is simple and clearly illustrates this concept. It
combines easy handling with automated data retrieval, lead-
ing to experimental results with reasonable accuracy for a
pedagogical experiment.
The treatment of the raw data to obtain the solubilities
is a good application and demonstration of some gas laws
(Dalton's law of partial pressures, Raoult's and Henry's laws).
This experiment also gives the opportunity to demonstrate
real behavior of a gas (through the fugacity concept) to the
students. The analysis of errors will allow evaluation of the
relative magnitudes of the uncertainties within the measured
variables and how they propagate to the final result, the
solubility.

ACKNOWLEDGMENT
This work was carried out under Research Project POCTI/
EQU 44056/2002 financed by FCT-FundaAio para a Cien-
cia e Tecnologia (Portugal) and FEDER. We thank Professor
Margarida F Costa Gomes from the Laboratoire de Ther-
modynamique des Solutions et des Polymeres, Universit6
Blaise Pascal, Clermont-Ferrand, for her advice related to
the assembly of the solubility apparatus.

NOMENCLATURE
A, B, and C parameters of Eq. (18)
fL fugacity of component 2 in the liquid phase [Pa]
f*,L fugacity of pure liquid 2 [Pa]
f fugacity of component 2 in the vapor phase [Pa]
fo standard-state fugacity of component 2 [Pa]
H2,1 Henry constant of component 2 in solvent 1 [MPa]
n amount of solvent [mol]
n2 amount of solute [mol]
np number of encoder pulses
P equilibrium pressure [Pa]
P2 partial pressure of component 2 [Pa]
P,2 vapor pressure of pure component 2 [Pa]
P vapor pressure of pure component 1 [Pa]
R ideal gas constant [J mol1 K 1]
T equilibrium temperature [K]
AV volume of dissolved gas [m3]
x2 mole fraction of component 2 in the liquid phase
y mole fraction of component 2 in the vapor phase


Greek symbols
Ah displacement of the piston [m]
(P2 fugacity coefficient of component 2 in the vapor phase

REFERENCES
1. Wilhelm, E., and R. Battino, Chem. Rev., 73, 1 (1973)
2. Letcher, M.T., and R. Battino, J.Chem. Educ., 78, 103 (2001)
3. Fonseca, I.M.A., J.PB. Almeida, and H.C. Fachada, J. Chem. Ther-
modynamics, 39, 1407 (2007)
4. Van Ness, H.C., and M.M. Abbott, Classical Thermodynamics of
Nonelectrolyte Solutions, Mc Graw Hill (1982)
5. Prausnitz, J.M., R.N. Lichtenthaler, and E.G. Azevedo, Molecular
Thermodynamics of Fluid-Phase Equilibria, 3rd Ed., Prentice Hall,
New Jersey (1999)
6. Gerrard, W., and PG.T. Fogg, Solubility of Gases in Liquids, Wiley,
New York (1991)
7. Taylor, J.R., An Introduction to Error Analysis, 2nd Ed., University
Science Books (1997)
8. Wilhelm, E., R. Battino, and R. Wilcock, Chem. Rev., 77, 219 (1977)
9. Prausnitz, J.M., B.E. Poling, and J.P O'Connell, The Properties of
Gases and Liquids, 5th Ed., Mc Graw Hill, New York (2001)
10. Dymond, E.H., and E.B. Smith, The Virial Coefficients of Pure Gases
and Mixtures, Clarendon, Oxford (1980)

APPENDIX A

The experimental procedure to determine the solubility of
a gas in a liquid using the apparatus shown in Figure 2 is as
follows:

1) Connect the lecture bottle to the solubility apparatus
by means of a pressure reducer.

2) With the valve of the lecture bottle closed, open the
valve of the pressure reducer.

3) Switch on the vacuum pump and open stopcocks V1,
V3, and V4 with V2 closed. Evacuate the whole ap-
paratus during 1 h.

4) Close the stopcocks. Remove the equilibrium vessel
from the line, lowering the level of the thermostat bath
in order for the connector of the equilibrium vessel to
be out of the water.

5) Introduce ~ 6 cm3 of water in the equilibrium vessel.

6) Connect EQ again to the vacuum line and re-establish
the initial position of the thermostat bath.

7) In order to degas the water, open V2 and V4. Switch
on the stirrer. This procedure lasts about 30 min until
the measured pressure equals the solvent vapor pres-
sure at the equilibrium temperature.

8) Close stopcocks V2 and V4 and remove the equilib-
rium vessel again from the line.

9) Weigh the equilibrium vessel and then connect it to
the vacuum line, re-establishing the level of water
bath using the elevator.

10) Open stopcock V2, (with V3 and V4 closed). Then
slowly open stopcock V1 to admit the gas to the equi-
librium vessel.


Chemical Engineering Education











11) Slightly open the valve of the pressure reducer and
then the valve of the lecture bottle in order to adjust
the total pressure to ~ 1 atm (reference pressure).

12) Close stopcock V1 and open V3.

13) Switch on the magnetic stirrer to promote contact
between the liquid and vapor phases.
14) The dissolution process lasts about 2 h. As the gas
dissolves the pressure decreases, which is detected by
PT. The piston comes slowly down to re-establish the
reference pressure.
15) Record the last constant value displayed by PC, which
is the number of pulses (np) of the linear actua-
tor. T /
16) To convert np in displacement of the piston 298
use the conversion factor indicated in the user
manual of the linear actuator. The value obtained must
be multiplied by the internal crosssection area of the
cylindrical gas burette to obtain the volume of the dis-
solved gas (AV).

APPENDIX B
(The data presented in the following tables was obtained
by a group of students.)
Entrance form for GIL solubility data
Gas: CO2
Solvent: water
Raw data: n is the number of encoder pulses; T is the
equilibrium temperature; P is the reference pressure; m1 is
the mass of the solvent.

T / K n P / Pa m, / kg
298.39 82346 100070 0.007737

Treatment of the experimental data
The value of n must be converted in Ah, the dis-


placement of the piston, using the conversion factor
indicated in the user manual of PC. Using Eq. (9) cal-
culate AV, the volume of gas dissolved in the liquid.
Write a simple computer program to calculate the
solubility of the gas and the Henry coefficient, at a
partial pressure of 101325 Pa.

Results: x'2 represents the mole fraction solubility calcu-
lated with Eqs. (9-17); (p the fugacity of the solvent in the
vapor phase obtained from Eq. (16); (p* the fugacity coef-
ficient of pure solvent in saturation conditions, from Eq. (17);
x, is the corrected solubility at partial pressure P2 = 101325

K x'2/104 9p 9' I x/ 104 H2,/(MPa)
.39 5.88 0.9880 0.9985 5.98 169.4


Pa; H,,1 is the Henry coefficient.

Error analysis

What are the uncertainties of the measured variables?

Using these uncertainties and the propagation law of
errors, determine the uncertainty in the final result, the
mole fraction solubility.

Which uncertainty most affects the solubility value?

Using your own results of solubility data and some of
the results of other groups (at least 6 pairs of experi-
mental data), make a fitting by a least squares method
of the equation:

R In x = A + B/T + C lnT

to represent the temperature dependence of the solu-
bility.
Calculate the average absolute deviation of the fitting.

Compare your solubility data with those of the litera-
ture. n


Vol. 42, No. 3, Summer 2008











|L =' laboratory




MIXING HOT AND COLD WATER STREAMS

AT A T-JUNCTION



DAVID SHARP, MINGQIAN ZHANG, ZHENGHE Xu, JIM RYAN, SIEGHARD WANKE, AND ARTIN AFACAN
University of Alberta Edmonton, Alberta, Canada T6G 2G6
chemical Engineering students at the University of Alberta
Sthe it er David Sharp is a faculty service officer at the University of Alberta. He
are taught how to write technical reports in the first term received his BS.c. and MS.c. in chemical engineering from the University
of their third year. At this point they have a very limited of Alberta. He has been teaching entry- and senior-level undergraduate
laboratory courses. His research interests are in the area of surface phe-
background, having only taken introductory thermodynamics and nomena, computational fluid dynamics, and separation processes.
a mass and energy balance course. The objectives of this course Mingqian Zhang is a chemical technologist at the University of Alberta.
are to provide a bridge between theoretical study and practical He obtained his MS.c. in physical chemistry from the Chinese Academy
applications, and to apply these principles to critical analysis of Sciences and his MS.c. in chemical engineering from the University of
applications, and to apply these principles to critical analysis Alberta. His work involves designing and testing of various undergraduate
of real experimental data in a professional, clearly written, and labs, and his major research interests include polymer characterization,
concise format. Furthermore, the experiment described in this polymer nanocomposites, catalysis, and reaction kinetics.
paper exposes many students to their first look at real equipment Zhenghe Xu is the Teck Cominco professor and Canada Research chair
in Mineral Processing at the University of Alberta. He received his BS.c.
and measuring devices. and MS.c. degrees in Minerals Engineering from the Central South Univer-
sity and Ph.D. from the Virginia Polytechnic Institute and State University.
To teach entry-level chemical engineering students with limited He has contributed over 200 publications on interfacial phenomena with
theoretical and statistical analysis background to write technical emphasis on mineral and materials processing.
reports and apply material and energy balance principles to a Jim Ryan is a professor Emeritus at the University of Alberta. He received
his Ph.D. degree in chemical engineering from the University of Missouri
critical analysis of real data, it is necessary to use a simple experi- (1965). He has taught thermodynamics, fluid mechanics, and process
ment. Most previous studies involving the mixing of heated water design for over 30 years.
require dynamic analysis of stirred-tank heaters;[1, 2] however, Sieghard Wanke is a professor at the University of Alberta. He received
students with a limited background would have trouble with the his BS.c. and MS.c. degrees in chemical engineering from the Univer-
tudents with a limited background would have trouble ith the sity of Alberta and his Ph.D. from the University of California, Davis. His
theory of such systems. A simpler experiment is the mixing of hot research area is heterogeneous catalysis with emphasis on supported
and cold water at a T-junction. This experiment can be used to metal and olefin polymerization catalysts.
demonstrate how to use steady-state material and energy balances Artin Afacan is a faculty service officer in the Department of Chemical
and Materials Engineering at the University of Alberta. He received his
as a troubleshooting tool to predict flow rates and temperatures BS.c. (1975) in chemical engineering from Istanbul Technical University.
of the mixed water stream. Also, this experiment emphasizes the Has been teaching entry- and senior-level undergraduate laboratory
courses. His research interests are in the areas of particle-fluid dynam-
importance of properly placing process measurement devices, ics as applied to slurry pipeline, cyclone separator in oil processing,
i.e., the thermocouples in the current experiment. Furthermore, tower internals, (tray and packing) efficiency, and catalytic distillation
in separation processes.
material and energy balances can also be used to check flow in separation processes.
rate calibration equations of the orifice meters for each of the
experimental streams. It is important for students to learn the importance of proper calibration in process measuring devices
since calibration equations can change over time due to corrosion, erosion, or scale buildup during its use.

THEORY
For any given continuous, nonreactive process at steady-state, the general material and energy balances can be written as
(Felder and Rousseau[3]):
mot mm =0 (1)

2 2
Q-W 0mot hout gz out m mm hm + gzm (2)


Copyright ChE Division of ASEE 2008
154 Chemical Engineering Education











In the present study, hot- and cold-water streams are mixed
at a T-junction to produce one mixed stream. Assuming the
hot- and cold-water streams are completely mixed at the T-
junction, the system is adiabatic, work is neither done by or
to the system, there are no frictional losses, and that kinetic
and potential energy changes are negligible, Eq. (1) and Eq.
(2) can be reduced to

mmi,pred cold +mhot (3)

m.,mea. h .. moldd hold mhothhot (4)
where h is the enthalpy of each stream and can be defined as
(Cengel and Boles 41):
h=C,(T-T0) (5)

By setting the reference temperature to 0C and assuming
the heat capacity, Cp, is constant for water in the temperature
range investigated in the experiment (15 to 500C), the tem-
perature of the mixed stream could be predicted by combining
Eq. (4) and Eq. (5) to give

T cold cold + mhotTht (6)
pred m
mIx,meas

The predicted temperature of the mixed stream using Eq.
(6) is dependent upon measured information of the streams
before and after the T-junction. In order to determine which
(if any) mass flow rate is wrong, Eq. (3) and Eq. (6) can be
solved simultaneously to predict the mixed stream tempera-
ture. For example, if the mass flow rate of the cold stream is
assumed to be wrong, the temperature of the mixed stream
can be predicted by replacing the cold stream flow rate using
Eq. (3) to give the predicted temperature as


T
pred,nom


(mmix,meas -mhot) cold +mhot hot
m
mllx .Illasg


Similarly, if the mass flow rate of the hot or mixed stream is
assumed to be wrong, the predicted mixed stream temperature
can be calculated by


T n
pred, no m,


cold old + (mmix,mnas
mi
mixmeas


moold) Thot


m T +motT
T cold cold h ot hot (9)
pred, no, (m + m, )
(mold + mhot)
If all three measured mass-flow rates are correct, then Eq.
(7), Eq. (8), and Eq. (9) will all give the same value for Tpred.mlx
If one calibration equation is incorrect, however, then only
one of those three equations will accurately predict the mixed
stream temperature, which will agree with the measured
temperatures after the T-junction and therefore indicate that
the mass flow rate of the missing stream in the equation is
incorrect.

EXPERIMENTAL SETUP AND PROCEDURE
The experimental setup is shown in Figure 1. The entire
setup is constructed using half-inch nominal copper pipe,
fittings, and brass valves. The feed lines are connected to the
domestic hot and cold water supply lines. The hot and cold
water streams are mixed at a T-junction before exiting into the
drain. As can been seen from this figure the experiment can
be done with the hot and cold water streams flowing through
either of the inlet lines by setting the appropriate valve com-
bination at the inlet manifold. The flow rate of the water is
controlled using globe valves and can be roughly set using


Tcqld Thot Tmixl Tmjx2 Tmix3 1. Globe valve
2. Pressure gauge
mcold .......... 3. Orifice meter
mhot .......... OPTO 22 4. Differential pressure cell
mmix .......... 5. J-type thermocouple

Figure 1. Schematic Diagram.

Vol. 42, No. 3, Summer 2008


175 mm 340 mm 325 mm 330 mm 190 mm 168 mm










the pressure gauges (Wika). The flow rates of the hot, cold,
and mixed streams are measured using a combination of an
orifice meter and differential pressure cell (Validyne). The
fluid temperatures are measured using J-type thermocouples.
The temperature of the mixed stream is measured at 60 mm
(Tmixl), 250 mm (Tmix2), and 400 mm (Tmix3) after the
T-junction. This is done to determine the proper location for
the thermocouple in order to measure the correct temperature
of the mixed stream. The detailed locations of other thermo-
couples, orifice meters, pressure gauges, and globes valves are
also shown in Figure 1. Analog signals from the differential
pressure cells and thermocouples are converted to digital sig-
nals using an OPTO 22 system. These signals are sent to the
personal computer, where they are stored and displayed using
LabView versionn. 1) software. In this experiment, the mass
flow rates of the hot, cold, and mixed streams are recorded
in terms of volts and temperatures are recorded in terms of
degrees Celsius. The following orifice calibration equations
are needed to convert volt readings to mass flow rates.

ml = 0.0265 7d (10)

mhot =0.0251 V/h (11)
m = 0.0473 V (12)

All three calibration equations have a systematic error
(accuracy) of 5 %. To run the experiment, the flow rates of
the hot and cold water must be set using the globe valves in
conjunction with the pressure gauges. The data acquisition
system must be initiated to record the five temperature and
three flow rate readings at a set time interval (usually set at
either 2 or 5 seconds). When the flow rate and temperature
profiles shown by the software remain constant, steady-state
can be assumed for that run. At least 100 seconds worth of
data for each run should be recorded to ensure the system
is at steady-state and to get sufficient sample points for a
reasonable analysis. The same experimental procedure is
then repeated at different flow rate settings as many times as
possible (this depends on class size and laboratory availabil-
ity), and if reproducibility is to be examined then at least one
flow rate setting must be repeated multiple times. The entire
experiment can be completed in 10 to 30 minutes, depending
on the number of runs students conduct. The short time span
of the experiment and large number of flow rate combinations
enables even large classes to do individual experiments in a
rather short time period. The data is recorded in a Microsoft
Excel file that contains the hot, cold, and mixed stream volt-
age readings from the DP/cell's and the temperatures in C
from the thermocouples.

RESULTS AND DISCUSSION
For this example, five separate runs at various hot- and cold-
water flow rates were conducted to illustrate the principles of
material and energy balances. Additionally, four more runs
were conducted to show the reproducibility of the data. When


comparing experimental and predicted results it is necessary
to do an error analysis on the variables being compared. The
total error for an experimental value can be determined from
the sum of the systematic (accuracy) and random (precision)
errors of the data. The accuracy error comes from the maxi-
mum absolute error in calibration of the measuring device.
For this study, the maximum absolute errors in the calibration
equations for the measured mass flow rates and temperatures
are 5% and 0.3C, respectively. The precision error can
be obtained directly from the standard deviation, o, of the
measured values. To determine the precision error, Coleman
and Steele5s] state that when the number of data points for
one time series is equal to or greater than 10, two times the
standard deviation gives a good approximation for the 95 %
confidence interval. Therefore, the total error in the measured
mass flow rates and temperatures is calculated by
'=m 0.05m+ 2( (13)

T =[0.3+ 2( (14)

In order to illustrate how material and energy balances
can be used to determine an incorrect calibration and to
predict exit stream temperature, orifice calibration Eq. (10)
for the cold stream was changed to (without the students
knowledge)

mol = 0.0168 V (15)
The students are then expected to find the incorrect calibra-
tion equation and to develop anew orifice calibration equation
to fit the data they have.
Figures 2 and 3 show the measured mass flow rates and
temperatures as well as the predicted values for all five runs
as a function of time, respectively. The measured cold, hot and
mixed stream flow rates were calculated using the calibration
Eq. (15), Eq. (11), and Eq. (12), respectively. The predicted
mass flow rate for the mixed stream was calculated using Eq.
(3). For these calculations the average hot- and cold-water
flow rates over each steady-state period were used. The pre-
dicted temperature of the mixed stream was calculated using
Eq. (6) and time-averaged mass flow rates and temperatures
for the hot, cold, and mixed stream water streams. The un-
certainty for the predicted mass flow rate and temperature of
the mixed stream were calculated using the method described
by Coleman and Steele5s] and Holman.J61 The uncertainty in
the predicted mass flow rate is based on experimental errors
in the measured hot- and cold-water stream flow rates and
can be calculated by


pm
mmx ,pred
am -m
cold


mm Lxpred (1
am hot (h6)
hot


Similarly, the uncertainty in the predicted temperatures
obtained from Eq. (6) to Eq. (9) can be determined from the
errors in the measured hot, cold, and mixed stream flow rates
Chemical Engineering Education











as well as the measured hot and cold stream temperatures. The uncertainty in the predicted temperature is given by


=T T
O- = -_-red


TT


i-i


o measured hot stream
+ measured cold stream
A measured mixed stream
-predicted mixed stream


0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time, t (s)
Figure 2. Mass flow rates of hot, cold, and measured and predicted mixed
streams for five steady-state runs.
50



45
hot

40
ATmixl
U Tmix2 d
35 x Tmix3predicted



E Al a1


25

cold
20



15


0 100 200 300 400 500 600
Time, t (s]


700 800 900 1000 1100 1200


'"WO148014*
*l^WIWl kp .f


Vol. 42, No. 3, Summer 2008


LT ,
pred
/0Trm ,,jmc


LT
pred
~9 jmn


pred
tmot


pred
Gm. t 2


(17)


E 200


150
,,
o

S100



50


0


From these figures it is evi-
dent that there were five distinct
steady-state periods having
time ranges of 0 to 130 s, 155
to 325 s, 390 to 625 s, 675 to
835 s, and 885 to 1135 s. The
constant flow rate profiles of
the measured values in Figure
2 indicate that the system was
indeed behaving at steady-state
for each of the five runs. This is
also evident in Figure 3 where
the measured temperatures
before and after the T-junction
remain constant for each stream
and steady-state period. Since
the system is non-reactive and
is at steady-state, the material
balance equation, shown in Eq.
(3) should be valid. Figure 2,
however, clearly shows that
the predicted flow rate of the
mixed stream does not agree,
within error, with the measured
mixed stream flow rate for the
first four runs-leading to the
conclusion that one, or more, of
the orifice calibration equations
must be incorrect.
The mixed stream tempera-
ture when measured only 60
mm (Tmixl) downstream of
the T-junction has larger error
bars and is always approxi-
mately 1C lower compared to
the temperatures measured at
270 mm (Tmix2) and 400 mm
(Tmix3) downstream of the T-
junction, as shown in Figure 3.
The error bars were calculated
by assuming each thermocou-
ple has the same systematic


4 Figure 3. Temperatures
of hot, cold, measured
mixed and predicted mixed
water streams for five
steady-state runs.


___


^Wsabfa@s6~)











(accuracy) error of +0.3 C.
Therefore, the larger total error
determined at the Tmixl loca-
tion must be only due to higher
standard deviations. This, com-
bined with the fact that the
measured average temperature
reading at the Tmixl location
is lower than the temperatures
measured at the Tmix2 and
Tmix3 locations, indicates that
at the thermocouple closest to
the T-junction the two streams
have not completely mixed.
Furthermore, the measured
temperatures at the Tmix2 and
Tmix3 locations are not only
very similar, but also have very
similar size error bars, which
indicates that a thermocouple
needs to be placed a minimum
of 270 mm downstream of the
T-junction to ensure complete
mixing. Figure 3 also shows
that thi ndAi;Itlt t dmnlt t"ll


TABLE 1
Average mixed stream temperatures after the T-junction for all five steady-state trials
Run Tmix2 (C) Td, nom.1 ( Td nom (C) Tdnom (C)
pe ,nomnlc 0 ) pred,nom t pred,nomnl ,
1 27.9 +0.5 27.6 +0.7 33.5+ 1.0 29.4 +0.7
2 30.3 0.6 30.2+ 1.0 34.9+ 1.0 32.5 +0.8
3 32.4 +0.8 32.4 1.2 36.4+ 1.0 34.8 +0.9
4 34.1 0.7 34.1+ 1.3 37.3+ 1.0 36.4+ 0.9
5 36.5+0.6 36.9+ 1.6 38.5 0.7 38.2 0.7

TABLE 2
Experimental average mass flow rates for all five steady-state runs
Run mh0t (kg/h) m1.d (kg/h) mmx (kg/h)
1 67.0 5.7 131.5 9.1 275.7 +16.4
Standard deviation 1.2 1.3 1.3
2 108.6+ 8.2 116.7+ 8.1 292.9+ 18.9
Standard deviation 1.4 1.2 2.1
3 145.0+ 11.2 94.7 6.4 298.4+ 19.5
Standard deviation 2.0 0.8 2.3
4 173.7+ 12.1 71.7+5.1 293.7+ 18.4
Standard deviation 1.7 0.8 1.9
5 205.5+ 16.2 44.0+3.6 272.1 18.6
Standard deviation 3.0 0.7 2.5


tlCIL Ltll Ul^Ul ^L-U t1,ll Cll U ^l ttul
does not agree within error with the measured
values at Tmix2 or Tmix3 for the first four
runs, again indicating that there is likely an Run
incorrect calibration equation. For each run, 3
the difference between the measured and 6
predicted temperatures is decreasing until the
7
fifth run, where the measured and predicted
values agree within experimental error. This
agreement is likely due to the decreasing cold 9
stream flow rate, which has less of an effect
on the predicted temperature of the mixed stream.

To determine which calibration equation was incorrect,
material and energy balances were solved simultaneously to
predict the mixed stream temperature, which can be compared
with the measured values. Table 1 shows the average tem-
perature measured at 270 mm downstream of the T-junction
(Tmix2) as well as three predicted mixed stream temperatures
calculated by eliminating one of the mass flow rate terms in
the energy balance equation. The mixed stream temperature
was calculated using Eq. (7), Eq. (8), or Eq. (9) by eliminating
the use of the flow rate readings of the cold, hot, or mixed
streams, respectively. From this table, it is evident that the
only predicted temperature that agrees within error of the
measured temperature for all five runs is the one in which
the flow rate of the cold stream, mold, is eliminated from the
calculation. This finding indicates that the cold stream flow
rate calibration equation was incorrect. Using linear regression
analysis between the corrected flow rates of the cold stream


TABLE 3
Reproducibility of the inlet conditions
mh0t (kg/h) m1.od (kg/h) Thot (C) T01,d (C)
145.0+ 11.2 94.7+ 6.4 42.8+ 1.1 22.5 +0.9
145.6+ 11.6 99.1 +8.6 41.7 +0.6 22.8 &0.9
144.1 +9.0 97.3+ 7.2 41.9+ 0.8 22.8 +0.8
144.4+ 13.2 97.4+ 6.9 42.1 0.5 23.1 0.6
148.0+ 10.1 96.5 +7.6 42.7 +0.8 22.7 &0.9


obtained from Eq. (3) and the corresponding average volt-
age readings of the cold stream for the five runs, the correct
calibration equation was determined to be

mol = 0.0268 V (18)
cold \ cold

This equation is valid for voltage readings between 0.5
and 4.7 Volts, corresponding to flow rates of 66 to 210 kg/h.
The R2 value is 0.996, so the calibration equation should be
quite accurate.
Students can also use error analysis to explain or discuss
trends observed in the experimental data. Table 2 shows the
measured average mass flow rate of the hot, cold, and mixed
streams with total errors and the corresponding standard de-
viations for each steady-state run. From this table, it can be
seen that the total error in the measured hot and cold streams
increase as the flow rates of both streams increases. Since the
total error for the experimental values was determined from
both the accuracy and precision of the data, the total error


Chemical Engineering Education










increase in the measured values is due to the increase in both
the flow rates and the standard deviations in both streams as
shown in Table 2. Similar observations and discussions can
be made for the mixed stream.
The reproducibility of the study was also examined by re-
peating one of the runs four times. The time average values of
the hot and cold stream flow rates and temperatures are shown
for the five runs in Table 3. The flow rate of each stream was
readjusted between runs and reset to the same value. Even
though the way in which the flow rate is set by the student
is very crude (a globe valve and a pressure gauge) and both
streams were connected to the domestic cold and hot water
supply lines, it can be seen that flow rates for the cold stream
and for the hot stream are all within experimental error of
each other-verifying the reproducibility of the flow rate set-
tings. Table 3 also shows that the hot and cold temperatures
agree within experimental error and the supply temperatures
remained relatively constant for all repeat runs.
It should be noted that the experiment outlined in this re-
port is only one of many possible ways in which the students
can be asked to analyze this system. A few other examples
include: 1) the thermocouples could be setup so that one or
more was malfunctioning, with students asked to determine
which one(s) are malfunctioning and why; 2) only one or
two calibration equations could be given to the students and
they then asked to determine the unknown ones; 3) all given
information could be correct allowing the students to test
material and energy balance principles used to predict a mixed
stream flow rate and temperature. Depending on the setup
and number of runs conducted, more or less emphasis could
be placed on reproducibility of data and/or error analysis. To
this end the experiment described here is fairly flexible and
allows the instructors) to vary the experiment from year to
year, while retaining the fundamentals.
CONCLUSIONS
A simple mixing of a hot- and cold-water stream at a T-junc-
tion was investigated. The main objective was to use mass
and energy balance equations to predict mass flow rates and
the temperature of the mixed stream after the T-junction, and
then compare these with the measured values. Furthermore,
the thermocouple location after the T-junction and the repro-
ducibility of the data were also investigated.
It was found that the predicted mixed stream flow rate
calculated using mass balance equations did not agree with
the measured mixed stream flow rate for all five runs. It was
concluded that one or more given orifice calibration equation
must be wrong. In order to determine which orifice calibration
equation was wrong, mass and energy balance equations were
solved simultaneously to predicted mixed stream temperature.
It was found that when only the cold stream flow rate was
eliminated from the energy balance, the predicted mixed
stream temperature was found to agree with all three measured


mixed stream temperatures within experimental error for all
five runs. This indicated that the given cold stream orifice
calibration was wrong.
The mixed stream temperature measured at 60 mm (Tmixl)
had a higher standard deviation error than the temperatures
measured at 230 mm (Tmix2) and 400 mm (Tmix3) down-
stream of the T-junction for all five runs. It was also found
that the temperatures measured at Tmix2 and Tmix3 locations
had similar absolute and standard deviations and error values.
Both observations indicated incomplete mixing at the Tmix 1
location. Therefore, to ensure complete mixing and minimize
heat losses, the thermocouple should be placed at least 230
mm downstream of the T-junction.
The reproducibility of the experimental data was also stud-
ied by repeating one of the runs four times. It was found that
the flow rates for hot and cold streams were all within error
of each other, verifying the reproducibility of the hot and cold
stream flow rate settings.

SUMMARY
In this paper, we proposed a simple experiment of mixing
a hot- and cold-water stream at a T- junction to demonstrate
how to use steady-state material and energy balance principles
in troubleshooting of an existing process and determining
the integrity and/or location of the measuring devices, such
as thermocouples and orifice meters. This experiment is
relatively inexpensive, requires little time to complete and
is conceptually simple to understand, making it ideal for the
undergraduate students who have a very limited chemical
engineering background.

NOMENCLUTURE
C P Constant pressure specific heat, J/(kg C)
g Gravitational acceleration, m/s2
m Mass flow rate, kg/s
h Enthalpy, J/kg
Q Heat transfer rate, W
T Temperature, C
To Reference Temperature, C
U Velocity, m/s
V Pressure drop across orifice, Volts
W Work input to the system, W
z Elevation, m
o Standard deviation of random variable
0o Error or uncertainty in a parameter

REFERENCES
1. Romagnoli, J.A., A. Palazoglu, and S. Whitaker, "Dynamics of Stirred-
Tank Heater Intuition and Analysis," Chem. Eng. Ed., 35, 46-49
(2001)
2. Muske, K.R., "Experience with Model Predictive Control in the Un-


Vol. 42, No. 3, Summer 2008











To teach entry-level

chemical engineering

students with limited

theoretical and statistical

analysis background to

write technical reports

and apply material and

energy balance principles

to a critical analysis of

real data, it is necessary to

use a simple experiment.


6

0
o

& o o
0 A 0A 0
I+

4

I 3
I 3 -t a






1 o hot stream
+






t+ cold stream
A mixed stream


0 200 400 600 800 1000 1200
Time, t (s)


1400 1600 1800 2000 2200


Figure Al. Voltage outputs from the DP/cells for the hot, cold,
and measured mixed streams for all nine steady-state runs.


dergraduate Laboratory," Comput. Appl. Eng.
Educ., 13(1), 40-47 (2005)
3. Felder, R.M., and R.W. Rousseau, Elementary
Principles of Chemical Processes, 3rd Ed., John 45
Wiley & Sons, New York, NY, pp. 86 and 323
(2000)
4. Cengel, Y.A., and M.A. Boles, Thermodynam- 40
ics: An Engineering Approach, McGraw- Hill,
New York, NY, p. 113 (1989)
5. Coleman H.W., andW. G. Steele, Experimenta- 35
tion and Uncertainty Analysis for Engineers,
2nd Ed., John Wiley & Sons, New York, NY,
pp 30 (1999)
6. Holman, J.P, Experimental Methods for Engi- E
neers, 7th Ed., McGraw-Hill, New York, NY,
p. 51(2001) 25

APPENDIX A: SAMPLE DATA,
CALCULATIONS AND ERROR 20
ANALYSIS
Sample Data: 15
15 ---
The raw data in terms of voltage vs. 0 200
time and temperature vs. time are shown
in Figures A1 and A2. Figu
Sample calculations:
Using the calibration equations to get mass flow rates from
known voltages (Figure Al):
Sample calculations based on the readings recorded at a
time of 10 s.
Hot stream voltage = 0.547 V
Cold stream voltage = 4.80 V
Mixed stream voltage = 2.90 V


400 600 800 1000 1200
Time, t (s)


1400 1600 1800 2000 2200


-e A2. Temperatures of hot, cold, and measured mixed
water streams for all nine steady-state runs.

To get the mass flow rates we use Eqs. (15), (11), and


(12)

mol =0.0168 V = 0.0168,4-80 =

mhot =0.0251 V =0.0251 0.547

m in.= 0.0473/- = 0.0473 29
mix,meas V mix


0.0368kg/s

0.0186 kg/s

= 0.0805kg/s


Chemical Engineering Education










The predicted mixed stream flow rate was calculated using the steady-state material balance (Eq. (3)) and average values for
flow rates. The average cold stream flow rate for the first run was 0.0365 kg/s, for the hot stream it was 0.0186 kg/s, and for the
measured mixed stream it was 0.0766 kg/s. Therefore, for the first run, the predicted mass flow rate of the mixed stream is

mmx,pred = old + mhot = 0.0365kg / s +0.0186kg / s = 0.0551kg / s

The predicted temperature of the mixed stream was calculated from Eq. (6) using average values for run 1.


mcoldTold + mhot hot
m
mix,meas


0.0365o22.9+0.0186o42.2 2
0.076621.2
0.0766


Assuming that the cold stream calibration equation is wrong, Eq. (7) can be used to predict the temperature from the average
values for the first run.


T no
pred,nomn


(mm..,ms -mhot)Told + mhothot (0.0766 0.0186)22.9 + 0.0186.42.2


m
mix,meas


0.0766


27.6C


Assuming that the hot stream calibration equation is wrong, Eq. (8) can be used to predict the temperature from the average
values for the first run.


'pred,nom,


moldTold + ixmea -mold)Thot 0.0365 @22.9+(0.0766 -0.0356)42.2
m 0.0766
mix ,meas


33.5C


Assuming that the mixed stream calibration equation is wrong, Eq, (9) can be used to predict the temperature from the aver-
age values for the first run


pred,nomr


mcoldTod + mhothot 0.0365 .22.9 +0.0186.42.2 = 29.4C
( old hot) 0.0365+0.0186


Error analysis:
The error in the measured flow rates is given as the maximum absolute error in the calibration equation plus two standard
deviations. For the mass flow rates the systematic error (accuracy) was 5 % for the full scale. For the first run it was deter-
mined that the standard deviations were 0.000355 kg/s, 0.000325 kg/s, and 0.000367 kg/s for the cold, hot, and mixed streams,
respectively. For the first run these standard deviations were based on 26 data points. Therefore, the total experimental error
for the mass flow rates is:


Cold Stream :

HotStream:w


0.05mold + 2(m

[0.05mhot +2(m


L 0.05.0.0365+ 2.0.000355J

0.05 0.0186 + 2 o 0.0003251 =


-0.00254 kg/s

:0.00158kg/s


Mixed Stream :


0.05m.....


0.05 0.0766 + 2 0.000367


For the temperatures the systematic error (accuracy) was + 0.30C for the full scale. For the first run it was determined that the
standard deviations were 0.200C, 0.190C, 0.400C, 0.120C, and 0.090C for the cold, hot, mixl, mix2, and mix3 thermocouples,
respectively. For the first run these standard deviations were based on 26 data points. Therefore, the total experimental error
for the mass flow rates is:


Cold thermocouple : w,

Hot thermocouple : T

Mixlthermocouple : wTmix1

Mix2 thermocouple : LTmix2

Mix3 thermocouple: LTm)3


0.3+ 2(T

0.3 +2cT


- 0.3 + 2 0.20j

10.3 +2.0.191=


0.3 + 2(m = + 0.3+2 0.40


0.3 + 2 Tmix2 _

0.3+ 2Tmix3


[0.3+2.0.12

[0.3+2.0.091


Vol. 42, No. 3, Summer 2008


:0.00456kg/s


-0.70 C

:0.68C

+1.10 C

=+0.54 C

=+0.48 C











Uncertainty in the predicted mixed stream mass flow rates is given by


mixpred
a old m


90m
mix,pred
amhot


in this case both partial derivatives are equal to 1 so the uncertainty in the predicted mixed stream mass flow rate becomes


w =L )2 +c2 = l 0.00254 2+(0.00158)
The uncertainty in the pr temperatures from Eqs. (6) to (9) is given by
The uncertainty in the predicted temperatures from Eqs. (6) to (9) is given by


WT T,
- [ mzpred
U)od= ,o---hot M


OT
prold
S cold


OT
preda
mix,meas


GT
prd
hot T


:0.0030kg/s


OT
pred
[ Told


For the uncertainty in Eq. (6) the partial derivatives are


&Tpred
mhot
OT d,
pred
cold


Tpred
am
mlx,meas


( m hot Thot c ldd )

(M-,-,M )2


OT d
sTpred
Thot
Opred
OTd
-ld


Thot
m
mix,meas
TId
cold
m
mixmeas


42.2 C
550.9-
0.0766 kg/s

22.9 C
-229 299.0-
0.0766 kg/s


-(0.0186 42.2 +0.0365.22.9)

(0.0766)2


mhot
m
m.x,meas

cold
m
mix,meas


oC
276.2-
kg/s


0.0186
= 0.2428
0.0766

0.0365
=3 0.4765
0.0766


Then, the uncertainty in Tpred is

O = i550.9* 0.00158)2 (299.0*0.00254 )2+

For the uncertainty in Eq. (7) the partial derivatives are


Trednomld

mhot


Thot Told
m
milx,meas


-276.2. 0.00456) +(0.2428. 0.68)2 +(0.4765.0.70)2


42.2-22.9 C
0 6 252.0k
0.0766 kg/s


G iT
pred,nomcld 0

amcold


Tpred,nomcld
am
mnimeas


mhot (Tld-T hot)

mixme.. )2


0.0186(22.9-42.2)

(0.0766)2


-61.2 C
kg/s


pred nom ld

&Thot


Opred,nomhot
Told
-ld


hot 0.0186 = 0.2428
m 0.0766
minx,mneas


m.x.ma mhot 0.0766 0.0186 0
= 0.7572
m 0.0766
mIix,mneas


Then, the uncertainty in Tpred is
T = (00.00158)2 -(-25200.00252 120100046)2 05234068)12 0.47650.7011/2
w^en = (0*0.00158) +-252.0*0.00254) +(l20.1* 0.00456) +(0.5234*0.68) +(0.4765*0.70)


1.0 C


Chemical Engineering Education


1.7C












This experiment is relatively inexpensive, requires little time to complete, and is

conceptually simple to understand, making it ideal for the undergraduate

students who have a very limited chemical engineering background.


For the uncertainty in Eq. (8) the partial derivatives are
dT
pred,nomhot 0
=0
amhot
OTprd,nomhot T old Thot 22.9 42.2 C25
-252
mold m... 0.0766 kg/s
cold mixmeas
OTprednmhot m cold(Thot Told) 0.0365(42.2-22.9) C
= 120.1
mx, (m...... 2 (0.0766)2 kg/s
Tpred. nomht In old 0.0766 0.0365 0.
= 0.5234
Thot m,.. 0.0766
ot mxmeas 0.0365

Told m 0.0766
cold mixmeas
Then, the uncertainty in Tpred is

OWT = (0.0.00158) -2252.0 0.002542 (120.1 0.00456) 0.5234*0.68) +(0.47650.70)2 2


For the uncertainty in Eq. (9) the partial derivatives are

T prednodmm coldd + hot Thot (hotThot coldTold
amhot cold hot)2

(0.0368 + 0.0186)42.2- 0.0186.42.2 0.036822.9)231
=-=231
(0.0368+0.0186)2
Tprednon (m hot cold) Told hothot cold Tld)

cold (cold + hot)2

(0.0186+0.0368)22.9-(0.0186.42.2 0.0368.22.9)
=-= -11
(0.0368+0.0186)2
T pred,nonmm
0-
Om
mix ,mieas


C
.41--
kg/s




C
7.0--
kg/s


Tred,nomm mhot
&Thot m old mhot
Tpred,nom ld cold
OTold cold mhot


0.0186 0.3357
0.0368+0.0186

0.0368 = 0.6643
0.0368+0.0186


Then, the uncertainty in Tpred is

O^- =~ (231.41.0.00158)2 +(-117.00.00254)2 + (0.004562 + (0.3357 0.69)2 + (0.6643 0.71)2


Vol. 42, No. 3, Summer 2008


:1.0C


:0.7C












TABLE Al
Corrected Mass Flow Rates and Corresponding Voltage Readings for the Cold Stream.
run Corrected cold stream mass flow rate, moo,d Average cold stream voltage, Vol,
c__orreted (kg/s)
10. 058 4.7
20. 051 3.7
30. 043 2.5
40. 033 1.4
50. 019 0.5

Since the calibration equation for the cold stream mass flow rate is incorrect, it was necessary to find a new, corrected calibra-
tion equation. This was done using the material balance and solving for the cold stream mass flow rate. For the first run

oold,corrected mx,meaured hot = 0.0766 -0.0186 = 0.058kg / s
The average voltage reading for the corresponding run was 4.7 V for the cold stream. Table Al shows the data for all five
runs with different flow settings.

Regression analysis was done between the corrected mass flow rate and the square root of the voltage reading. The new
calibration equation is

mold = 0.0268 Vd
J


Chemical Engineering Education







Visit
us
on the
Web
at




Full Text

PAGE 1

Vol. 42, No. 3, Summer 2008 117 Chemical Engineering Education Volume 42 Number 3 Summer 2008 CHEMICAL ENGINEERING EDUCATION (ISSN 0009-2479) is published quarterly by the Chemical Engineering Division, American Society for Engineering Education, and is edited at the University of Florida. Co r respondence regarding editorial matter, circulation, and changes of address should be sent to CEE, Chemical Engineering Department, University of Florida, Gainesville, FL 32611-6005. Copyright 2008 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 120 days of pu b lication. Write for information on subscription costs and for back copy costs and availability. POSTMA S TER: Send address changes to Chemical Engineering Education, Chemical Engineering Department., University of Florida, PUBLICATIONS BOARD EDITORIAL AND BUSINESS ADDRESS: Chemical Engineering Education Department of Chemical Engineering University of Florida Gainesville, FL 32611 PHONE and FAX : 352-392-0861 EDITOR Tim Anderson ASSOCIATE EDITOR Phillip C. Wankat MANAGING EDITOR Lynn Heasley PROBLEM EDITOR James O. Wilkes, U. Michigan LEARNING IN INDUSTRY EDITOR William J. Koros, Georgia Institute of Technology CHAIRMAN John P. OConnell University of Virginia VICE CHAIRMAN C. Stewart Slater Rowan University MEMBERS Kristi Anseth University of Colorado Jennifer Curtis University of Florida Rob Davis University of Colorado Pablo Debenedetti Princeton University Dianne Dorland Rowan Thomas F. Edgar University of Texas at Austin Stephanie Farrell Rowan University Richard M. Felder North Carolina State University H. Scott Fogler University of Michigan Jim Henry University of Tennessee, Chattanooga Jason Keith Michigan Technological University Steve LeBlanc University of Toledo Ron Miller Colorado School of Mines Susan Montgomery University of Michigan Lorenzo Saliceti University of Puerto Rico Stan Sandler University of Delaware Donald R. Woods McMaster University DEPARTMENT 118 Chemical Engineering at Tennessee Technological University Joseph J. Biernacki, with faculty and staff RANDOM THOUGHTS 139 How To Write Anything Richard M. Felder, Rebecca Brent OUTREACH 125 Finger Kits: An Interactive Demonstration of Biomaterials and Engineering for Elementary School Students Heather E. Canavan, Michael Stanton, Kaori Lpez, Catherine Grubin, and Daniel J. Graham CURRICULUM 141 Incorporating Risk Assessment and Inherently Safer Design Practices into Chemical Engineering Education Jeffrey R. Seay and Mario R. Eden LABORATORY 154 Mixing Hot and Cold Water Streams at a T-junction David Sharp, Mingqian Zhang, Zhenghe Xu, Jim Ryan, Sieghard Wanke, and Artin Afacan 147 A Lab Experiment to Introduce Gas/Liquid Solubility I.M.A. Fonseca, J.P.B. Almeida, and H.C. Fachada CLASS AND HOME PROBLEMS 132 Geothermal Cogeneration: Icelands Nesjavellir Power Plant Edward M. Rosen

PAGE 2

Chemical Engineering Education 118 T ennessee Technological University (Tennessee Tech) began its life as Dixie College in 1909, with a few small but elegant Georgian buildings on 7th Street and Dixie Avenue in Cookeville, Tennessee. The tiny, thenprivate college* evolved into Tennessees only technologi cal university with a strong engineering, science, business, and education emphasis. By the end of the 1940s the seed for what would eventually become chemical engineering was planted within the Department of Chemistry, but the sapling soon withered and became dormant for another two and a half decades. In 1966, a young man named John C. McGee a Ph.D. from North Carolina State University, was hired. Kindled by the presence of a growing chemical indus try in Tennessee, McGee and a colleague from the already became the Chemical Engineering Department and McGee W.D. (Denny) Holland (Ph.D., Georgia Tech) was hired, followed by David W. Yarbrough (Ph.D., Georgia Tech) and Clayton P. Kerr (Ph.D., Louisiana State University). These four men, young and energetic, would build the department strong unit-operations tradition, construct a lasting laboratory infrastructure, cultivate the masters program, embrace the college-level Ph.D. when introduced in 1971, pioneer compu tational techniques that were budding ideas at the timeand become respected and dedicated educators, researchers, and friends for the next three-plus decades, thus sowing the seeds that would grow into the present-day Department of Chemi cal Engineering. While Professors McGee, Holland, Yarbrough, and Kerr would, from time to time, be joined by other faculty, they alone would remain for a lifes career at Tennessee Tech. These historical notes are brief yet important. The present faculty acknowledges and owes much of the departments ongoing success to the foundation that Professors Emeriti McGee, Holland, Yarbrough, and Kerr established. In 1999, both McGee and Holland retired, followed by Yarbrough and Kerr in 2001 and 2002, respectively. Since then, the depart ments faculty has been re-created, ushering in a new wave of excitement and productivity. Much as the departments founders laid the cornerstones and built a tradition and legacy, the present faculty has initiated a renaissance: introducing new research thrusts, modernizing both the undergraduate and graduate curricula in content and pedagogy, and stepping forward in service to regional outreach and their respective professional communities. The remainder of this article deals with the present, and to some extent, a vision for the future. ChE at Tennessee Technological UniversityJOSEPH J. BIERNACKI, WITH FACULTY AND STAFF Dixie College, formally Dixie University, was founded by the Church of Christ in 1909. ChE department The clock tower at Derryberry Hall. Copyright ChE Division of ASEE 2008

PAGE 3

Vol. 42, No. 3, Summer 2008 119 PLAYING LIKE A TEAMINSIGHTS FROM OUR VISITING SPEAKERS Each academic term the department hosts what has become a model research seminar series within the university, bring ing in eight to 10 regional and national speakers. If you have been one of our speakers, you know that this will not be a day of rest for you. A full day of interaction with the students and faculty will be carefully planned and integrated with your seminar and, in the end, you will likely know who we are and we will know something about you. Consistently and almost universally, at the end of the day our guests tell us that the single most striking characteristic of our department is how the faculty clearly demonstrates collegiality, teamwork, and a sense of scholarly community. This is a posture that we cultivate and strive to perfect. Pedro Arce the departments chairperson, is a well-known educator, having developed several strategies for teaching and learning including The Coach Model. [1, 2] It is no coincidence that we view the department as a team. Arce mentors the junior faculty, gives them opportunity to grow as team members, listens to his players, is sensitive to the environment of the game and, in the end, lets his team play. Formally, one might call our management structure distrib uted and unoriginal. Faculty empowerment is the factor that changes the equation. Our faculty members represent the department and carry son and each other. Each faculty member is empowered to play his/her position and to pass or shoot when he or she sees the opportunity. team has two additional individuals without whom the game could not be played: Rebecca (Becky) Asher, our depart Perry Melton our laboratory and machine shop technician. While Asher is holding the department together by responding to the many requests of the faculty and day-to-day student needs, Melton is keeping the unit operations running and working with students to build equipment that they design, as well as helping the faculty and graduate students with research labs.EDUCA TIONAL OBJECTIVESINSIGHTS FROM OUR CONSTITUENTS As part of our recent Southern Association of Colleges and Schools (SACS) accreditation, Tennessee Tech has a newly established Quality Enhancement Program (QEP) and, accordingly, a QEP Committee. This committee surveyed the constituency of the universityour students, alumni, employers of our students, and the facultyand found that unilaterally, what really counts are critical thinking skills and the ability to solve real-world problems. [3] When it came time to review our departmental Program Educational Objectives (PEOs) it was simple: We would integrate critical thinking and real-world problem solving in some way, and we would write PEOs that were timeless. The result is our present statement of PEOs that are our driv ing force and motivation: will collectively exhibit the following traits: be critical thinkers be real-world problem solvers have continued their formal education be working at the frontiers in chemical engineering In addition to real-world problem solving and critical think ing, we have chosen to explicitly call for the continuation of formal education and working at the frontiers in chemical engineering. These two objectives complete the characteristics of our program: an environment that empowers students to take responsibility for their own learning (lifelong learning) wherein research (the frontier) is highly integrated with, and pushes the boundaries of, undergraduate educationmaking it compatible and forming a continuum with graduate studies. Technically, the department uses a transformational-based manage rial structure with a strong team-based component. Yearbook photos (ca. 1976) of (clockwise from upper left) early faculty W. Denny Holland, John C. McGee, Clayton P. Kerr, and David W. Yarbrough, superimposed on a picture of Prescott Hall, home of the Department of Chemical Engineering since 1971. Several ChE faculty members have received seed grants from the University-wide effort to integrate critical thinking and real-world problem solving activities across campus.

PAGE 4

Chemical Engineering Education 120 THE F ACUL TYINSIGHTS FROM OUR AMBIDEXTROUS SCHOLARS If you ask any one of us what characterizes the departmental faculty best, we will tell you it is balance. We strive to create an effective balance between research and education so that students are exposed to an environment that maximizes their learning, and we do it across the department as a pervasive way of being. Such balance between excellence in teaching and research is taken seriously, and our faculty members demonstrate this characteristic by being active and visible in both arenasendeavoring to integrate research and educa tion in unique ways and to create new paradigms for student achievement. Professor Donald Visco for example, has re ceived both a Presidential Early Career Award for Scientists and Engineers (PECASE) for his research on solving inverse design problems, and the American Society for Engineering Education ChE Divisions Ray W. Fahien Award for his vision and contribution to chemical engineering educa tion. So, what are the character istics that lead to a balanced faculty member? The an swer, apparently, is that many boundary conditions can lead to similar outcomes. Chairperson and Profes sor Pedro Arce was born in Argentina and received his undergraduate education in his homelands practiceoriented engineering system at the Universidad Nacional del Litoral (Santa Fe). He started his transformation as a member of the presti gious National Council of Research (CONICET) at one of the leading research and development institutes (INTEC) of Argentina before coming to the United States in 1983. Arces transforma tion was completed by the great research and education scholars at Purdue Univer sity, fusingas his mentors havethe desire to achieve the perfect balance between the two ideologies. Professor Biernacki is the undergraduate product of Case Western Reserve Universitys research-driven program of the late s and the more applied Doctor of Engineering (DRE) program at Cleveland State University. He has 15 years of industrial experience, yet retains a fundamental approach to his research and says that Teaching is a performance, and I simply love the audience, the stage, and the script. Assistant Professor Ileana Carpen her third year at Tennessee Tech, represents a great milestone for the department. With a B.S. from Stanford, a Ph.D. from Caltech, and a post-doctoral appointment at the University of Twente, this faculty member demonstrates that it is possible program because it places an equal emphasis on both educa tion and scholarly research. Assistant Professor Holly Stretz is the rarest of alla high school teacher turned Ph.D. chemical engineer. Stretz has a TABLE 1 Various Awards and Honors of the Faculty Educational Scholarship and Related Service 2008 Outstanding Teaching Award, ASEE Southeastern Section 2008, 2007 Outstanding Faculty Award for Teaching, Tennessee Tech 2008, 2007, 2006 Tennessee Tech College of Engineering Brown-Henderson Award, for outstanding Engineering faculty 2008, [9] 2006 [12] ASEE, Southeast Section, Thomas C. Evans Award, for best paper 2007 Outstanding Campus Representative, Zone 2, ASEE 2007 Outstanding Campus Representative ASEE Southeastern Section 2006 [12] ASEE Corcoran Award, for best paper 2006 Quality Enhancement Program Award for Innovative Instruction, Tennessee Tech 2006 [13] Annals of Research in Engineering Education, invited feature article 2006 Ray E. Fahien Award, ASEE 2004 Outstanding Campus Representative, 2nd Place, ASEE Southeastern Section 2001 ASEE Membership Award, ASEE Southeastern Section Research Scholarship and Related Service 2007 Distinguished Faculty Fellow, Tennessee Tech 2007 American Concrete Institute (ACI), Fellow 2007 2007 Oronzio De Nora Postdoctoral Fellowship, The Electrochemical Society 2006 Invited Visiting Professor, University of Wollongong, Australia 2006 Oronzio De Nora Young Author Award, International Society of Electrochemistry 2006, 2005, 2002 Leighton E. Sissom Innovation and Creativity Award, Tennessee Tech 2006, 2005 College of Engineering Deans Advisory Board Award for Excellence 2005 New Faculty Research Award (1st Place), ASEE Southeastern Section 2004 Presidential Early Career Award for Scientists and Engineers, DOE 2004 NNSA DOE-DP Early Career Scientist and Engineer Award, DOE 2003 New Faculty Research Award (2nd Place), ASEE Southeastern Section 2003 Outstanding Faculty Award for Professional Service, Tennessee Tech 2002 [14] Sigma Xi Research Award, Tennessee Tech 2002, [15] 2000 [16] Kinslow Engineering Research Award, for best paper, Tennessee Tech

PAGE 5

Vol. 42, No. 3, Summer 2008 121 B.S. degree in chemistry from Texas A&M, worked in the high school in the public school system, and studied in the world-renowned research laboratories of Don Paul [4] at the University of Texas, Austin. Associate Professor Venkat Subramanian came to the United States with an undergraduate degree from Indias distinguished Central Electrochemical Research Institute, to join Ralph Whites group at the University of South Carolina. Subramanian merges his passion for applied mathematics with electrochemistry and likely has one of the most productive research groups on campus. He is presently digesting his work into a text for undergraduate and graduate students. Professor Visco, mentioned previously, studied at the Uni versity at Buffalo, State University of New York. He tutored as an undergraduate student, an experience that he reports would ultimately shape his career. Visco had an extended industrial internship, served in the U.S. Navy, and is now ments undergraduate program coordinator for the past four yearsweaving the fabric of our curriculum and creating new and exciting opportunities in undergraduate chemical and biomolecular engineering. The department is also home to two other faculty members. Adjunct Assistant Professor Mario Oyanader (originally from Chile) is both an outstanding researcher and rising young educator. Trained in Arces own group at Florida State University (where Arce taught prior to becoming our chair), Oyanader has the characteristics of both a scholar and an educator who is all about critical thinking. He is leading the renaissance effort in process design and helping to re-cast the Unit Operations Laboratory role within the new integrated curriculum. Research Assistant Professor Vijayasekaran (Vijay) Boovaragavan joined Subramanians group as a post-doctoral researcher and was recently hired to his current position. He received two international competitive distinc tions for his research while at Tennessee Tech and presently co-teaches our Operations course. The ChE Department at Tennessee Tech offers another answer to the old question, Research or education? Schol arship in both is achievable, although a balance must be ac cepted. Collectively, the Tennessee Tech chemical engineering faculty have earned 35 awards and distinctions since 2000 (see Table 1). Many of these are top Tennessee Tech honors, oth ers are national recognitions, and yet others are international distinctionsthe sum of which paint a picture of balance, combining elements of research scholarship and excellence in education. This does not happen by chance; emphasis must be placed on maintaining a rational balance. Furthermore, our experience is that an undergraduate programs excellence is enhanced by strengthening the graduate program. The adage one cannot have a strong graduate program if too much at tention is paid to undergraduate education is simply contrary to the Tennessee Tech experience. STUDENT -CENTERED LEARNINGINSIGHTS FROM HOW OUR STUDENTS LEARN It is well accepted that students learn best by doing, [5] and while there are a range of learning styles, most education researchers would agree that active participation is key to retention and ultimate internalization of learned informa tion. [6] Furthermore, most would also agree that educators must relinquish their tradition of teacher-centered control and place control in the hands of the students who must be empowered to learn, [7] i.e. we must provide student-centered learning environments. [8] These pedagogical principles guide the many changes that we are presently implementing across the curriculum. Active learning is one of several guiding principles being The faculty and staff of the Department of Chemical Engineer ing, from left to right, front row: Ileana Carpen, Rebecca Asher, and Holly Stretz; back row: Don Visco, Vijay Boovaragavan, Perry Melton, Venkat Subramanian, Pedro Arce, Joe Biernacki, and Mario Oyanader.

PAGE 6

Chemical Engineering Education 122 advocated at Tennessee Tech on a universitywide basis. The Department of Chemical Engineering, however, has adopted active and collaborative (team-based) learning at large, and all of us are using some form of these approaches in our classes. The most ambitious of these efforts is growing out of our laboratory-and-lecture integration initiative. Our depart mental founders established and passed on what we refer to as a laboratory-oriented tradition. Four laboratory courses, the one-hour sophomore lab, the two-hour junior lab, and lecture courses across the curriculum. The remaining onehour, second-semester senior lab was preserved for what is now called Capstone Lab. This bold move was met with some effort of the faculty to embrace this initiative has shaped and Motivated by the observation that students are unable to independently synthesize theory, computation practices, and the real world, Biernacki proposed to adopt the lab-inte gration concept across the entire ChE curriculum in 2005. Having pioneered small-scale labs and demonstrations as a regular part of some of their courses already ( e.g. courses on momentum and heat transfer, reaction engineering, thermody namics, and process controls), the faculty team agreed that it could be done on a larger, curriculumwide scale. Thanks to the energetic leadership of Undergraduate Program Coordinator six courses: Introduction to Mass Balances, all three transfer science courses, Chemical Kinetics, and Solution Thermody namics/Separations. Newly named Curriculum Coordinator Stretz is working diligently with Oyanader and other faculty to re-create the role of the unit operations laboratory (UOL) within the new curriculum. Figure 1 illustrates the concept. Here, Level 1 experiences are related to our integrated labs. These integrated activities draw from a variety of physi cal resources including a growing lab tool-kit, the existing UOL, and New Frontiers chemical engineering stations. By the time students reach the Capstone experience (Level 2), they have numerous examples of how theory, computation, and observation (experimentation) work together, thus they are prepared to transition to a more independent, open-ended capstone experience. intended to be a course with a lab section ( i.e. a lecture with a lab). The single most important outcome of the lab-lecture integration is to break down the barriers between theory, computational practices, and the real-world. In Arces view, [7] Traditional lectures give way to integrated environments with a seamless transition from class to lab where the student is in the learning drivers seat. The approach effectively uses these three key elements of engineering education (classroom, simulation, and lab activities) to create a continuum that is seamlessly woven together, each providing input for valida tion of the other, and all working together. Certainly, some individuals are gifted experimentalists and others, theoreti cians, yet the broader general education of the undergradu The 2005 National Cham pion AIChE Chem-E-Car Team from Tennessee Tech, from left to right: Jonathan Phillips, Braxton Sluder, Jason Miller, Regan Chandler, Robert Phillips, Jennifer Pascal, and Haley Hunter. Figure 1. The integrated lab-lecture (ILL) concept.

PAGE 7

Vol. 42, No. 3, Summer 2008 123 ate must include a continuum that explores the relationship between concepts (theory), calculations (the computer), and the behavior of real things (the real world). THE STUDENT CONTINUUM, RESEARCH AND EDUCA TIONINSIGHTS FROM LESSONS IN SCALING Arce was recently a recipient of the 2008 ASEE-SE Sections Thomas C. Evans Award for best paper published in engineering education during 2007. Arce is the only threetime winner of this prestigious award, this time sharing the honor with two co-authors, departmental adjunct faculty member Mario Oyanader, and Steven Whitaker of University of California at Davis, for their paper entitled, The Catalytic Pellet: A Rich Prototype for Up-Scaling. [9] In this paper, Arce, Oyanader, and Whitaker explain that the traditional chemical engineering curriculum and, for that matter, the traditional engineering curriculum, attempts to teach designa study creating a sort of step function in design content of the traditional curriculum. This approach effectively expects that students will synthesize everything they have learned in the past three years during two semesters, thereby transform ing them into engineers. This approach, the paper argues, advocated design across the curriculum, and similar pro grams, with some success. The papers concept is to exploit distributed laboratory courses and use real-world activities ( e.g. the experimental prototype [7, 10] ), thereby introducing concepts of scale and effectively scaling up the knowledge of students as they move through the curriculuminstead of trying to accomplish the same all in one year. The approach introduces a progressive type of curriculum that will require new didactic materials ( e.g. textbooks, simulations) and a new vertical integration of the curriculum that, for Transport Phenomena, is already in place at Tennessee Tech. We take the concept even further in the department and, in fact, view the student body, both graduate students and undergraduates, as a continuum in the lifelong-learning process. Barriers among student groupsjuniors, seniors, masters degree-seeking graduate students, Ph.D.-seeking students, etc.are considered obstacles to learning, growth, and scholarly productivity. While there are a number of effec tive tools that can be used to unify the student body, we feel that research is by far the most productive and learning-rich vehicle. Research not only promotes critical thinking and facilitates learning development, it also promotes the idea of a community of learners among undergraduates, gradu ates, postdoctoral students, and faculty. Although not every undergraduate will engage (formally) in research, we feel that ates every classroom/lab and catalyzes as many elements of the curriculum as possible, facilitating student learning and critical thinking. To this end we have initiated a number of integrating elements to our curriculum, and we advocate that students engage in them along with university-led research activities. While there are numerous aspects of the program that em phasize integration, several merit special mention in addition to the lab-lecture integration: 1) the Distinction in the Major (DITM) option, a formal ized and intensive undergraduate research track that leads to a written and oral thesis defense; 2) the Research Seminar Series, [11] above; 3) the Chemical Engineering Graduate Research As sociation (CEGRA), a student-governed organization that serves the needs of the graduate student body. We feel these aspects are critical to the departmental suc cess. In combination with pedagogy that empowers students to take charge of their own learning, we hope these aspects will create a culture of scholarship emphasizing critical thinking, problem solving, lifelong learning, and extending the frontiers of knowledge. Our faculty, although relatively small compared to others in the Southeast, is able to offer Tennessee Techs students a cross-section of frontier areas in which to work. Arce and Oyanader are interested in electric e.g. electrokinetic hydrodynamics, corona discharge processing. Biernackis main research focus is experimental reaction kinetics, most recently emphasizing portland cement-based materials. Carpens interests focus and polymer composites) and on biomedical systems (tumor growth and tissue engineering). Stretzs efforts focus on the experimental characterization of nano-particle polymer composite behavior, and nano-particle ordering and high temperature behavior of similar materials. Subramanians computational research brings together electrochemistry, com plex transport modeling, applied mathematics, and computer for real-time batteries and fuel cell performance prediction, system optimization, and control. Subramanian is joined by Boovaragavan, who is presently funded by a prestigious in ternational grant award from the The Electrochemical Society. Finally, Viscos work involves a spectrum of computational thermochemistry and molecular design initiatives as well as laboratory research on phase equilibrium. Collectively the faculty are or have been funded by numerous government and private-sector organizations working closely with Tennessee Techs three state-funded research centers, and have many ongoing or prior research collaborations. Such collabora tions enhance opportunities for our students and faculty to The concept parallels the High Performance Learning Environment (HiPeLE) developed by Arce and collaborators, 2004, CEE.

PAGE 8

Chemical Engineering Education 124 infrastructure in the world. undergraduates have amassed top honors in numerous com petitions: Our Chem-E-Car team has consistently been in the vehicles, having established the current national record second at the AIChE Southeast Regional Meeting this in the competition. Collectively, our students have received top paper and poster awards at regional and national competitions. In 2006, Jennifer Pascal ates and a current Ph.D. student) received the AIChE Othmer Award, and Hope Sedrick was selected as one of 10 students to receive a NIST SURF internship. Recent graduate students have performed similarly well. Ph.D. student Vinten Diwakar received The Electro chemical Societys Industrial Electrolysis and Electro chemical Engineering Division Student Achievement award in 2006 for his research on battery and fuel cell modeling, (Subramanian, his Ph.D. advisor, was also a recipient of this prestigious award as a student). Ph.D. student Pravin Kanan was selected for the BASF International Summer Course in Bohn, Germany, in recognition of his research on polystyrene foam thermal decomposition. Masters student John M. Richardson received an NSF Graduate Research Fellowship in 2002 for his research on nano-pore structure of hydrated portland cement, a In addition, our B.S. and M.S. graduates are being hired by leading companies. Recent Ph.D. graduate Dr. Baburao was highly sought by design companies and Dr. Swaminathan is currently a post-doctoral research associate at the Technical University of Denmark. The list of similar awards and achievements is long, but these illustrate the excitement and success among our students and, we believe, the result that is achievable with a program that focuses on scholarship in both education and research through integration rather than separation. ABOUT TOMORROWINSIGHTS FROM OUR VISION OF THE FUTURE It seems appropriate to dream and to speculate just a bit at this point in the departments story. We recently revised our vision statement as well as our PEOs, and after considerable debate over two wordswill bewe were convinced by our Board of Advisors to phrase our vision statement in the sion now reads as follows: The Department of Chemical Engineering is a recognized leader in chemical engineering education through excellence in teaching, research, and service. This statement is a vision of the future for ussimple, yet a bold supposition that we believe will be. The path from here to there for us is clear: (1) continue to respect and build upon the foundations of our legacy; (2) develop and grow a faculty that plays like a team; (3) have clearly stated educational objectives that are simple and timeless; (4) main tain a balance between research scholarship and education, and strive to excel in both; (5) fervently maintain a studentcentered learning environment; and (6) integrate the student body, integrate research with education, and do not let size scale of our productivity.REFERENCES 1. Arce, P.E., Colloquial and Coach Approach Environments in Engi Proceeding of the 8th Latin American Congress on Heat/Mass Transfer LATCYM 2001, 551-554, Puerto Veracruz, Mexico, February 2001 2. Arce-Trigatti, M.P., and P.E. Arce, Parallel Between Active Learn ing and Coaching Team Sport Techniques: Analysis and Selected Examples, Annual Conference Proceedings, American Society for Engineering Education (CD-ROM), June 2000 3. , slide #15 of Tennessee Technological University Quality Enhancement Plan 4. Koros, W.J., Don Paul of the University of Texas at Austin, Chem. Eng. Ed. 35 (2) 86 (2001) 5. Cardellini, L., An Interview with Richard M. Felder, J. Science Ed. 3 (2), 62 (2002). 6. Johnson, D.W., R.T. Johnson, and K.A. Smith, Active Learning: Co operation in the College Classroom Interaction Books, Edina, MN (1991) 7. Arce, P.E., and L.B. Schreiber, High-Performance Learning Environ ments, Chem. Eng. Ed. 38 (4) 286 (2004) 8. Creighton, L., Kicking Old Habits, ASEE Prism p. 33, April (2001) 9. Arce, P, M.A. Oyanader, and S. Whitaker, The Catalytic Pellet: A Rich Prototype for Up-Scaling, Chem. Eng. Ed. 41 (3), 187 (2007) and Plenary Lecture at the ASEE SE Section Annual Meeting, the University of Memphis, April 6, 2008. 10. Arce, P., J.J. Biernacki, and P. Melton, The Experimental Prototype: Critical Thinking and Real-World Problem Solving in Engineering Education, presented at the AIChE Annual Meeting, San Francisco, November 2006 11. Tennessee Tech Department of Chemical Engineering, Research Seminar Series, 12. Biernacki, J.J., A Course-Level Strategy for Continuous Improve ment, Chem. Eng. Ed. 39 (3) 186 (2005) and Plenary Lecture at the ASEE SE Section Annual Meeting, the University of Alabama, April 2006. 13. 14. Visco, D.P. Jr., R.S. Pophale, M.D. Rintoul, and J.L. Faulon, Devel oping a Methodology for an Inverse Quantitative Structure-Activity Relationship Using the Signature Molecular Descriptor, J. Molecular Graphics and Modelling 20 429 (2002) 15. Biernacki, J.J., P. Stutzman, and P.J. Williams, Kinetics of the Reac tion Between Fly Ash and Calcium Hydroxide, ACI Mat. J. 98 (4), 340 (2001) 16. Visco, D.P. Jr., and D.A. Kofke, Modeling the Monte Carlo Simulation of Associating Fluids, J. Chem. Phys. 110 5493 (1999)

PAGE 9

Vol. 42, No. 3, Summer 2008 F rom personal computers to cell phones, there is no dispute that technology developed by scientists and and engineering have on our lives, however, the numbers of undergraduate and graduate degrees awarded in these areas to U.S. students have decreased steadily for the past several decades. [1] In addition, the numbers of women and minorities ics ( e.g., New Mexico is now a minority-majority state). [2] For these reasons, many efforts have emerged with the goal of attracting students into engineering and science disciplines, including outreach efforts such as those sponsored by the National Science Foundation. The authors of the present work include researchers from both the University of New Mexico (UNM) and the University of Washington (UW). The technical expertise of the authors materials with biological systems. The term biomateri als, therefore, encompasses a number of research interests including microbially induced corrosion of ship hulls, the development of DNA microarrays, and the optimization of materials used for biological implants. While it is unlikely lowing demonstration) are familiar with DNA microarrays, FINGER KITS: HEATHER E. CANAVAN, MICHAEL STANTON* University of New Mexico Albuquerque, NMKAORI LPEZ Albuquerque Public Schools Albuquerque, NMCATHERINE GRUBIN University of Washington Seattle, WADANIEL J. GRAHAM Asemblon Corporation Seattle, WA Copyright ChE Division of ASEE 2008 ChE outreach

PAGE 10

Chemical Engineering Education 126 many have already been exposed to implanted materials, as they or members of their families may have contact lenses, use a glucose monitor for their diabetes, or have hip or other implants. Therefore, giving students a project that emphasizes biomaterials taps into something with which they already may be familiar and that is becoming increasingly important in the lives of millions of people around the world. As noted by Tobias, [3] tying science to societal issues via cooperative and interactive learning styles may increase participation by women and other under-represented minorities in science and engineering. Furthermore, as more people are going to be affected by biomaterials, there will be more opportunities has been estimated that by the end of this decade, there will have been more than a 30% increase in bioengineering-related employment positions. [4] To capture part of this excitement, we present a real-world and the students in the class need to design an implant to replace it. After presenting the problem, we discuss how joint. In order to do this, the students need to understand what materials are available that match the properties of the e.g., that an object). At UNM, we made a number of adaptations to the original design. For instance, to better communicate with the bilingual and monolingual (Spanish-only speaking) students within the Albuquerque Public School (APS) district, we translated the text used in our brief presentation into Span ish and recruited Spanish-speaking volunteers. Furthermore, our par ticipating volunteers are typically a combination of undergraduates, graduate students, postdocs and faculty to show representatives of women and minorities that have gone on to successful engineering careers. This hands-on activity en gages the students creativity while also teaching them a basic under standing about what biomaterials are and how one would go about designing and building them. BEFORE THE VISIT Prior to any outreach events, the following should be addressed: 1) provisions of key ideas and vocabu lary necessary to understand the les son, 2) assembly of kits, and 3) training of volunteers. Stanton), when scheduling outreach visits, provides teachers and principals with vocabulary words and concepts necessary to understand the lesson. Table 1 presents the key vocabulary that the outreach director (co-author Canavan) and elementary UNM visit to Lpezs classroom. These vocabulary words are discussed. In addition, each of the visits is scheduled such that the parts of the body especially important when considering comprehensive list ( e.g., body parts that make primary contributions to the function of the joint, which is the focus of the lesson. For example, it is the contraction and expansion of the muscles that lead to joint movement, and bone which provides structural support. Table 3 lists the contents to be included in each kit. No additional supplies are required to perform this activity. For each of the suggested materials ( e.g., chalk), a potential use e.g., bone). Therefore, each of the materials listed in Table 3 should approximate those of the human body listed in Table 2. It is important to note that we provide this list to aid volunteers using this demon stration, but we do not provide it to the students themselves. provided that we had not initially envisioned ( e.g. the use T ABLE 1 Vocabulary given to 5th grade teachers prior to demonstration to prepare for visit. Vocabulary word Biology The study of living organisms. Engineer (noun) A person who designs, builds, or maintains engines or machines. Implant Material The substance of which a thing is made, such as wood, glass, or metal. Prosthetic T ABLE 2 Parts of the body taught in 5th grade module prior to visit by UNM. Additional body parts may be added by students ( e.g., Part of the Finger Function Arteries and veins Bone Structural support, mechanical strength Muscles Contract and extend to move joints Nerves Sensation of heat/cold, movement of body via attachment to muscles Skin External surface of body Tendons Attach muscles to bone

PAGE 11

Vol. 42, No. 3, Summer 2008 127 more pleasing, instead of as a structural element to hold the design together). Although most of the parts can be re-used ( e.g., pipe cleaners, straw, etc.) and the kits recycled through many events, at UNM each kit is used only once, and the students retain possession of their designs. Therefore, to maximize the number of students we can interact with over that can be purchased in bulk ( e.g., Popsicle sticks vs. tongue depressors). We estimated that, when the contents of the kits are purchased in bulk ( e.g., to make 400 kits), the cost of the kits fall to ~$1/student. It is advisable to train all volunteers prior to the outreach season so they understand what to expect from the events. In particular, the volunteers should understand how the ma terials in the kit can be used in designs. Such an event will also yield a number of diverse designs (as illustrated in Figure 1), demonstrating to the volunteers that they may see many different ways the materials will be used by the elementary school students. Also, the volunteers should be briefed about what to expect from a visit to an elementary school, including any necessary information about the schools dress code. If possible, the teachers participating in outreach visits should be invited to speak with the volunteers about the general level of knowledge and understanding, as well as modes of learning, that young students demonstrate. DURING THE VISIT The outreach visit contains several elements: 1) a brief pre sentation outlining the topic and project (~15-20 minutes); 2) discussion and formulation of the design and test parameters (~2-5 minutes); 3) fabrication of the designs by the students (~30 minutes); and 4) evaluation of the design according to the parameters previously outlined (~10 minutes). In addition, improvements to be made as needed. At the beginning of the visit, the lead volunteer will give a brief talk [5] to introduce the range of topics in bioengineer T ABLE 3 Contents of the Finger Kit. Note that all materials are meant to be commonly commercially available and of low cost. Substitutions, deletions, and additions to the kit may be made to accommodate the preferences of the demonstrators, as well as the availability of material. Qty Item Potential Use in Finger Design 1 Zip-close sandwich bag None (used to contain other parts listed below) 3 enough for dowel pieces/chalk to slip into Skin that holds all parts together 1 Tongue depressor/Popsicle stick 4 Toothpicks 1 Pipe cleaner Actuator (muscle/tendon) or body part (blood vessel/nerve) 1 Piece of copper wire 10 long (~24 gauge) Actuator (muscle/tendon) or body part (blood vessel/nerve) (muscle/tendon) or body part (blood vessel/nerve) 1 Flexible straw Actuator (muscle/tendon) or body part (blood vessel/nerve) 4 Rubber bands Actuator (muscle/tendon), body part (blood vessel/nerve), or to hold design together (tendons) 3 Wooden dowels, ~1.5 long x 0.25 diameter Bones 3 Half-pieces of chalk Bones 2 Small paper clips 2 Large paper clips a) b) c) d) Figure 1. Examples of nished nger designs. Note that the resulting design will reect the design consid erations agreed to by the students at the beginning of class, as well as their own individual preferences. For example, design a) uses a Popsicle stick as a nger nail, while design d) uses it as a brace to prevent backward movement.

PAGE 12

Chemical Engineering Education 128 ing, including examples of biochemical engineering ( e.g., pharmaceuticals and dialysis), biomechanics ( e.g., grafting procedures, prosthetics, and implants), and biomaterials ( e.g., to the students using the vocabulary list (see Table 1), are used to lay the groundwork for the design students will be performing. Therefore, it is imperative they understand what grade students are (at best) unfamiliar with even the most basic language traditionally used in bioengineering, and often hold erroneous beliefs (such that the term engineer solely applies to people who repair car engines). We have found that the best way to engage the students attention and get these concepts across is to make the talk highly interactive, with the lead volunteer asking the students questions and listening to their responses. For instance, the students become quite animated when asked: Do you know anyone who has contact lenses or hearing aids? Such ques tions solidify ideas in the students minds, as they are able to make connections between the new material and something they are already familiar with. In fact, according to elementary school teacher and co-author Lpez, the most important factor for a successful demonstration is to pay attention to cues from the students to determine if they are understanding the mate rial, and therefore connecting bioengineering to their lives. Later in the talk, the materials used in biomaterials are dis function. For instance, metal implants are often used in bone replacements (due to their mechanical strength), whereas the idea that a design may be perfected over time (such as the early use of ear trumpets prior to the invention of a hearing aid) can show students that rarely is a design perfect from Finally, we use the talk as a chance to educate the students about the career path that bioengineers take, from their cur rent position to college and graduate school. Many of our participating volunteers are from the local community (20% of UNM volunteers attended APS as students, and 78% of APS participants attended APS as students). As importantly, many volunteers are members of groups traditionally underrepresented in science and engineering (45% are women, and 50% identify themselves as Latino, Hispanic, or Chicano) and are Spanish speakers (35%). At the conclusion of the talk, the students are told they will not have to wait until college or graduate school to start their research career, and that they will be bioengineers for a day. and will test their designs according to parameters they agree upon. This leads to a discussion to elicit design goals from the students themselves. For instance, the volunteers may they should bend? [In one direction only.] What could you do to allow them to bend, but in just one direction? [This is where students tend to start equating the materials to what the students to come up with their own rubric for a good design is far more engaging than providing these parameters directly. In addition, this method requires higher-order think ing skills and often causes a great deal of excitement among the students. Figure 2 shows a page from a students notebook should have, as well as the parts of the body that should be included (mitiriols [ sic ]). Once the design parameters have been agreed upon, the volunteers distribute the kits to the students, and fabrication of the designs begins. During this portion of the visit, the volunteers and teacher should circulate among the students. While we discourage the volunteers from telling the students how to make the design, they can provide guidance, remind ing students about the parameters ( e.g., but does so in many directions, it hasnt been optimized). If students are stumped, the volunteers can help provide prompts to get the students working on their designs. For instance, Figure 2. Page from a students notebook at Longfellow El ementary. The student has listed the design considerations the students agreed a nger should do, as well as the parts of the body that should be included (materials). She has also illustrated her own nger for the design.

PAGE 13

Vol. 42, No. 3, Summer 2008 129 [Bone.] Is there anything in this kit that reminds you of that part? [Chalk or wooden dowel.] This also gives the students a chance to interact with their potential role models and ask them questions about themselves while they work on the de signs. Furthermore, this gives the volunteers a chance to stress the creative aspects of science and engineering. In addition, the students often begin to observe the designs used by others to overcome a particular challenge ( e.g., the use of toothpicks as braces to prevent their design from bending backward) and incorporate it into their own. Such peer mentoring is a natural occurrence in this environment. Figure 3 illustrates each of these modes of learning during a visit from the UNM Biomaterials Engineering Outreach Program to Longfellow Elementary School. At the end of the visit (or in a post-visit session with their teacher), students evaluate their designs according to the parameters to which they previously agreed. Table 4 is an example of a rubric designed by the students in co-author Lpezs fifth-grade class, including the design parameters (or criterion) con sidered important by the students, as well as their standards for the design. For example, the students considered the ability to hold something up as an important criterion, and a design that is capable of may be considered an advanced design. Often, the students come up with new designs or improve ments not initially listed as criteria in the rubric. For instance, some students will attempt to improve the aesthetics of the design by including a paperclip as remind students that many designs are improved over time for aesthetic reasons ( e.g., less noticeable AFTER THE VISIT of the formal visit, and the students are allowed to keep their designs. As well, some teachers may want to display the top designs or keep them for followup. In most cases, students from the classrooms we visited sent thank you letters to the demonstrators, relating what they learned, how much they enjoyed T ABLE 4 posed and agreed upon by the students at the beginning of the demonstration. Criterion Emergent (1) Advanced (4) Moves like a Doesnt move. Moves somewhat, but in too many directions that are not appropriate. Moves forward. Moves forward, not back wards. Does not move to can move side to side. Has two places where it can bend. Has components Has components that dont function like a not in the right place. Has only two compo nents: possibly bone and muscle or skin and bone. Components may not be in the right place. Has bone, skin, muscles, nents are generally in the correct place. Has bone, skin, muscle, nerves and other sensors. Components are in the cor rect place. Holds something up (e.g., pencil). Doesnt hold it up at all. Barely holds it upfalls quickly. Holds up pencil. Holds up pencil in the a) b) c) d) Figure 3. Examples of the different modes of learning during a visit to fth-grade students at Longfellow Elementary School (Albuquer que, NM). UNM volunteers circulate throughout the classroom giving guidance and providing translation into Spanish, where necessary (a); students often teach each other lessons learned from their designs, a classic example of peer learning (b); a student consults the hand of a skeleton to determine the parts of the nger (c); the students teacher reminds the fth-grade bioengineers that one criterion of their design is that it must bend (d).

PAGE 14

Chemical Engineering Education 130 the visit, and their interest in science. The letters are always appreciated by the volunteers, and help reinforce the value of outreach activities to them. In addition, it is also a chance for the outreach organizers to get valuable feedback. For instance, due to the large number of students who mentioned in their letters how much they wanted to keep their designs (to show their parents or siblings, or to improve on the design), we at UNM now make kits for each student, rather than attempting to recycle them (as had been done previously). EXPECTED IMP ACT As previously stated, one of the primary complaints that elementary school teachers have about outreach projects is that they are often considered stand-alone demonstrations with little thought to how they will be integrated into the regular demonstration addresses a number of benchmarks from the State of New Mexico [6] as well as the Project 2061 Bench marks for Science Literacy (from the American Association for the Advancement of Science, AAAS), after which many states have modeled their standards. [7] The individual bench ing and Practice): By observing and experimenting on their model, and analyzing their product (using the rubric), the class discussion at the end of the event, and learn that their conclusions are subject to peer review. Strand II/Standard I/Benchmarks II & III (Physical Sci project addresses the state benchmark pertaining to forces and motions. While studying the action of muscles and tendons upon an object, it will move in a different direction. Strand II/Standard II/Benchmark III (Life Science): By replacing that function using a prosthetic), the students learn the properties, structures, and processes of living things, and how cells and tissues are related to the behavior of an entire organism. Strand III/Standard I/Benchmark I (Science and Society): With its emphasis on how machines have been engineered to aid in human health ( e.g., glucose monitors, hearing aids, etc.), the introductory talk demonstrates to students how technology has affected the lives of individuals. Emphasizing how rudimentary prosthetics have evolved allows students to Furthermore, in an integrated elementary curriculum, the progress of bioengineers could be connected to the effect on social issues. Developing the rubric to evaluate their designs allows students ent forms, including observing what things are like or what is happening somewhere, collecting specimens for analysis, and doing experiments. Enterprise): Discussing their results reinforces to students that clearly communicating their results is an essential part of doing science. Because the volunteers participating in this event are both men and women of many different ages and backgrounds, it is reinforced that people who perform Benchmark 3A (The Nature of Technology/Technology and students learn that technology extends the ability of people to change the world, often in response to the need to meet basic survival needs. Benchmark 3B and C (The Nature of Technology/Design Constraints and Systems and Issues in Technology): While building their designs, the students rapidly grasp the need to match properties of materials and engineering principles while designing solutions to problems. Simultaneously, the design with the best side-to-side stability may aesthetically be the least pleasing. Benchmark 6C (The Human Organism/Basic Functions): As they learn how nerves stimulate the muscles in a joint to contract, the students learn how the human body functions as a systemwith the brain giving signals to the body to stimulate movement. Benchmark 8F (The Designed World/Health Technology): By learning about prosthetics and designing a replacement body part, students learn that technology has made it possible the State of New Mexico [6] as well as the Project 2061 Benchmarks for Science Lit eracy (from the American Association for the Advancement of Science, AAAS), after which many states have modeled their standards. [7]

PAGE 15

Vol. 42, No. 3, Summer 2008 131 to repair and sometimes replace some body parts. Benchmark 11A (Common Themes/Systems): By learning about how the materials in their kits (and the parts of the body) work together as a system, students learn that the parts of a are broken, worn out, or misconnected. This is applicable to both their own creation as well as the body part it is meant to replace. Benchmark 12C (Habits of Mind/Manipulation and Obser vation): By asking the students to relate the material properties of the objects in their kits ( e.g., chalk) to those in the human e.g., bone), the students learn to choose appropriate common materials for making simple mechanical construc tions and repairing things. CONCLUSIONS This work describes one hands-on activity and demonstra of the project is to provide a hands-on experience with an engineering project. While the project itself is goal-oriented, it is also creative and open-ended, with many possible so lutions to the problem presented. In this way, the creativ ity involved in the project is emphasized, rather than only relying on science and math ability. In our experience, the demonstration works best when it is tailored to suit the needs of the community, and we recommend that anyone adopt ing this outreach demonstration take the time to do so with their own community. It is for these reasons that we at UNM chose to focus our activities on one grade level (5th grade) of the community ( e.g., bilingual students in NM) should be addressed ( e.g., translation of the slides into Spanish). Finally, we chose simple, low-cost materials to maximize the number of students that can be reached with the activity. Ultimately, we strove to develop a fun experience that will get students excited about career opportunities in science and engineering. After all, although not all students will ultimately pursue science and engineering-related careers, we feel that a general population more educated in the area of science and engineering is also a valuable pursuit. ACKNOWLEDGMENTS This work was supported by NSF-Partnerships for Research and Education in Materials (PREM) Program grant # DMR0611616 to the UNM Biomaterials Engineering Outreach program, and NSF-Engineering Research Center (ERC) Pro gram grant # EEC-9529161 to the University of Washington Engineered Biomaterials Outreach (UWEB) program. We also thank the many volunteers who have participated in outreach events in Washington and New Mexico, including the UNM Student Chapter of the Biomedical Engineering Society. In particular, we thank Danielle Garca, Rosalba Rincn, and Ulises Martinez for assistance translating our presentation into Spanish. REFERENCES 1. Mannix, M., Getting It Right, Prism 10 (7), 14 (2001) 2. Smith, T.Y., Science, Mathematics, Engineering, and Technology Retention Database, Making Strides 2 (2) (2000) 3. Tobias, S., Theyre Not Dumb, Theyre Different: Stalking the Second Tier Tucson, AZ Research Corporation. 70 (1990) 4. Employment Outlook: 2000-2010, U.S. Dept. of Labor, Ed., Monthly Labor Review (2000) 5. UNM Biomaterials Engineering Outreach Materials Online: 6. New Mexico Public Education Department: 7. AAAS Benchmarks Online:

PAGE 16

Chemical Engineering Education 132 E nergy use in Iceland (population 283,000) is higher per capita than in any other country in the world. [1] Some 53.2% of the energy is geothermal, which supplies farms, snow melting, greenhouses, and space heating. The Nesjavellir Power Plant is a major geothermal facility, supplying both electricity and heated water to Reykjavik. The purpose of this paper is to interest students in geothermal energy, describe a simulation of this plant, and determine the plants suitability for classroom study. PLANT DESCRIPTION The plant (commissioned in 1998 [2] ) is located near one of [3] minerals that the waters cannot be used directly in the distri bution system. [4] Its high pressure and thermal energy, how ever, makes it suitable for heating fresh water and generating electricity. Ballzus, et. al., [2] names ( e.g. {S1}, {S2}). In addition, the heat exchangers are labeled ({HX1}, {HX2}, {HX3}). Steam mixed with water {S1} is conveyed from boreholes through collection pipes to the separation station, where the water is separated from the steam. Excess steam and unused water go into a steam exhaust outside the separation station. From the separation station, steam and water proceed by separate pipes to the power plant at a pressure of about 12 through a mist eliminator) is conveyed to steam turbines, where electricity is generated. Each turbine (two of them) produce 30 MW of electricity (MWe). The object of this column is to enhance our readers collections of interesting and novel prob lems in chemical engineering. Problems of the type that can be used to motivate the student by presenting a particular principle in class, or in a new light, or that can be assigned as a novel home problem, are requested, as well as those that are more traditional in nature and that elucidate dif (e-mail: wilkes@umich.edu), Chemical Engineering Department, University of Michigan, Ann Arbor, MI 48109-2136. ChE class and home problemsGEOTHERMAL COGENERA TION: ICELANDS NESJAVELLIR POWER PLANTEDWARD M. ROSEN Copyright ChE Division of ASEE 2008

PAGE 17

Vol. 42, No. 3, Summer 2008 133 Figure 1. The Nesjavellir Geothermal Plant Process Flow Diagram (Adapted from Ballzus [2] ), placed vertically to preserve clarity.

PAGE 18

Chemical Engineering Education 134 Figure 2. VMGSim Flowchart of Icelands Nesjavellir Power Plant (placed vertically to preserve clarity).

PAGE 19

Vol. 42, No. 3, Summer 2008 In the condenser {HX1} the steam exhaust from the tur bines is utilized to preheat cold water {S35}. This cold water is then further heated in the heat exchanger {HX3} by the {HX2} can be utilized to preheat a portion of the cold water simulation, however, {HX2} is not utilized). Since the min exchanger pipes, steel particles are allowed to circulate in the stream, impacting against the pipes to remove any scaling as it occurs. [5] The cold water {S21} is saturated with dissolved oxygen that corrodes steel after being heated. To rid of the oxygen, the water is sent to a vacuum deaerator. [6] {S11} enters the central part of the deaerator. The water boils Steam and gas rise to the top. The steam is condensed through the injection of cold water {S30} before the gas is ejected. Finally, a very small quantity of steam containing acid gases {S37} is mixed with the water to eliminate the last traces of dissolved oxygen and lower the pH of the water in order to prevent precipitation in the distribution system. The following reaction takes place. [7] 2H S (g ) + O (a q) = => 2H O (a q) + 2S ( s) 2 2 2 Small quantities of H 2 S ensure the dissolved oxygen that could get into the storage tanks is eliminated The H 2 S also gives the water the good smell for which the water from the water supply system in Reykjavik is known today. THE VMGSIM SYSTEM [8] The VMGSim system is a modern interactive process simulation system. One of the partners of VMG (Virtual Materials Group) founded Hyprotech and another created and wrote most of HYSIM. As a general policy, VMGSim is provided to universities free of charge when used for aca demic purposes. The system uses Microsoft Visio for the graphical input engine. A menu is provided that allows the user to drag streams and unit operations onto a graphical screen to build a complex system. The system uses the interactive calculation principles of nonsequential unit operation calculations with partial data creating and evaluating process models. Equilibrium stream The physical property system has been carefully crafted simple click of the mouse will allow the user to evaluate dif ferent physical properties for his/her simulation. Similarly, different units (SI, Field, etc) can be implemented with a simple click of the mouse. Custom models can be created using Excel (VBA). T ABLE 1 Composition of Geothermal Fluid Stream {S1} Vapor Fraction 0.3527 Temperature (Deg C) 189.2 Pressure (kPa) 1235 Flow (kg/s) 326 Enthalpy (kJ/kg) 1500 Water (kg/s) 325.56 0.1495 Carbon Dioxide (kg/s) 0.2875 Oxygen (kg/s) 0 Sulfur (kg/s) 0 T ABLE 2 Composition of Cold Water Stream {S21} Vapor Fraction 0 Temperature (Deg C) 5 Pressure (kPa) 101.33 Flow (kg/s) 1129 Enthalpy (kJ/kg) 21.1 Water (kg/s) 1128.985 0 Carbon Dioxide (kg/s) 9.889 E-04 Oxygen (kg/s) 0.0144416 Sulfur (kg/s) 0 T ABLE 3 Pump Pressure P1 82.33 85 P2 150 80 P3 150 75 P4 75 75 Heat Exchangers HX1 30 1 HX2 30 20 HX3 30 20 Valves V1 (Mist Eliminator) 35 V2 to V9 68.94

PAGE 20

Chemical Engineering Education 136 use. It allows very rapid evaluation and optimization of dif ferent cases. PLANT SIMULA TION VMGSim is used to simulate the plant and match the data given in Figure 1. Steam Table is selected as the physical property system. The components in the simulation are: 1. Water 3. Carbon Dioxide 4. Oxygen 5 Sulfur depiction of the process There are two feed streams to the plant. In H 2 S, and CO 2 Table 1 (p. 135) gives the stream composition based on the values of H 2 S and CO 2 (CO 2 2500 ppm, H 2 S 1300 ppm) in the highpressure steam {S2}. [9] The VMGSim system is used to determine (by iteration) the values of the temperature and pressure of {S1} from the composition of the high pressure steam, the enthalpy and vapor fraction (=115/326) of {S1} The cold water (Table 2, p. 135) at 1 atm and 5 dioxide. Values of Henrys Law constants (H) are taken from Perry: [10] H(O 2 ) = 29100 atm/mole fraction (air 20.94% oxygen) H(CO 2 ) = 878 atm/mole fraction (air 0.0314% carbon dioxide) where partial pressure(atm) = H x (mole frac tion) The pressure drops throughout the system are is the pressure drop across the tubine: 12 bara 0.2 bara ). As a result, literature suggestions [11, 12] for pressure drops in the valves and heat ex changers and pressure rises in the pumps (arbi trary) are used as shown in Table 3 (p. 135). Exit streams are assumed to be at about atmospheric pressure. The mist eliminator is simulated as a valve {V1}. The steam turbine {Ex1} is simulated by an expander, and the electrical energy (MWe) is T ABLE 4 Expander {Ex1}, Condenser {Hx1} and {Hx3} Expander 1180 60 82.45 Condenser (Hx1} [14] 50 Heat Exchnger {Hx3} [14] 96.9 Figure 3. Equations to determine deaerator performance.

PAGE 21

Vol. 42, No. 3, Summer 2008 137 as a heat exchanger ({Hx1}) and a separator that purges the noncondensable gases (Table 4) VMGSim does not have a model of a vacuum deaerator. [6] It is simulated, however, with a mixer {M4} and a Compo nent Splitter {CSP1} A determination is made [13] amount of water and O 2 that the deaerator is required to purge 2 {S26} and the H 2 S content of the water delivered to Reykjavik {S28}. (All CO 2 is assumed to go overhead) Figure 3 is a set of eight equations (in eight unknowns), but with just three manipulated variables to achieve three 1. Fraction of high pressure steam that goes to the reac T ABLE 5 Calculated Streams Compared to Reference Number 2 Icelands Nesjavellir Co-Generation Power Plant Stream Description Flow kg/s Temp Deg C Enthalpy kJ/kg Pressure kPa Frac Ref [2] Ref [2] Ref [2] Ref [2] Vapor S1 Geothermal Fluid 326 326 189.2 1500 1500 1235 0.3527 S2 High Pressure Steam 115.13 115 189.2 2775.36 1235 1 S3 Geothermal Fluid 210.86 211 189.2 803.57 1235 0 S4 High Pressure Steam 115.13 188.2 188 2775.96 1200 1200 1 S5 Low Pressure Steam 114.53 115 60 2251.9 20 20 0.8519 S6 Condensate 112.62 56.3 60 235.74 19 0 S9 Warm Water 667 667 55 230.39 221.33 0 S11 Warm Water 667 667 86.4 88 361.96 122.89 0 S14 High Pressure Steam 0 0 S15 Geothermal Fluid 0 0 S20 Geothermal Fluid 323.48 326 79.9 81 334.65 101.33 0 S21 Cold Water 1129 1129 5 21.1 101.33 0 S28 Warm Water 709 709 81.7 83 342.19 118.45 0 S30 Cold Water 42 42 5 21.3 182.39 0 S35 Cold Water 1087 1087 5 21.29 251.33 0 S38 Warm Water 420 420 55 230.39 152.39 0 S54 Geothermal Fluid 219.86 92.3 92 387.41 1146.06 0 Turbine Output: Ref [2]= 60 MWe, Simulation = 60 MWe ; Thermal MWt: Ref [2] = 127 MWt, Simulation = 123.88 MWt T ABLE 6 Distribution of Noncondensable Gases H 2 S CO 2 O 2 In Out In Out In Out Stream kg/h kg/h Stream kg/h kg/h Stream kg/h kg/h S1 543.67 S1 51.99 S20 12.34 S1 1035.09 S25 32.52 S28 2.55 S21 3.56 S32 19.34 S39 528.51 S20 2.75 Reaction 0.13 Reaction 0.27 S25 2.24 S28 5.43 S38 1.32 S39 1026.91 Sum 543.67 543.67 Sum 1038.65 1038.65 Sum 51.99 51.99

PAGE 22

Chemical Engineering Education 138 tor in separator {SP4} 2. The amount of water purged in the deaerator {S25} 3. The amount O 2 purged in the deaerator {S25}. It is assumed all the CO 2 will be purged. 1. The ppb of O 2 in the liquid leaving the deaerator {S26} 50 ppb [5] 2. The ppm of H 2 S in the exit water {S28) 1 ppm [5] The results of the computation are used to enter the fractions overhead into the Component Splitter Block in the VMGSim determined by an enthalpy balance around the deaerator. heat capacity (Out put Temperature 40 ) = 123.88 Table 5 (p. 137) gives the results of the simulation. Numbers in bold are those taken from Figure 1. Other values are results of the simulation. DISCUSSION OF THE SIMULA TION As shown in Table 5 the VMGSim simulation matches the indicated conditions [2] reasonably well. Two important fac tors, however, impact the comparison of the simulation and the data of Figure 1. 1. The plant data of Figure 1 does not indicate any vent ing from the condenser {Sep2} or specify the amount of high pressure steam in stream {S37}. The simula tion calculates both {S37} and {S39}. 2. The plant data of Figure 1 does not indicate any venting from the deaerator. The deaerator vents both water and noncondensable gases. able changes in downstream streams {S5}, {S6}, and {S20}. Similarly, small changes in the concentration of H 2 S in the heated water {S28} greatly affect the amount of water purged in the deaerator. The deaerator design is based on data suggested by an author [5] other than Ballzus. [2] The distribution of the noncondensable gases was not ad dressed in Figure 1 but is discussed by Gislason. [14] Table 6 (p. 137) lists the distribution in this simulation. A comparison 2 S and CO 2 together and indicates different amounts of the noncondens able gases in the entering streams ( {S1} and {S21} ) than used in this study. Also, Gislason does not account for O 2 CONCLUSIONS Study of Icelands Nesjavellir Power Plant appears to be well suited for classroom instruction and inclusion in under graduate energy courses. [15] Such a study illustrates both the advantages of geothermal energy as well as indicating some of its limitations in terms of the suitability and source of Carrying out a simulation draws attention to a variety of energy tradeoff issues, material balance questions, physi cal property estimates, equipment design selection, water chemistry, and environmental control. Interest in geothermal energy generated by this study can be pursued by searching ( e.g., on the Internet) for other ways of using this source of energy. [16] ACKNOWLEDGMENT The author would like to thank Gerald Jacobs of Virtual Materials Group for making the VMGSim system available to the author to conduct this study. REFERENCES 1. Gunnlaugsson, E., A. Ragnarsson, and V. Stefannson, Geothermal Energy in Iceland, International Symposium in Izmir, Turkey 4-5(October 2001) 2. Ballzus, C., H. Frimannson, G. Gunnarsson, and I. Hrolfsson, The Geothermal Power Plant at Nesjavvellir, Iceland, Proceedings World Geothermal Congress 2000 Kyushu Tohoku, Japan, May 28 (June 10, 2000) 3. Reykjavik Energy, Nesjavellir Power Plant 4. Geothermal Resources in Iceland, 5. How the Plant Works, 8. Virtual Materials Group, Inc. Version 3.1.44 (January, 2008) 9. Gunnarsson, A., B.S. Steingrimsson, E. Gunnlaugsson, J. Magnusson, and R. Maack, Nesjavellir Geothermal Co-Generation Power Plant, Geothermics 21 (4), 559 (1992) 10. Perry, J.H., Chemical Engineers Handbook 3 rd Ed., 673-674 (1950) 11. 12. 13. 14. Gislason, G., Nesjavellir Co-Generation Plant, Iceland. Flow of Geothermal Steam and Non-Condensable Gases, Proceedings World Geothermal Congress 2000 Kyushu Tohoku, Japan, May 28 (June 10, 2000) 15. Edgar, T., A Course on Energy Technology and Policy, Chem. Eng. Ed. 41 (3), 195 (2007) 16.

PAGE 23

Vol. 42, No. 3, Summer 2008 139 Random Thoughts . H eres the situation. Youre working on a big writing projecta proposal, paper, book, dissertation, what to get done is one measly paragraph. Youre long past the just looked at your to-do list and reminded yourself that this is only one of several writing projects on your plate and you havent even started most of the others. If youre frequently in that situation (and weve never met a faculty member who isnt) weve got a remedy for you. First, though, lets do some truth in advertising. Lots of books and articles have been written about how to write clear and persuasive papers, proposals, dissertations, lab reports, techni cal memos, love letters, and practically everything else you might ever need to write. Were not going to talk about that stuff: youre on your own when it comes to anything having to do with writing quality. All were going to try to do here is help you get a complete draft in a reasonable period of time, because that usually turns out to be the make-or-break step in big writing projects. Unless youre a pathological perfectionist once youve got a draft, theres an excellent chance that a far behind. We have two suggestions for getting a major document writ ten in this lifetime: (1) commit to working on it regularly, and (2) keep the creating and editing functions separate.* Dedicate short and frequent periods of time to your major writing projects Copyright ChE Division of ASEE 2008 HOW T O WRITE ANYTHINGRICHARD M. FELDERREBECCA BRENT I write when Im inspired, and I see to it that Im inspired at nine oclock every morning. (Peter De Vries) See if this little monologue sounds familiar. I dont have time to work on the proposal nowIve got to get Wednesdays lecture ready and theres a ton of e-mail to answer and Ive got to pick the kids up after school tomorrow . BUT, as soon as fall break (or Christmas, or summer, or my sabbati cal) comes, Ill get to it. Its natural to give top priority to the tasks that can be done quickly or are due soon, whether theyre important (preparing Wednesdays lecture) or not (answering most e-mails), and so the longer-range projects keep getting put off as the weeks date, you panic when it approaches and quickly knock some many references on writing. A particularly good one is Robert Boice, Professors as Writers Stillwater, OK: New Forums Press, 1990.

PAGE 24

Chemical Engineering Education 140 All of the Random Thoughts columns are now available on the World Wide Web at thing out well below the best you can do. If its a proposal or paper, subsequent rejection should not come as a surprise. If the book youve been working on for the last 10 years never gets into print, or your graduate students leave school with their research completed but without their Ph.D.s because The strategy of waiting for large blocks of time to work on it can take hours or days to build momentum again and youre likely to run out of time before much gets written. Also, as your family, whom youve been neglecting for months and who now legitimately think its their turn. A much more effective strategy is to make a commitment to regularly devote short periods of time to major writing projects. Thirty minutes a day is plenty, or maybe an hour youre at your peak, during which you close your door, ignore your phone, and do nothing but work on the project. Alterna tively, you might take a few 10 minute breaks during the daytimes when you would ordinarily check your e-mail or surf the Web or play Sudokuand use them to work on the project instead. Either way, when you start to write youll quickly remember where you left off last time and jump in with little wasted motion. When youve put in your budgeted time for the day, you can (and generally should) stop and go back to the rest of your life. These short writing interludes wont make much differ astounded when you look back after a week or two and see how much youve gotten done on the projectand when a larger block of time opens up, youll be able to use it effec you also take our next suggestion. Do your creating and editing sequentially, not simul taneously Heres another common scenario that might ring a bell. sentence. You look at it, change some words, add a phrase, rewrite it three or four times, put in a comma here, take one to rewrite it again . and you work on those two sentences repeat the process . and an hour or two later you may have a paragraph to show for your efforts. If that sounds like your process, its little wonder that you you spend hours on every paragraph, the 25-page proposal or 350-page dissertation can take forever, and youre likely to become frustrated and quit before youre even close to a At this point youre ready for our second tip, which is to keep the creating and editing processes separate The routine we just described does the opposite: Even before you complete doing that, write whatever comes into your head, without looking back. If you have trouble getting a session started, write anything random words, if necessaryand after a from outlines, start with an outline; if the project is not huge like a book or dissertation and you dont like outlines, just plunge in. If youre not sure how to begin a project, start with the introduction later. Throughout this process you will, of course, hear the usual voice in your head telling you that what youre writing is pure garbagesloppy, confusing, trivial, etc. Ignore it! Write the as much written as your budgeted time allows. Then, when you come back to the project the next day (remember, you committed to it), you can either continue writing or go back and edit what youve already gotand then (and only then) is the time to worry about grammar and syntax and style and all that. Heres what will almost certainly happen if you follow that indeed be garbage, but the rest will invariably be much bet ter than you thought while you were writing it. Youll crank much easier and faster to edit it all at once rather than in tiny completed manuscript in a small fraction of the time it would take with one-sentence-at-a-time editing. Were not suggesting that working a little on big projects every day is easy. It isnt for most people, and days will inevitably come when the pressure to work only on urgent tasks is overwhelming. When it happens, just do what you have to do without beating yourself up about it and resume your commitment the next day. It may be tough but its do able, and it works.

PAGE 25

Vol. 42, No. 3, Summer 2008 141 P rocess safety is a fundamental component of sound process design. Although the chemical industry has demonstrated an excellent safety record over the years, [1] the quantities and hazardous nature of many of the substances typically handled by chemical manufacturers make the potential for large-scale disasters a constant con cern. Because safety is so critical in industry, it is vital to introduce the concept of safe process design practices during undergraduate chemical engineering education. From famous historic disasters such as Flixborough and Bhopal to recent the importance of process safety in chemical process design is abundantly clear. An appreciation of this gained during a chemical engineers education can only enhance chemical manufacturing safety in the future. in the concept of risk. From government regulatory require ments, such as those outlined by OSHA and the EPA, [2-4] to industry initiatives such as Responsible Care, the requirement of quantifying and managing risk is paramount. In addition to working within economic and environmental constraints, the process design engineer is also tasked with reducing the risk of operating a chemical manufacturing process to an acceptable level for employees, regulatory authorities, insur ance underwriters, and the community at large. Therefore, a holistic approach to process safety as an integral component of sound process design is critical. In addition to the study of toxicological impacts and quan tifying release scenarios, an understanding of how risk is process design engineers to mitigate those risks at the earliest ChE Copyright ChE Division of ASEE 2008 INCORPORA TING RISK ASSESSMENT AND INHERENTLY SAFER DESIGN PRACTICES JEFFREY R. SEAY University of Kentucky Paducah, KY 42002 MARIO R. EDEN Auburn University Auburn, AL 36849

PAGE 26

Chemical Engineering Education 142 stage of conceptual process developmentthe stage where design. This paper will present, by case-study example, how the fundamental concepts of inherently safer process design can be integrated into chemical engineering education.RISK ASSESSMENT METHODOLOGY safer process design, the chemical engineering student must stood by both the general public and students of chemical engineering. It is important to separate the concept of risk from the concept of hazard. While the concept of hazard re lates to the potential for adverse consequences, risk is rather a combination of both the severity of the consequences of an upset scenario and the likelihood of that scenarios initiating cause. This is an important distinction. The potential hazard associated with a substance or process is an inherent property that cannot be changed. The risk associated with handling a substance or operating a process can be high or low, depending upon the safeguards included in the design. Thus, for chemical engineers, the most important distinction between hazard and risk is that risk can be reduced through process design. In order to begin to discuss risk, the process design engi words, answer the question, What is the worst thing that can happen? Answers to this question typically involve loss of containment of a process chemical with causes ranging from failure of control loops and operator errors to external events answers to the aforementioned question must be considered independently of the likelihood of the worst-case scenario oc curring. Again, it is the combination of both the severity and the likelihood that deter mines the risk. In order to ensure a complete and consistent assess ment of potential upset scenarios, a structured approach must be ap plied. The need for such an approach is the basis for a Process Hazard Analysis. A Process Hazard Analysis (PHA) is a methodology for review ing and assessing the potential hazards of a chemical process by us ing a structured, facili tated, team brainstorm ing approach. A PHA is typically facilitated by a trained team leader and attended by a wide variety of plant person nel, including engineers, managers, operators, maintenance technicians and safety, health, and environmental (SHE) personnel. Although several techniques are available for performing PHAs, [3] the goal of the PHA is always the same T ABLE 1 Example of a hazard scenario using What If...? methodology What If? Initiating Cause Consequence Safeguards 1. There is High Pressure in the Cyclohexane Storage Tank? 1.1 Failure of the pressure regulator on nitrogen supply line to Cyclohexane Storage Tank. 1.1 Potential for pressure in nitrogen pad gas through failed regulator. Potential to exceed design pressure of storage tank. Potential tank leak or rupture ignition source be present. Po tential personnel injury should exposure occur. 2.1 Potential environmental release requir ing reporting and remediation. 1. Conservation vent sized to relieve over pressure due to this scenario. 2. Pressure transmitter with high alarm set to indicate high pressure in Cyclohexane Storage Tank. T ABLE 2 Inherently Safer Design Choices for Common Design Applications Hazard Scenario Process Operation Potential Upset Case Inherently Safer Design Overpressure Filling a process vessel with a pump. Overpressure by pump 1. Vessel design pres sure greater than pump deadhead pressure 2. Static head due to vessel elevation plus vessel design pres sure greater than pump deadhead pressure. Overpressure Operating a vessel un der inert gas pressure. Failure of inlet gas regula tor leading to overpressure. 1.Vessel design pressure greater than inert gas supply pressure. Underpresure Emptying a process vessel with a pump. Blocked vent leading to vessel collapse due to vacuum pulled during pump out. 1.Vessel designed for full vacuum Underpressure Draining an elevated process vessel by gravity. Blocked vent leading to vessel collapse due to vacuum pulled during draining. 1.Vessel designed for full vacuum. 2. Liquid drain lined sized to be self-venting.

PAGE 27

Vol. 42, No. 3, Summer 2008 143 to identify the potential hazards of a process and determine hazards.CLASSROOM EXAMPLE OF APPLYING PHA METHODOLOGY The following is a simple example that can be used to illus trate the basic concepts of a PHA in the chemical engineering classroom. Consider a low design pressure API storage tank designed for no more than 2.5 pounds of pressure and only a few inches of water of vacuum. Therefore, careful control of pressure is critical. Furthermore, assume that the storage tank is equipped with a pad/de-pad vent system to control pres sure, and is located in a diked tank farm. Table 1 illustrates a typical scenario that might be developed during a PHA using the What If? methodology. In Table 1, the listed safeguards would be effective means of mitigating the personnel exposure and environmental impact cause illustrated, other causes of high pressure that might be considered by a PHA Team include the following: pressure in the storage tank. by deadhead pump pressure. If the safeguards iden tified by the PHA team are not deemed adequate, recommendations are made for the implemen tation of additional safe guards. This technique, called Layer of Protec tion Analysis (LOPA), is often employed by PHA teams to quantitatively assess the risk associated with an upset scenario so that appropriate layers of protection can be applied to adequately mitigate the risk. [5] Hazard assessment and layer of protection analy sis are complex subjects. As such, a formal hazard analysis is typically not performed during the con ceptual phase of process design. In most cases, the PHA is performed during the engineering phases of a project. A basic understanding of the fundamentals of risk assessment, however, is extremely during the conceptual phase of process design. To make inher ently safer design choices during conceptual development of a process, the design engineering student must be aware of piece of equipment or system. Inherently safe process design practices can generally be [6, 7] Substitution Attenuation Limitation of effects Some examples of inherently safer design choices for typi cal process applications are included in Table 2. Typically, however, these types of design choices are made in later stages of engineering development. Although these begin evaluating inherently safer design strategies at the earli est stages of process development, when the process design engineer has the greatest opportunity to affect the safety T ABLE 3 Potential opportunities for making inherently safer design choices [6] Process Design Choice Inherently Safe Design Category Potential Process Safety Impact Reactor type Continuous reactors are typically smaller than batch reac tors for a given production volume. Feed stocks Substitution Less hazardous raw materials may be available to make the same products. Process solvents Substitution Less hazardous and/or less volatile solvents may be available. Reaction mechanism Attenuation Endothermic reactions present less potential for runaway. Operating conditions Attenuation Temperatures and pressure close to ambient are typically less hazardous. Process utilities Attenuation Low pressure utilities such as hot oil may be a safer choice than high pressure steam. Alternative technology Attenuation Use of alternative technology, for example pervaporation instead of azeotropic distillation using a solvent entrainer. Production rate Limitation of effects A continuous process making just what is required can be safer that a batch process with a large hold-up volume. Storage volume Limitation of effects Minimization of volume limits the potential effects of a release. Equipment layout equipment. Cooling by natural convection eliminates the potential for process upsets due to loss of utilities.

PAGE 28

Chemical Engineering Education 144 aspects of the process. Some examples of design choices that are typically made at the onset of conceptual engineering are illustrated in Table 3 (previous page). Initially, inherently safer designs may seem to be more expensive than applying traditional safeguards to processes. When the total cost of the process is considered, however, the inherently safer design is often more cost effective. Installing and maintaining multiple independent layers of protection can be quite expensive, but these costs are often ignored during initial cost estimates. Conceptual phase cost estimates are usually based on stand-alone major equipment costs that are simply multiplied by factors to obtain the total installed cost. These factors are intended to account for instrumentation and controls, among other items needed for the complete process installation. To apply the same factors to traditional and inherently safer processes, however, can lead to an er roneous comparison and conclusion. Inherently safer processes will typically require fewer safety controls, which leads to lower installation and op erating costs. These factors should be considered when evaluating processes during a hierarchical approach to process design. Additional cost sav ings for inherently safer processes that are often overlooked include insurance costs and costs associated with regulatory compliance. The following case study is presented as a classroom engineering design problem to illus trate the techniques of applying inherently safer design choices. Consider a chemical process using 1-propanol as a solvent. Currently, the waste solvent ends up as a waste-water stream for disposal. The task for the process design engineer is to develop a process to recover the 1-propanol from the wastewater stream. This separation is complicated by the fact that water and 1-propanol form a mini mum-boiling azeotrope. Therefore, separation by ordinary distillation is not possible. The traditional method employed for breaking this azeotrope uses a third solvent, or entrainer. For the water/1-propanol system, cyclohexane diagram of the azeotropic distillation process is given in Figure 1. In this process, the minimum-boiling azeotrope is separated from the water in the Azeotrope Col umn and is collected as an overhead product. The azeotrope is then mixed with the cyclohexane in the Entrainer Vessel. The 1-propanol is soluble in cyclohexane, while the water is not. The water phase, with a small amount of 1-propanol, is then recycled back to the Azeotrope Column, while the cyclohexane/1-pro panol mixture is fed to the Solvent Column, where 1-propanol is recovered as a bottoms product and the cyclohexanewith a small amount of 1-propanolis recycled to the Entrainer Vessel. This simple system is easily modeled using any pro cess simulation software package. From a process design perspective, this process is certainly acceptable. From the perspective of safety, however, some highly volatile solvent, cyclohexane, is introduced to the pro cess. A sample of some of the potential hazard scenarios that might be generated during a PHA is illustrated in Table 4. Some potential safeguards that might be used to mitigate these hazards include safety relief valves, redundant in Figure 2. Flow diagram of inherently safer solvent recovery process. Figure 1. Flow diagram of traditional solvent recovery process.

PAGE 29

Vol. 42, No. 3, Summer 2008 strumentation, and hardwired interlocks independent from the primary basic process control system (BPCS). All of these safeguards would be applied to the process during later stages of process design, as considering inherently safer design choices could make such safeguards unnecessary. An inherently safer approach to this design problem will in clude technology to break the azeotrope without introducing to the process. One possible solution is the use of a pervapo ration membrane. A pervaporation membrane separates two liquids by partial vaporization through a nonporous mem brane, such as ceramic. The pervaporation membrane is able to break azeotropes due to its ability to separate components based on polarity differences between the molecules, rather than relying on differences in vapor pressure, like distilla tion does. Although the pervaporation technology could be used to completely separate 1-propanol from the water in one step, such a sharp split would most likely prove to be prohibitively expensive. An optimum design using a combination of dis tillation and pervaporation can be achieved, as illustrated in Figure 2. In this design, the azeotrope is again separated from the water as an overhead product in the Azeotrope Column, but in this process, instead of using an entrainer, the Pervaporization Unit is used to separate the liquids. Because this technology is used in conjunction with distillation, a sharp split is not needed. The water-rich phase leaving the Pervaporation Unit is returned to the Azeotrope Column, and the 1-propanol is recovered as a bottoms product from the Solvent Column, with the azeotrope being collected overhead and returned to the Azeotrope column. This design is advantageous because it can be optimized to minimize the impact of the cost of the Pervaporization Unit. The inherently safer design has the obvious advantage of the process. Taking a wider view, not only is the cyclohexane eliminated from the process itself, but also from storage ar eas, unloading areas, and waste treatment. In addition to the material through the system. Therefore, the column and as sociated heat exchangers are smaller than with the traditional the inherently safer process are illustrated in Table 5. process are clear. The pervaporation process addresses three of due to the elimination of the entrainment solvent from the process, and Limitation of Effects is due to the smaller equipment and chemical inventories. Of course, 1-propanol is a narios listed in Table 4 would still need to be considered, but by eliminating the cyclohexane from the process, the overall severity of the consequences would be reduced. Since, as discussed previously, risk is a combination of both severity and likelihood, the overall risk of the inher T ABLE 4 Potential Hazard Scenarios for Traditional Azeotropic Distillation Process What If? Initiating Cause Consequence 1.There is higher pres sure in the Entrainment Vessel? process area. 1.1 Potential increased temperature and pressure leading to possible vessel leak or material to the atmosphere. Potential personnel injury due to exposure. 1.2 Pressure regulator for inert gas pad fails open. 1.2 Potential for vessel pressure to increase up to the inert gas supply pressure. Poten tial vessel leak or rupture leading to release Potential personnel injury due to exposure. 2. There is higher level in the Entrainer Vessel? 2.1 Vessel level trans mitter fails and indicates lower than actual volume. the vent gas incinerator. Potential to overwhelm incinerator leading to possible explosion. Potential personnel injury due to exposure. T ABLE 5 Summary of Process Safety Implications of Design Choices for Case Study Example. Upset Scenario Traditional Process Inherently Safer Process External Fire liquid circulating in process. Flammable volume limited to recovered solvent only. Cyclohexane entrainer more volatile than 1-propanol. Minimal liquid hold up in Pervaporation Unit. Overpressure Larger liquid hold-up leads to higher severity in the event of a release. Volume limited to solvent distillation hold-up.

PAGE 30

Chemical Engineering Education 146 ently safer design would be reduced. Although a more in-depth study would be required before making the choice of which solvent recovery process is preferred, it should be clear that these decisions must be made in the early stages of conceptual CONCLUSIONS One of the responsibilities of every chemical engineer is to ensure that the excellent safety record enjoyed by the chemical process industry is maintained. Therefore it is important to begin introducing the fundamentals of process safety during undergraduate chemical engineering education. The purpose of this work has been to underscore, by case-study example, the natural relationship between inherently safe process design and conceptual process development, and describe how it can be integrated into undergraduate process design education. As has been illustrated by this case study, taking a holistic approach to process safety education can serve to reinforce the decisions made during conceptual process development. choices during conceptual process development, students of process engineering will be better prepared for the chal lenges of meeting the high standards of safety set by todays chemical industry.ACKNOWLEDGMENTS The authors would like to acknowledge Felicia Foster and Robert DAlessandro of Evonik Degussa Corporation for providing valuable insight and guidance on the industrial applications of process safety.REFERENCES 1. Sanders, R., Chemical Process Safety Learning from Case Histories 3rd Ed., Elsevier, Inc., (2005) 2. Nelson, D., Managing Chemical Safety Government Institutes, (2003) 3. Environmental Protection Agency, Process Hazard Analysis, 40 CFR 68.67 (2005) 4. Occupational Safety and Health Administration, Process Safety Man agement of Highly Hazardous Chemicals, 29 CFR 1910.119 (2005) 5. Center for Chemical Process Safety, Layer of Protection Analysis AIChE (2001) 6. Kletz, T., Process Plants: A Handbook for Inherently Safety Design Taylor and Francis (1998) 7. Center for Chemical Process Safety, Guidelines for Engineering Design for Process Safety AIChE, 1993.

PAGE 31

Vol. 42, No. 3, Summer 2008 147 T he concept of the solubility of a gas in a liquid is familiar in many everyday ways. When we drink a bottle of carbonated beverage, we understand it con tains dissolved carbon dioxide, and we know animals can live in oceans and rivers because oxygen dissolves in water and animal life depends on the dissolution of oxygen in blood. For chemical engineers, this everyday phenomenon is important from both practical and theoretical points of view. [1, 2] From a practical point of view, this concept is used in the design of absorption columns where a gaseous mixture is separated by contact with suitable solvents that dissolve its components differently. Further, knowledge of gas solubility in water is important in the processes that control environmental distri bution of contaminants, such as halogenated hydrocarbons. From the theoretical point of view, the solubility of gases in liquids is an excellent tool to investigate solute-solvent intermolecular forces in the liquid state, since solute-solute interactions are almost negligible. oped by Fonseca, et al, [3] for experimental determination of the solubility of a gas in a liquid. This experiment is imple mented in the chemical engineering department of Coimbra, and is a part of one of the third-year laboratory courses in phase equilibria domain. The experiments are carried out by groups of a maximum of three students during approximately four hours. The full report of the experiment is completed by the group at home and must be presented to the teacher ChE laboratoryA LAB EXPERIMENT T O INTRODUCE GAS/LIQUID SOLUBILITYI.M.A. FONSECA AND J.P.B. ALMEIDA Universidade de Coimbra 3030-290 Coimbra, PortugalH.C. FACHADA Instituto Politcnico de Coimbra 3030-199 Coimbra, Portugal Copyright ChE Division of ASEE 2008

PAGE 32

Chemical Engineering Education 148 the following week and discussed one week later. The report consists of the following sections: introduction, experiment objectives, background of the experiment, description of the apparatus, experimental procedure, and a discussion/results section. The analysis of uncertainties must also be done using the propagation law of errors, and the relative contribution of i.e. the solubilityevaluated. The students must also include in the report a section of conclusions, where they can also present comments and recommendations for improvements. THEORY When a gas is in contact with a liquid it tends to dissolve in the liquid and the liquid evaporates until equilibrium is reached. Consider a binary mixture of components 1 and 2 at temperature T and pressure P, represented schematically in Figure 1. The component 2 is near or above its critical tem perature, which means this component is a gas at temperature and a vapor below T c ). When the liquid-phase mole fraction, x 2 is small and the equilibrium vapor-phase mole fraction, y 2 is large (near unity), it is conventional to call species 2 a dissolved gas, and to label the physical situation one of gas solubility. [4] Gas/liquid solubility is a particular case of vapor/liquid equilibrium; therefore, the classical treatment of this subject is similar in some important aspects. Lets consider species 1 the major component of the liquid phase (the solvent) and species 2 the dissolved gas (the solute). The liquid phase mole fraction, x 2 represents the solubility of gaseous solute 2 in liquid solvent 1. For component 2, the thermodynamic condition of phase equilibrium states: [5] ff GL 22 1 ( ) where f G 2 and f L 2 are the fugacities of component 2 in the vapor and liquid phases, respectively. f G 2 is given by the expression: fy P G 22 2 2 ( ) where 2 in the vapor phase and P is the total pressure. f L 2 is repre sented by: fx f L 22 22 0 3 ( ) 2 and f 2 0 the standard-state fugacity of the same component, which is usually assumed to be the fugacity of the pure liquid at the temperature and pressure of equilibrium ff L 2 0 2 *, The substitution of Eqs. (2) and (3) in Eq. (1) gives: yP xf L 22 22 2 4 *, ( ) following assumptions: [5] 2 1 = 1), i.e. the liquid phase is an ideal solution; (ii) 2 = 1 ( 2 = 1) i.e., the vapor phase is an ideal mixture; (iii) fP P L 22 2 *, ** ( is the pure vapor pressure of compo nent 2 at temperature T) i.e., the effect of pressure on the fugacity of the pure liquid phase is negligible (Poynting correction =1) at moderate pressures; and 2 = 1, ( 2 is the P 2 is low at temperature T. Therefore Eq. (4) reduces to Raoults law: Py Px P 22 22 5 ( ) where P 2 represents the partial pressure of component 2. The solubility x 2 as given by Eq. (5), is called the ideal solubility of the gas. This expression states that the solubility is independent of the solvent and for a given gas, at a constant partial pressure, the solubility always decreases with rising temperature, which is not always true. Because of these dis advantages, the ideal solubility expression usually gives no more than a rough estimate of gas solubility. The solubility of gas in a liquid is often proportional to its fugacity in the vapor phase. This situation is described by a more realistic expression, Henrys law: fH x L 22 12 6 , ( ) where H 2,1 is a constant that for a given solute and solvent depends only on temperature. Again the assumptions leading to Eq. (6) can be readily recognized by comparing it with Eq. (4). The more important (independent of the composition), which means that Henrys ing Eq. (6) in more convenient form: H f x x L 21 0 2 2 2 7 lim ( ) which is equivalent to H yP x x 21 0 22 2 2 8 lim ( ) taking into account Eqs. (1) and (2). Figure 1. Schematic representation of gas/liquid solubility.

PAGE 33

Vol. 42, No. 3, Summer 2008 149 EXPERIMENT AL SECTION APP ARA TUS The determination of the solubility of gas in a liquid is made using a volumetric method. [6] The principle of this procedure is to bring a measured volume of liquid into contact with a known volume of gas at a given temperature and pressure. After equilibrium has been attained, the change in the gas volume yields the amount of gas dissolved in the liquid and hence the solubility. The solubility apparatus is shown schematically in Figure 2. The apparatus is housed in a water thermostat where the temperature is maintained constant using a temperature controller, TC. The temperature is measured of the thermostat bath can be adjusted using an elevator, E, in order to immerse the whole vacuum line. The main features of the apparatus are: EQ, equilibrium vessel, (~50 cm 3 ) where the dissolution of the gas takes place; GB, gas burette, which consists in a piston-cylinder arrange ment; PT, pressure transducer, for total pressure readings; LA, linear actuator, which moves the piston in the gas burette; and PC, pressure controller. PROCEDURE The experimental procedure begins with the evacuation of the whole apparatus. After this two fundamental steps in any gas solubility measurement need to be performed: (i) degassing of the solvent (ii) dissolution of the gas To accomplish step (i), the equilibrium vessel is removed from the line, lowering the level of the thermostat bath, and it is again connected to the line and the position of the thermostat bath re-established. The stopcocks V2 and V4 are opened to degas the solvent in EQ during about 10 min. The magnetic stirrer must be on. This degassing procedure should be repeated two or three times until the measured pressure equals the solvent vapor pressure at the equilibrium temperature. Then EQ must be removed from the line in order to be weighed. After this procedure, EQ is connected again to the vacuum line. Step (ii) begins with slowly opening stopcock V1, with V2 opened (V3 and V4 must be closed) to admit the gas to the equilibrium vessel. The total pressure is adjusted to ca 1 atm and, after this, stopcock V1 is closed and V3 is opened. This pressure acts as a reference value for the pressure controller, which commands the linear actuator, LA. As the gas dissolves the pressure decreases and this is detected by PT. The linear actuator, LA, drives the piston down the cylinder to maintain the pressure constant at the reference value. The number of encoder pulses is counted and displayed, and a conversion is made to determine the volume of gas displaced from the precision-bore tube that comprises that cylinder of the burette. This volume represents the volume of the gas dissolved. The detailed experimental procedure is described in Appendix A. ADDITIONAL INFORMA TION FOR INSTRUCT ORS In this experiment it is important to use gases, such as CO 2 N 2 O or CH 3 has the disadvantage of being very expensive. Less soluble gases (O 2 N 2 etc.) should not be used in this experiment since the time of dissolution increases substantially. For these kinds of gases an equilibrium vessel with a greater volume should be used. [3] Other solvents such as primary alcohols ( e.g. methanol, ethanol, propan-1-ol, butan-1-ol) can also be used. They must have purities greater than 99.8 percent. The accuracy of the method can be improved using lecture bottles of gases, which have higher purity (>99.5%) than other commercial gases. A pressure reducer should be used connected to the lecture bottle. The lecture bottles are quite ap propriate to classroom experiments and must be used in these experiments. An adequate pressure reducer must be coupled to the lecture bottle. In the present experiments we have used a pressure reducer HSB 280 5 (from PRAXAIR), but other reducers can be adapted depending on the mark of the lecture bottle. This allows a safe use of the lecture bottles. The linear actuator, LA, consists of a permanent magnetic DC motor, which drives a worm screw coupled to an optical encoder. The worm screw moves the piston inside the cylinder of the gas burette and the displacement is proportional to the number of encoder pulses (n p ). The proportionality constant Figure 2. Solubility apparatus: TB, thermostated bath; TC, tem perature controller; PT, pressure transducer; LA, linear actuator; PC, pressure controller; EQ, equilibrium vessel with connector; GB, gas burette; V1,...V4, high vacuum Teon stopcocks; AGIT, magnetic stirrer; E, elevator. EQ V9

PAGE 34

Chemical Engineering Education indicated in the manual of the linear actuator should be given to the students. In the present apparatus, the displacement is -4 n p In this experi ment we have used a pressure transducer from Honeywell, model PPT0015AWN2VA-A. The thermostat bath can be built using Perspex or glass (71 26 38 cm) where a thermostatic control unit is immersed. The whole setup costs about 1,800 euros.CALCULA TIONS To simplify the treatment of the raw data we consider some assumptions: neglect the volume change of the liquid sample during saturation and ideal solution behavior. The raw data obtained from experiment are: n p the number of encoder pulses from the linear actuator, the equilibrium temperature, T, the reference pressure, P, and m 1 the mass of displacement of the piston, using a conversion factor indicated in the user manual of the pressure controller. The change of the volume of gas in the gas burette due to gas dissolution is obtained from: Vr h 2 9 ( ) where r represents the internal radius of the gas burette. The quantity of the gas absorbed in the liquid, n 2 (in moles), can PV RT BR TP G mi x m ix /( ) / ,( ) 1 1 0 where V G mix represents the molar volume in the vapor phase and B mix given by the expression, By By By yB mi x 1 2 11 2 2 22 12 12 2 1 1 ,( ) where B 11 B 22 and B 12 respectively. Substituting, in Eq. (10), V G mix G where n G repre sents the total number of moles contained in the V volume, one obtains, Vn RT PB G mi x /( /) ( ) 12 which multiplied by (1/y 2 ) gives, ny PV RT BP mi x 22 13 /( ). () The value of the solubility, x 2 is then obtained from: xn nn 22 12 14 /( ), () where n 1 and n 2 represent the amount of solvent and solute in moles in the liquid phase, respectively. The n 1 is obtained directly from n 1 = m 1 / M 1 Since we need to know y 2 to obtain n 2 from Eq. (13), this calculation requires an iterative proce dure. The calculations begin with estimates of the vapour and liquid phases obtained from Raoult and Dalton laws. In the following iteractions, these compositions are improved using Eqs. (13) and (14) and the following expression: yx P P 2 2 11 1 11 15 () ( ) ** which results from the thermodynamic condition of phase equilibrium written for the solvent. The 1 phase, which is given by, 1 1 12 2 12 22 11 2 ex p( /) ,( PR TB yB BB 1 16 ) and 1 saturation conditions obtained from, 1 1 11 17 * ex p/ () ( ) BP RT where P 1 is the vapor pressure of pure component 1 at the equilibrium temperature, which can be obtained using a vapor pressure equation. Figure 3. Solubility of carbon dioxide () and nitrous ox ide ( ) in water. The curves were obtained from Eq.(18). The students must also include in the report a section of conclusions, where they can also present comments and recommendations for improvements.

PAGE 35

Vol. 42, No. 3, Summer 2008 The calculation ends when convergence is obtained between two consecutive x 2 values. The determination of Henrys constant, H 2,1 is then straightforward from Eq. (8). All the solubilities found in this work were corrected to 1 atm partial pressure using Henrys law, since the literature values are referred to this pressure. ERROR ANALYSIS The analysis of uncertainties or errors is a vital part subject to some uncertainties. In this experiment the de result, the mole fraction solubility, provides use of the propagation law of errors. [7] This law gives the relative magnitude of the uncertainties of the measured variables, i.e., the student will be able to tell which of the experimental errors affect the solubility value more. Appendix B presents an entrance form distributed to each group of students in order to guide them in treating the raw data and analyzing the uncertainties of the results. The data was obtained by one of these groups to determine the solubility of CO 2 in water, at the temperature 298.39 K. RESUL TS AND DISCUSSION The experimental solubility data and the values reported in the literature for the sys tems CO 2 /H 2 O and N 2 O/H 2 O are shown in Table 1. The accuracy of the experimental method is found to be about 1 percent. We have also determined the ideal solubili ties of both gases at P 2 = 101325 Pa using Eq. (5). The vapor pressures of the pure components were obtained from the Wagner equation. [9] The value determined for CO 2 is x 2 = 1.57 -2 (T = 298.39 K) and for N 2 O is x 2 = 1.91 -2 (T = 298.13 K). These values are quite different from the experimental ones, since Raoults law gives only a rough estimate of the solubility of a gas independent of the solvent. taken from the Dymond and Smith compila tion. [10] The dependence of the solubility on tem perature has been represented by: Rx AB TC T ln /l n, () 2 18 least-squares method. The optimized param eters of Eq. (18) and the average absolute T ABLE 1 Solubility of CO 2 and N 2 O in water, expressed as mole fraction, x 2 at a partial pressure P 2 = 101325 Pa. H 2,1 Solute T/K x 2 /10 -4 x 2lit /10 -4 (a) H 2,1 / (MPa) CO 2 290.27 7.70 7.61 1.1 131.7 291.49 7.36 7.34 0.3 137.7 292.11 7.22 7.21 0.2 140.4 293.39 6.90 6.94 0.6 146.8 294.58 6.65 6.71 0.9 152.3 295.15 6.66 6.60 0.9 152.1 296.19 6.36 6.42 0.8 159.2 297.19 6.19 6.24 0.9 163.8 298.39 5.98 6.04 1.0 169.4 299.37 5.89 5.89 0.0 172.0 300.15 5.80 5.77 0.5 174.8 301.10 5.61 5.63 0.3 180.5 302.13 5.43 5.48 1.1 186.7 302.93 5.35 5.38 0.5 189.5 N 2 O 290.36 5.49 5.53 0.6 184.4 291.36 5.38 5.36 0.4 188.4 292.36 5.17 5.19 0.4 196.1 293.30 5.03 5.04 0.2 201.5 294.20 4.93 4.90 0.5 205.5 295.40 4.71 4.73 0.3 214.9 296.15 4.66 4.63 0.8 217.2 297.25 4.49 4.48 0.3 225.6 298.13 4.38 4.37 0.2 231.5 299.07 4.28 4.25 0.7 236.6 300.16 4.13 4.12 0.2 245.2 301.15 3.97 4.01 1.0 255.0 302.15 3.84 3.90 1.7 263.9 303.13 3.81 3.80 0.0 266.3 (a) (%) xx x x lit lit 22 2 100 where x 2 represents the solubility value found in this work and x 2lit is obtained from Reference 8. T ABLE 2 Parameters in the equation Rln x 2 = A + B/T + ClnT. Solute A/(J K -1 mol -1 ) B/(J mol -1 ) C/(J K -1 mol -1 ) AAD(%) CO 2 -2785.2 138287.6 396.617 0.5 N 2 O 890.1 -23763.0 -153.526 0.5

PAGE 36

Chemical Engineering Education deviation of x 2 AA DM xx cal cx x (/ )( exp) () /( exp) 1 1 0 2 2 2 0 01 9 () where M is the number of experimental points, are listed in Table 2 (previous page). In Figure 3 (p. 150) we have plotted the experimental agreement is good. CONCLUSIONS The apparatus used here to measure the solubility of gas in a liquid is simple and clearly illustrates this concept. It combines easy handling with automated data retrieval, lead ing to experimental results with reasonable accuracy for a pedagogical experiment. The treatment of the raw data to obtain the solubilities is a good application and demonstration of some gas laws (Daltons law of partial pressures, Raoults and Henrys laws). This experiment also gives the opportunity to demonstrate real behavior of a gas (through the fugacity concept) to the students. The analysis of errors will allow evaluation of the relative magnitudes of the uncertainties within the measured solubility. ACKNOWLEDGMENT This work was carried out under Research Project POCTI/ cia e Tecnologia (Portugal) and FEDER. We thank Professor Margarida F. Costa Gomes from the Laboratoire de Ther modynamique des Solutions et des Polymres, Universit Blaise Pascal, Clermont-Ferrand, for her advice related to the assembly of the solubility apparatus. NOMENCLA TURE A, B, and C parameters of Eq. (18) f L 2 fugacity of component 2 in the liquid phase [Pa] f L 2 *, fugacity of pure liquid 2 [Pa] f G 2 fugacity of component 2 in the vapor phase [Pa] f 2 0 standard-state fugacity of component 2 [Pa] H 2,1 Henry constant of component 2 in solvent 1 [MPa] n 1 amount of solvent [mol] n 2 amount of solute [mol] n p number of encoder pulses P equilibrium pressure [Pa] P2 partial pressure of component 2 [Pa] P 2 vapor pressure of pure component 2 [Pa] P 1 vapor pressure of pure component 1 [Pa] R ideal gas constant [J mol -1 K -1 ] T equilibrium temperature [K] 3 ] x 2 mole fraction of component 2 in the liquid phase y 2 mole fraction of component 2 in the vapor phase 2 REFERENCES 1. Wilhelm, E., and R. Battino, Chem. Rev. 73 1 (1973) 2. Letcher, M.T., and R. Battino, J.Chem. Educ. 78 103 (2001) 3. Fonseca, I.M.A., J.P.B. Almeida, and H.C. Fachada, J. Chem. Ther modynamics 39 1407 (2007) 4. Van Ness, H.C., and M.M. Abbott, Classical Thermodynamics of Nonelectrolyte Solutions Mc Graw Hill (1982) 5. Prausnitz, J.M., R.N. Lichtenthaler, and E.G. Azevedo, Molecular Thermodynamics of FluidPhase Equilibria 3rd Ed., Prentice Hall, New Jersey (1999) 6. Gerrard, W., and P.G.T. Fogg, Solubility of Gases in Liquids Wiley, New York (1991) 7. Taylor, J.R., An Introduction to Error Analysis 2nd Ed., University Science Books (1997) 8. Wilhelm, E., R. Battino, and R. Wilcock, Chem. Rev. 77 219 (1977) 9. Prausnitz, J.M., B.E. Poling, and J.P. OConnell, The Properties of Gases and Liquids 5th Ed., Mc Graw Hill, New York (2001) 10. Dymond, E.H., and E.B. Smith, and Mixtures Clarendon, Oxford (1980) APPENDIX A The experimental procedure to determine the solubility of a gas in a liquid using the apparatus shown in Figure 2 is as follows: 1) Connect the lecture bottle to the solubility apparatus by means of a pressure reducer. 2) With the valve of the lecture bottle closed, open the valve of the pressure reducer. 3) Switch on the vacuum pump and open stopcocks V1, V3, and V4 with V2 closed. Evacuate the whole ap paratus during 1 h. 4) Close the stopcocks. Remove the equilibrium vessel from the line, lowering the level of the thermostat bath in order for the connector of the equilibrium vessel to be out of the water. 5) Introduce ~ 6 cm 3 of water in the equilibrium vessel. 6) Connect EQ again to the vacuum line and re-establish the initial position of the thermostat bath. 7) In order to degas the water, open V2 and V4. Switch on the stirrer. This procedure lasts about 30 min until the measured pressure equals the solvent vapor pres sure at the equilibrium temperature. 8) Close stopcocks V2 and V4 and remove the equilib rium vessel again from the line. 9) Weigh the equilibrium vessel and then connect it to the vacuum line, re-establishing the level of water bath using the elevator. 10) Open stopcock V2, (with V3 and V4 closed). Then slowly open stopcock V1 to admit the gas to the equi librium vessel.

PAGE 37

Vol. 42, No. 3, Summer 2008 11) Slightly open the valve of the pressure reducer and then the valve of the lecture bottle in order to adjust the total pressure to ~ 1 atm (reference pressure). 12) Close stopcock V1 and open V3. 13) Switch on the magnetic stirrer to promote contact between the liquid and vapor phases. 14) The dissolution process lasts about 2 h. As the gas dissolves the pressure decreases, which is detected by PT. The piston comes slowly down to re-establish the reference pressure. 15) Record the last constant value displayed by PC, which is the number of pulses (n p ) of the linear actua tor. 16) To convert n p in displacement of the piston use the conversion factor indicated in the user manual of the linear actuator. The value obtained must be multiplied by the internal crosssection area of the cylindrical gas burette to obtain the volume of the dis APPENDIX B (The data presented in the following tables was obtained by a group of students.) Gas: CO 2 Solvent: water Raw data: n p is the number of encoder pulses; T is the equilibrium temperature; P is the reference pressure; m 1 is the mass of the solvent. T / K n p P / Pa m 1 / kg 298.39 82346 100070 0.007737 The value of n p placement of the piston, using the conversion factor indicated in the user manual of PC. Using Eq. (9) cal Write a simple computer program to calculate the partial pressure of 101325 Pa. Results: x 2 represents the mole fraction solubility calcu lated with Eqs. (9-17); 1 the fugacity of the solvent in the vapor phase obtained from Eq. (16); 1 the fugacity coef x 2 is the corrected solubility at partial pressure P 2 = 101325 Pa; H 2,1 What are the uncertainties of the measured variables? Using these uncertainties and the propagation law of mole fraction solubility. Which uncertainty most affects the solubility value? Using your own results of solubility data and some of the results of other groups (at least 6 pairs of experi of the equation: R ln x 2 = A + B/T + C ln T to represent the temperature dependence of the solu bility. Compare your solubility data with those of the litera ture. T / K x 2 /10 -4 1 x 2 / 10 -4 H 2,1 /(MPa) 298.39 5.88 0.9880 0.9985 5.98 169.4

PAGE 38

Chemical Engineering Education C hemical Engineering students at the University of Alberta of their third year. At this point they have a very limited background, having only taken introductory thermodynamics and a mass and energy balance course. The objectives of this course are to provide a bridge between theoretical study and practical applications, and to apply these principles to critical analysis of real experimental data in a professional, clearly written, and concise format. Furthermore, the experiment described in this and measuring devices. To teach entry-level chemical engineering students with limited theoretical and statistical analysis background to write technical reports and apply material and energy balance principles to a critical analysis of real data, it is necessary to use a simple experi ment. Most previous studies involving the mixing of heated water require dynamic analysis of stirred-tank heaters; [1, 2] however, students with a limited background would have trouble with the theory of such systems. A simpler experiment is the mixing of hot and cold water at a T-junction. This experiment can be used to demonstrate how to use steady-state material and energy balances of the mixed water stream. Also, this experiment emphasizes the importance of properly placing process measurement devices, i.e. the thermocouples in the current experiment. Furthermore, experimental streams. It is important for students to learn the importance of proper calibration in process measuring devices since calibration equations can change over time due to corrosion, erosion, or scale buildup during its use. THEORY For any given continuous, nonreactive process at steady-state, the general material and energy balances can be written as (Felder and Rousseau [3] ): mm QW mh U gz out in out out out out ( ) 0 1 2 2 m h U gz in in in in 2 2 () 2 ChE laboratoryMIXING HOT AND COLD WA TER STREAMS A T A T-JUNCTIONDAVID SHARP, MINGQIAN ZHANG, ZHENGHE XU, JIM RYAN, SIEGHARD WANKE, AND ARTIN AFACAN University of Alberta Edmonton, Alberta, Canada T6G 2G6 Copyright ChE Division of ASEE 2008

PAGE 39

Vol. 42, No. 3, Summer 2008 In the present study, hotand cold-water streams are mixed at a T-junction to produce one mixed stream. Assuming the hotand cold-water streams are completely mixed at the Tjunction, the system is adiabatic, work is neither done by or to the system, there are no frictional losses, and that kinetic and potential energy changes are negligible, Eq. (1) and Eq. (2) can be reduced to mm m mh mh mi xp re dc old hot mi xm eas mi xc ol d , () 3 c cold hot hot mh () 4 (Cengel and Boles [4] ): hC TT P () () 0 5 By setting the reference temperature to 0C and assuming the heat capacity, C p is constant for water in the temperature range investigated in the experiment (15 to 50C), the tem perature of the mixed stream could be predicted by combining Eq. (4) and Eq. (5) to give T mT mT m pr ed cold cold hot hot mi xm eas () 6 The predicted temperature of the mixed stream using Eq. (6) is dependent upon measured information of the streams before and after the T-junction. In order to determine which solved simultaneously to predict the mixed stream tempera assumed to be wrong, the temperature of the mixed stream Eq. (3) to give the predicted temperature as T mm Tm T pr ed no m mi xm eas hot cold hot ho co ld , t t mi xm eas m () 7 assumed to be wrong, the predicted mixed stream temperature can be calculated by T mT mm pr ed no m cold cold mi xm eas cold hot , ( ) T m T mT hot mi xm eas pr ed no m cold cold mix , () 8 mT mm hot hot cold hot () () 9 (7), Eq. (8), and Eq. (9) will all give the same value for T pred,mix If one calibration equation is incorrect, however, then only one of those three equations will accurately predict the mixed stream temperature, which will agree with the measured temperatures after the T-junction and therefore indicate that incorrect. EXPERIMENT AL SETUP AND PROCEDURE The experimental setup is shown in Figure 1. The entire setup is constructed using half-inch nominal copper pipe, domestic hot and cold water supply lines. The hot and cold water streams are mixed at a T-junction before exiting into the either of the inlet lines by setting the appropriate valve com controlled using globe valves and can be roughly set using 325 mm 330 mm 60 mm 190 mm 150 mm 200 mm 1. Globe valve 2. Pressure gauge 3. Orifice meter 4. Differential pressure cell 5. J-type thermocouple OPTO 22 3 T T T To drain 3 4 5 5 5 1 3 2 T 4 5 T P 1 2 4 5 m cold m hot m mix T cold T hot Tmix1 Tmix2 Tmix3 P 175 mm 340 mm 168 mm hot cold Figure 1. Schematic Diagram.

PAGE 40

Chemical Engineering Education and mixed streams are measured using a combination of an The temperature of the mixed stream is measured at 60 mm (Tmix1), 250 mm (Tmix2), and 400 mm (Tmix3) after the T-junction. This is done to determine the proper location for the thermocouple in order to measure the correct temperature of the mixed stream. The detailed locations of other thermo also shown in Figure 1. Analog signals from the differential pressure cells and thermocouples are converted to digital sig nals using an OPTO 22 system. These signals are sent to the personal computer, where they are stored and displayed using LabView (version7.1) software. In this experiment, the mass in terms of volts and temperatures are recorded in terms of m V m V m cold cold hot hot m 0 0265 10 0 0251 11 ( ) ( ) i ix m eas mi x V ( ) 0 0473 12 All three calibration equations have a systematic error the hot and cold water must be set using the globe valves in conjunction with the pressure gauges. The data acquisition can be assumed for that run. At least 100 seconds worth of data for each run should be recorded to ensure the system reasonable analysis. The same experimental procedure is possible (this depends on class size and laboratory availabil ity), and if reproducibility is to be examined then at least one experiment can be completed in 10 to 30 minutes, depending on the number of runs students conduct. The short time span enables even large classes to do individual experiments in a rather short time period. The data is recorded in a Microsoft age readings from the DP/cells and the temperatures in C from the thermocouples. RESUL TS AND DISCUSSION material and energy balances. Additionally, four more runs were conducted to show the reproducibility of the data. When comparing experimental and predicted results it is necessary to do an error analysis on the variables being compared. The total error for an experimental value can be determined from the sum of the systematic (accuracy) and random (precision) errors of the data. The accuracy error comes from the maxi mum absolute error in calibration of the measuring device. For this study, the maximum absolute errors in the calibration are 5% and 0.3C, respectively. The precision error can measured values. To determine the precision error, Coleman and Steele [5] state that when the number of data points for one time series is equal to or greater than 10, two times the standard deviation gives a good approximation for the 95 % m T m 00 52 13 03 2 1 4 ( ) ( ) In order to illustrate how material and energy balances can be used to determine an incorrect calibration and to for the cold stream was changed to (without the students knowledge) m V cold cold 0 0168 15 ( ) as a function of time, respectively. The measured cold, hot and Eq. (15), Eq. (11), and Eq. (12), respectively. The predicted (3). For these calculations the average hotand cold-water dicted temperature of the mixed stream was calculated using for the hot, cold, and mixed stream water streams. The un the mixed stream were calculated using the method described by Coleman and Steele [5] and Holman. [6] The uncertainty in can be calculated by m mi xp re d cold m mix pr ed co ld m m , 2 2 m m mi xp re d hot m hot 12 16 / () Similarly, the uncertainty in the predicted temperatures obtained from Eq. (6) to Eq. (9) can be determined from the

PAGE 41

Vol. 42, No. 3, Summer 2008 as well as the measured hot and cold stream temperatures. The uncertainty in the predicted temperature is given by T pr ed hot m p pr ed hot T m T 2 r re d cold m pr ed mi x m T m co ld 2 ,, m eas m pr ed hot mix m eas T m 2 T pr ed hot T hot hot T m 2 2 12 17 / () steady-state periods having time ranges of 0 to 130 s, 155 to 325 s, 390 to 625 s, 675 to 835 s, and 885 to 1135 s. The the measured values in Figure 2 indicate that the system was indeed behaving at steady-state also evident in Figure 3 where the measured temperatures before and after the T-junction remain constant for each stream and steady-state period. Since the system is non-reactive and is at steady-state, the material balance equation, shown in Eq. (3) should be valid. Figure 2, however, clearly shows that mixed stream does not agree, within error, with the measured conclusion that one, or more, of must be incorrect. The mixed stream temperature when measured only 60 mm (Tmix1) downstream of the T-junction has larger error bars and is always approximately 1C lower compared to the temperatures measured at 270 mm (Tmix2) and 400 mm (Tmix3) downstream of the Tjunction, as shown in Figure 3. The error bars were calculated by assuming each thermocouple has the same systematic 0 50 100 150 200 250 300 350 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 Mass flow rate, m (kg/h) Time, t (s) measured hot stream measured cold stream measured mixed stream predicted mixed stream Figure 2. Mass ow rates of hot, cold, and measured and predicted mixed streams for ve steady-state runs. 15 20 25 30 35 40 45 50 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 Temperature, T ( o C) Time, t (s) Tmix1 Tmix2 Tmix3 hot cold predicted Figure 3. Temperatures of hot, cold, measured mixed and predicted mixed water streams for ve steady-state runs.

PAGE 42

Chemical Engineering Education (accuracy) error of 0.3 C. Therefore, the larger total error determined at the Tmix1 loca tion must be only due to higher standard deviations. This, com bined with the fact that the measured average temperature reading at the Tmix1 location is lower than the temperatures measured at the Tmix2 and Tmix3 locations, indicates that at the thermocouple closest to the T-junction the two streams have not completely mixed. Furthermore, the measured temperatures at the Tmix2 and Tmix3 locations are not only very similar, but also have very similar size error bars, which indicates that a thermocouple needs to be placed a minimum of 270 mm downstream of the T-junction to ensure complete mixing. Figure 3 also shows that the predicted temperature does not agree within error with the measured runs, again indicating that there is likely an incorrect calibration equation. For each run, the difference between the measured and predicted temperatures is decreasing until the values agree within experimental error. This agreement is likely due to the decreasing cold on the predicted temperature of the mixed stream. To determine which calibration equation was incorrect, material and energy balances were solved simultaneously to predict the mixed stream temperature, which can be compared with the measured values. Table 1 shows the average tem perature measured at 270 mm downstream of the T-junction (Tmix2) as well as three predicted mixed stream temperatures the energy balance equation. The mixed stream temperature was calculated using Eq. (7), Eq. (8), or Eq. (9) by eliminating streams, respectively. From this table, it is evident that the only predicted temperature that agrees within error of the cold is eliminated from the rate calibration equation was incorrect. Using linear regression obtained from Eq. (3) and the corresponding average volt calibration equation was determined to be m V cold cold 0 0268 18 ( ) This equation is valid for voltage readings between 0.5 The R 2 value is 0.996, so the calibration equation should be quite accurate. Students can also use error analysis to explain or discuss trends observed in the experimental data. Table 2 shows the streams with total errors and the corresponding standard de viations for each steady-state run. From this table, it can be seen that the total error in the measured hot and cold streams total error for the experimental values was determined from both the accuracy and precision of the data, the total error T ABLE 2 Run m hot (kg/h) m cold (kg/h) m mix (kg/h) 1 Standard deviation 67.0 5.7 1.2 131.5 9.1 1.3 275.7 16.4 1.3 2 Standard deviation 108.6 8.2 1.4 116.7 8.1 1.2 292.9 18.9 2.1 3 Standard deviation 145.0 11.2 2.0 94.7 6.4 0.8 298.4 19.5 2.3 4 Standard deviation 173.7 12.1 1.7 71.7 5.1 0.8 293.7 18.4 1.9 5 Standard deviation 205.5 16.2 3.0 44.0 3.6 0.7 272.1 18.6 2.5 T ABLE 3 Reproducibility of the inlet conditions Run m hot (kg/h) m cold (kg/h) T hot (C) T cold (C) 3 145.0 11.2 94.7 6.4 42.8 1.1 22.5 0.9 6 145.6 11.6 99.1 8.6 41.7 0.6 22.8 0.9 7 144.1 9.0 97.3 7.2 41.9 0.8 22.8 0.8 8 144.4 13.2 97.4 6.9 42.1 0.5 23.1 0.6 9 148.0 10.1 96.5 7.6 42.7 0.8 22.7 0.9 T ABLE 1 Run Tmix2 (C) T pr ed no m cold (C) T pr ed no m hot (C) T pr ed no m mi x (C) 1 27.9 0.5 27.6 0.7 33.5 1.0 29.4 0.7 2 30.3 0.6 30.2 1.0 34.9 1.0 32.5 0.8 3 32.4 0.8 32.4 1.2 36.4 1.0 34.8 0.9 4 34.1 0.7 34.1 1.3 37.3 1.0 36.4 0.9 5 36.5 0.6 36.9 1.6 38.5 0.7 38.2 0.7

PAGE 43

Vol. 42, No. 3, Summer 2008 increase in the measured values is due to the increase in both shown in Table 2. Similar observations and discussions can be made for the mixed stream. The reproducibility of the study was also examined by re peating one of the runs four times. The time average values of readjusted between runs and reset to the same value. Even is very crude (a globe valve and a pressure gauge) and both streams were connected to the domestic cold and hot water and for the hot stream are all within experimental error of tings. Table 3 also shows that the hot and cold temperatures agree within experimental error and the supply temperatures remained relatively constant for all repeat runs. It should be noted that the experiment outlined in this re port is only one of many possible ways in which the students can be asked to analyze this system. A few other examples include: 1) the thermocouples could be setup so that one or more was malfunctioning, with students asked to determine which one(s) are malfunctioning and why; 2) only one or two calibration equations could be given to the students and they then asked to determine the unknown ones; 3) all given information could be correct allowing the students to test material and energy balance principles used to predict a mixed and number of runs conducted, more or less emphasis could be placed on reproducibility of data and/or error analysis. To allows the instructor(s) to vary the experiment from year to year, while retaining the fundamentals. CONCLUSIONS A simple mixing of a hotand cold-water stream at a T-junc tion was investigated. The main objective was to use mass the temperature of the mixed stream after the T-junction, and then compare these with the measured values. Furthermore, the thermocouple location after the T-junction and the repro ducibility of the data were also investigated. calculated using mass balance equations did not agree with equation was wrong, mass and energy balance equations were solved simultaneously to predicte mixed stream temperature. eliminated from the energy balance, the predicted mixed stream temperature was found to agree with all three measured mixed stream temperatures within experimental error for all calibration was wrong. The mixed stream temperature measured at 60 mm (Tmix1) had a higher standard deviation error than the temperatures measured at 230 mm (Tmix2) and 400 mm (Tmix3) down that the temperatures measured at Tmix2 and Tmix3 locations had similar absolute and standard deviations and error values. Both observations indicated incomplete mixing at the Tmix1 location. Therefore, to ensure complete mixing and minimize heat losses, the thermocouple should be placed at least 230 mm downstream of the T-junction. The reproducibility of the experimental data was also stud ied by repeating one of the runs four times. It was found that of each other, verifying the reproducibility of the hot and cold SUMMARY In this paper, we proposed a simple experiment of mixing a hotand cold-water stream at a Tjunction to demonstrate how to use steady-state material and energy balance principles in troubleshooting of an existing process and determining the integrity and/or location of the measuring devices, such relatively inexpensive, requires little time to complete and is conceptually simple to understand, making it ideal for the undergraduate students who have a very limited chemical engineering background. NOMENCLUTURE C p g Gravitational acceleration, m/s 2 h Enthalpy, J/kg Q Heat transfer rate, W T Temperature, C T 0 Reference Temperature, C U Velocity, m/s W Work input to the system, W z Elevation, m REFERENCES 1. Romagnoli, J.A., A. Palazoglu, and S. Whitaker, Dynamics of StirredTank Heater Intuition and Analysis, Chem. Eng. Ed. 35 46-49 (2001) 2. Muske, K.R., Experience with Model Predictive Control in the Un

PAGE 44

Chemical Engineering Education 160 dergraduate Laboratory, Comput. Appl. Eng. Educ. 13 (1), 40-47 (2005) 3. Felder, R.M., and R.W. Rousseau, Elementary Principles of Chemical Processes 3rd Ed., John Wiley & Sons, New York, NY, pp. 86 and 323 (2000) 4. Cengel, Y.A., and M.A. Boles, Thermodynamics: An Engineering Approach McGrawHill, New York, NY, p. 113 (1989) 5. Coleman H.W., and W. G. Steele, Experimentation and Uncertainty Analysis for Engineers 2nd Ed., John Wiley & Sons, New York, NY, pp 30 (1999) 6. Holman, J.P., Experimental Methods for Engineers 7th Ed., McGraw-Hill, New York, NY, p. 51 (2001) APPENDIX A: SAMPLE DATA, CALCULATIONS AND ERROR ANALYSIS The raw data in terms of voltage vs. time and temperature vs. time are shown in Figures A1 and A2. known voltages (Figure A1): Sample calculations based on the readings recorded at a time of 10 s. Hot stream voltage = 0.547 V Cold stream voltage = 4.80 V Mixed stream voltage = 2.90 V 0 1 2 3 4 5 6 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 Voltage readings, V (V) Time, t (s) hot stream cold stream mixed stream 15 20 25 30 35 40 45 50 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 Temperature, T ( o C) Time, t (s) Tmix1 Tmix2 Tmix3 hot cold Figure A1. Voltage outputs from the DP/cells for the hot, cold, and measured mixed streams for all nine steady-state runs. Figure A2. Temperatures of hot, cold, and measured mixed water streams for all nine steady-state runs. (12) m V kg s m cold cold hot 0 0168 0 0168 48 00 0368 . .. / 0 0251 0 0251 0 547 0 0186 . .. / V k gs m hot mi xm ea s s m ix V k gs 0 0473 0 0473 29 0 0805 . .. / To teach entry-level chemical engineering students with limited theoretical and statistical analysis background to write technical reports and apply material and energy balance principles to a critical analysis of real data, it is necessary to use a simple experiment.

PAGE 45

Vol. 42, No. 3, Summer 2008 161 mm mk gs kg s mi xp re dc old hot ./ ./ 0 0365 0 0186 0 0 0551 ./ kg s The predicted temperature of the mixed stream was calculated from Eq. (6) using average values for run 1. T mT mT m pr ed cold cold hot hot mi xm eas 0 0365 2 2 29 0 0186 42 2 0 0766 21 2 .. . Assuming that the cold stream calibration equation is wrong, Eq. (7) can be used to predict the temperature from the average T mm Tm T pr ed no m mi xm eas hot cold hot ho co ld , t t mi xm eas m .. .. 0 0766 0 0186 22 90 0186 42 2 0 0 0766 27 6 . C Assuming that the hot stream calibration equation is wrong, Eq. (8) can be used to predict the temperature from the average T mT mm T pr ed no m cold cold mi xm eas cold h hot , o ot mi xm eas m .. .. 0 0365 22 90 0766 0 0356 42 2 2 0 0766 33 5 . C Assuming that the mixed stream calibration equation is wrong, Eq, (9) can be used to predict the temperature from the aver T mT mT mm pr ed no m cold cold hot hot cold hot mix 0 0365 22 90 0186 42 2 0 0365 0 0186 29 .. .. .. .4 4 C mined that the standard deviations were 0.000355 kg/s, 0.000325 kg/s, and 0.000367 kg/s for the cold, hot, and mixed streams, Cold Stream m m c ol dm co ld co ld :. 00 52 00 50 0365 20 000355 0 00254 .. . / kg s H Ho tS tr eam m m hot m hot hot :. 00 52 00 05 0 0186 20 000325 0 00158 .. ./ kg s Mi xe ed Stream m m m ix m eas m mix m eas mix : , 00 52 m m eas 00 50 0766 20 000367 .. 0 0 00456 ./ kg s standard deviations were 0.20C, 0.19C, 0.40C, 0.12C, and 0.09C for the cold, hot, mix1, mix2, and mix3 thermocouples, Cold ther mocoupl e T T co ld co ld :. 03 2 03 20 20 07 0 .. : C Ho tt he rm ocoupl e T T T hot hot 03 20 32 01 9 . 0 06 8 1 0 32 1 1 : C Mi xt he rm ocoupl e Tmix Tmix 03 20 40 11 0 2 .. C Mi xt he rm oc coupl e Tmix Tmix : .. 2 2 03 20 32 0 1 12 05 4 3 0 3 : C Mi xt he rm ocoupl e Tmix .. .. 32 03 20 09 04 8 3 Tmix C

PAGE 46

Chemical Engineering Education 162 m mi xp re d cold m mix pr ed co ld m m , 2 2 m m mi xp re d hot m hot 12 / m m m mix pr ed co ld hot / 2 2 12 0 0 00254 0 00158 0 0030 2 2 12 . ./ / kg s The uncertainty in the predicted temperatures from Eqs. (6) to (9) is given by T pr ed hot m p pr ed hot T m T 2 r re d cold m pr ed mi x m T m co ld 2 ,, m eas m pr ed hot mix m eas T T 2 T pr ed cold T hot co ld T T 2 2 12 / For the uncertainty in Eq. (6) the partial derivatives are T m T m pr ed hot hot mi xm eas . 42 2 0 0766 550 9 C C kg s T m T m pr ed cold cold mi xm eas / . 22 9 0 076 6 6 299 0 / C kg s T m mT m pr ed mi xm eas hot hot c o ol dc ol d mi xm eas T M .. 2 0 0186 42 20 0365 22 9 0 0766 276 2 2 . / C kg s T T m pr ed hot h hot mi xm eas pr ed co m T T . 0 0186 0 0766 0 2428 l ld cold mi xm eas m m . 00 365 0 0766 0 4765 Then, the uncertainty in T pred is T pr ed 550 90 00158 299 00 00254 2 2 .. .. 2 276 20 00456 0 2428 06 80 4765 07 2 2 .. .. .. 0 01 7 2 C For the uncertainty in Eq. (7) the partial derivatives are T m TT m pr ed no m hot hot cold mi xm eas co ld , 42 2 22 29 0 0766 252 0 . / C kg s T m pr ed no m co co ld l ld pr ed no m mi xm eas hot cold ho T m mT T co ld 0 , t t mi xm eas m .. . 2 2 0 0186 22 94 22 0 0766 61 2 / , C kg s T T m m pr ed no m hot hot mi xm co ld e eas pr ed no m cold T T hot 0 0186 0 0766 0 2428 . mm m mi xm eas hot mi xm eas , .. 0 0766 0 0186 00 7 6 66 0 7572 Then, the uncertainty in T pred is T pr ed no m co ld . 00 00158 252 00 0025 2 4 4 120 10 00456 0 5234 06 80 476 2 2 2 .. .. .5 50 70 10 2 12 .. / C

PAGE 47

Vol. 42, No. 3, Summer 2008 163 For the uncertainty in Eq. (8) the partial derivatives are T m T m T pr ed no m hot pr ed no m cold co hot hot , 0 l ld hot mi xm eas T m C kg .. / 22 94 22 0 0766 252 s s T m mT T pr ed no m mi xm eas cold hot cold hot , m m mi xm eas .. . 2 2 0 0365 42 22 29 0 0766 12 0 01 / , C kg s T T mm pr ed no m hot mi xm eas co l hot d d mi xm eas pr e m T .. . 0 0766 0 0365 0 0766 0 5234 d dn om cold cold mi xm eas co ld T m m , . 0 0365 0 076 6 6 0 4765 Then, the uncertainty in T pred is T pr ed no m hot . 00 00158 252 00 00254 2 2 2 2 120 10 00456 0 5234 06 80 4765 .. .. .0 07 01 0 2 12 .. / C For the uncertainty in Eq. (9) the partial derivatives are T m mm Tm T pr ed no m hot cold hot hot hot ho mix t tc ol dc ol d cold hot mT mm 2 0 0368 0 0186 .. 42 20 0186 42 20 0368 22 9 0 0368 00 1 .. .. .. 8 86 231 41 2 / C kg s T m m pr ed no m cold hot mix mT mT mT mm cold cold hot hot cold cold cold h hot 2 0 0186 0 0368 22 90 0186 42 20 0 .. .. .. 3 368 22 9 0 0368 0 0186 117 0 2 .. / C kg s T p r re dn om mi xm eas pr ed no m hot mix mix m T T m , 0 h hot cold hot mm T 0 0186 0 0368 0 0186 0 335 7 .. p pr ed no m cold cold cold hot co ld T m mm 0 0368 0 .. . 0368 0 0186 0 6643 Then, the uncertainty in T pred is T pr ed no m hot .. .. 231 41 0 00158 117 00 0 2 0 0254 00 00456 0 3357 06 90 6643 2 2 2 . .. 07 10 7 2 .. C This experiment is relatively inexpensive, requires little time to complete, and is conceptually simple to understand, making it ideal for the undergraduate students who have a very limited chemical engineering background.

PAGE 48

Chemical Engineering Education 164 mm m cold co rr ected mi xm easur ed hot , 0 0766 0. .. / 0186 0 058 kg s calibration equation is m V cold cold 0 0268 T ABLE A1 Corrected Mass Flow Rates and Corresponding Voltage Readings for the Cold Stream. run cold, corrected (kg/s) Average cold stream voltage, V cold 1 0. 058 4.7 2 0. 051 3.7 3 0. 043 2.5 4 0. 033 1.4 5 0. 019 0.5