Chemical engineering education

Material Information

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


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


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

Record Information

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

UFDC Membership

Chemical Engineering Documents


This item has the following downloads:

Full Text


.. 0 ..... (..) I.. 'll 'll .. 0,() I.. ,:, 'll (..) 0 .. i::s (..) ., I.. "' 'll I.. !: 'll 'll .. :: 0.() 0 "' ;,. (..) Cl !: 'll -:: I.. I.,) 'll 'll ...... o .. 'll 0,() ..... :: .... ..... "' i::s (..) !: .. 'll i::s -:: (..) I.,) I.. 'll !: r. ., _ -.. 1, i chemical engineering education Particle Dynamics in Fluidization and Fluid-Particle Systems: Teaching Examples. Liang-Shih Fun of II Old Purdue" [I Toward Technical Understanding: Part 5. General Hierarchy Applied to Engineering Education (p.138) ............................................ Haile [I Random Thoughts: The Scholarship of Teaching (p. 144) ........................... Felda [I Teaching PDE-Based Modeling to ChE Undergraduates: Overcoming Conceptual and Computational Bariers (p. 146) ............ Tlw111p.1011 [I An '"Open-Ended Estimation" Design Project for Thermodynamics Students (p. 154) ......................................... .......... Lo111bardo [I Low-Cost Mass Transfer Experiments: Part 6. Determination of Vapor Diffusion Coefficient (p. 158) ................ ..... Nirdosh. Garred. Baird [I Is Matter Converted to Energy in Reactions'? (p. 168) .............................. Andersen [I Incorporating Molecular Modeling into the ChE Curriculum(p. 162) ............................................ Baldl!'in. Ely, War. Daniel [I A Laboratory for Gaseous Diffusion through Permeable Solids: The Time Lag (p. I 72) ....................................... Dufaud, Fm Te, Vincent [I Use of an Emission Analyzer to Demonstrate Basic Principles (p. 178) ....... Lodge ... and chemical engineering of THE FUTURE OF ENGINEERING EDUCATION Developing Critical Skills (p.108) Woods, Felder Rugarcia, Stice Learning How to Teach (p. 118) Stice. Felder, Woods, Rugarcia


2000 ASEE Annual Conf ere nee Chemical Engineering Division Program Jun e 1 8 2 1 20 00 St. L o u is, M O Pr e Co nf e r e n ce W ork s h o p "Modern ChE and ME L abora t ory C h E Di visi on L e ctur es hip In strumen t a ti on" Monday Jun e 19, 4:30-6:00 pm Sunday Jun e 1 8, 8:30 am t o 4: 1 5 pm Moderator Mi chae l Cutlip Coordinators: W as hin g t o n University Jim Henr y and C har l es Knight Workshop Cost $65 Techn i cal Sess ion s Sess i o n 2213, "C h E In s t r u c ti o n in th e F utu re Tue sday, Jun e 20, 8:30 -10:15 a m : CoM o d era t o r s Kirk Schulz a nd Christopher Wi ege n ste in P aper I : "Approac h es o r L earning a nd L ea rnin g Environments and Lecture Evironments, Donald Woods Andrew Hrym ak, and H ea th er Wright Pap er 2: L ect ur es o r Electrons: Which W o rk s B e tter for C h em i cal Engineering Fundamen tal s C la ss?" Billy Cry n es, Connie and Barb ara Gr ee n e Pap er 3: An I ndu c tiv e Approach t o T eac hin g H ea t a nd M ass Transfer ," Robert H eske th and Stephanie Farrell Pap e r 4 : "Expa nd able P o l ys t y r e n e Batch Reactor D esig n : A n Academic/Industrial Collabora tion in Teaching Reaction E n gi n ee rin g," R obert Baral a n d R o n a ld Gabbard Pa per 5: 'The Student Co n sultant: Enha n cing Communica ti on Skills in the Undergraduate L a b oratory," D en ni s Mill er Sess i o n 23 1 3, I ns tru c ti o n a l Tec hn o l ogy: T h e F utu re of C h E In st ru c ti on?" Tu esday Jun e 20 10:30 am-12:00 pm : Co-Moderators Thoma Edgar a nd Scott F ogler P aper I: D eve l opment of a n Extended Campus Chemical Engineering Pro gram, Jimm y Smart William Murph y, G. Lin eberry a nd Bonit a Lykin s P aper 2: "Molec ul a r Simulation Via Web-Based In struc ti on ," Peter Cumming s, D avid Kofke and Ri c hard R ow le y Pap er 3: "A nal ys i s of I n st ructi o nal T ec hn o l ogy Usage in th e Introductor y Chemical Engineeering Cour se," Richard F e ld e r Amy Michel, and Jan Genzer Pap er 4: In fo rm a t ion Technology and C h emica l Engineering Education: Evol uti on or R evo luti on?" Thom as Edgar Sessio n 25 1 3 T h e G r ee nin g of th e C h E C u r ri c ulu m Tu es da y, Jun e 20, 2:30 4: 15 pm : Co-Moderators Denni s Sourlas a n d Ashish Gupta P aper / : "Produc ti o n of Clean Fu e l : A Bi oc hemic a l Experiment for Unit Operation s Lab ora tory D eve l oped Through Undergrad u a t e Research Project s," Muthanna Al-Dahhan P aper 2: Minimizing the Environmental Imp act of C h emica l Manufacturing Pr ocesses Jo se ph Shaeiwitz, Ro ger Schmitz, M a rk M cC re ady, Joan Br e nn ecke, M ark Stadtherr Ri c h a rd Turton and Wallace Whiting P aper 3: D eve lopm e nt of a n Elective Course o n Pollution Pr eve nti on," D en ni s Sourlas a nd Ashish Gupta Sess i on 2613. I mp l e m e ntin g Soft Sk ill s I nto ChE Curr i c ulum Tuesday June 20, 4:30-6:00 pm: Co-Modera t ors D ouglas Ludl ow and James Newell P aper I: Inte gra tin g Soft Criteria into the C urri cul um W. Nicholas D e lg ass, Philip W a n kat and Frank Oreovicz Pap e r 2: Trainin g in Multidisciplinarianism, D a ina Bri ed i s and R Mark Worden Pap er 3: "An Indu s trial Int erns hip Program t o Enhance Student Learning a nd Marketability, Zenida K ei l and Melanie Basanti s P aper 4: "An In vestigatio n of the Communication Culture of a n Intr oductory Chemical Engineering Class H eat h er Cornell, Wade Kenn y, a nd Kevin M ye r s Soc ial Activ i ties C h E D iv i sio n Aw ard s Banqu e t Tuesda y, Jun e 20 6:30-8:00 pm ChE Division Reception Moderator Mi chael Cut lip M onday, Jun e 1 9, 6:308:00 pm Missouri B o tani ca l Garden Spink Pavillion Moderator R obert Ybarra Guest Speaker Dr. P eter R ave n Washington U n ivers it y Director of Missouri Botanical Garden Cost: $48 P aper 5: Th e Bu s in ess Meeting: A n Alternati ve to the C la ss i c De s i g n Presentation ," Jame s Newell Sess i o n 33 1 3, "T h e F utur e o f E n g in ee rin g E du ca t io n W ed ne sday June 21 10:30 am-12:00 pm: Plenary Session Moderator Dendy Sloan Speakers: Donald Woods Intrin sic a nd Extrinsic Re wa rd s of Teaching Excellence "; Richard F e ld er, 'Teaching M e th o d s That Work ; J a m es Stice, Leaming H ow T o T eac h "; and Armando Ru ga r c ia A Vi s i on for A New Century. Sessio n 34 1 3, "ChE Labo r a t or i es i n the Next M ill e nniu m Wedne sday, June 2 1 12:30-2: 1 5 pm: Co-Moderator s Ri c har d Gilbert and Steve LeBlanc Pap er I : "A Laboratory for Enh a ncing Proce ss Co nt rol Courses Using Real-Time MATLAB / Simulink ," B abu Joseph D ee pak Srinivasagupta and Chao -Ming Yin g Pap er 2: "A Fluidiz e d Pol y m er Coating Experiment ," C. Stewart Slater Rob e rt H esket h and Mi c hael Camey Paper 3: Enhancement of In s trumentation and Pro cess Contro l Studies at the U nd ergrad u ate Level," Rebbec ca Toghiani H osse in To ghiani Donald Hill and Craig Wierenga P aper 4: D eve l o pment of Unit Operation s Fermentation L abora t ory Experiment Us in g Indu s trial Collaboration ," G Dale Wes so n William Muth Bryan Land en and Egwu Kalu Pa per 5: Introdu cing Fre s hmen to Drug Deliver y," Stephanie Farrell and Robert H eske th Paper 6: In corporation of Graduate Facilities I nto Undergraduate U nit Operation s Labora tory ," Muthanna AI-Dahhan Sess i o n 3513, "ChE Ed u cat i o n : How D o We Assess It?" Wedn esday, Jun e 2 1 2:30 -4 : 15 pm: Co-Moderator s Dain a Bri edis and Su s a n M o nt gomery P aper I : Student P ortfo li os: Asses s ing Criteria 2000 ," Caro l y n e Garcia and Edgar Clau se n P aper 2: Th e Pro cess of L ea rnin g Chemical Engineering : What Work s a nd What Doe s n t ," David Dibi as i o William Clark Anthony Di xon, and Lisa Comparini P aper 3: A ssess ing Chemical Engineering Education A s It I s Delivered ," Joseph Sha e i witz Paper 4: "P rin cipal Objects of Knowledge ( POK 's) in Colloquial Approac h Enviro nm ents, Pedro Arce Session 3613 "C h E, Comp u ters a n d t h e Next M ill e nni um Wedne s d ay, Jun e 21 4:30-6:00 pm : Co-Mo d erato r s Skip R oc h efort a nd Valerie Young Paper I : A Virtual Reality-Based Safety and H azard Analysi s Simulation, John B e ll and Scoll Fogler Pap e r 2: Combining Hi g h -Level P rogrammi n g and Spread s heet s: An Alternative Route for Teaching Pro cess S y nth esis a n d De sig n Jorge Gatic a, Mauricio Colombo, a n d Marla H e rn nde z P ape r 3: "24x7: Lab Experiment s Access o n th e Web All th e Time, Jim H e nr y Pap e r 4: MATLAB A ppli ca ti o n in Re actor D esign and Simulation ," Char le s Okonkwo and Gbekeloluw a O g untimien Pap er 5: Profe ss ion al Simul a tion P ackages a s Effective Teaching T oo l s in U ndergraduate ChE Curriculum ,'' D avid Di xon, Jan Pu szy n s ki. a nd Larry Bauer


EDITORIAL AND BUSINESS ADDRESS: Ch e mical Engineering Education Department of Chemical Engineering University of Florida Gainesville, FL 32611 PHO NE and FAX: 352-392-0861 e -mail : cee@c h e. ufl .ed u Web Pag e : http ://www.c /r e. 11fl .e d11 !ceel EDITOR T. ]. Anderson ASSOCIATE EDITOR Phillip C. Wankat MANAGING EDITOR Carole Yocum PROBLEM EDITOR James 0. Wilkes U. Michigan LEARNING IN INDUSTRY EDITOR William]. Koros University of Te xas, Austin PUBLICATIONS BOARD CHAIRMAN E. D e ndy Sloan Jr Co l orado School of Min es MEMBERS Pablo Debenedetti Princ eton Unive r sity Dianne Dorland Un i ve r sity of Minn esota, Duluth Thomas F. Edgar University of T exas at Austin Richard M Felder North Carolina State University Bruc e A. Finla yso n University of Washin g t on H. Scott Fogl e r University of Mi c hi gan William J Kara s University of Texas at Austin David F. Olli s North Carolina Sta t e University Angelo J. P e ma New J ersey In stitute of T ec hn ology Ronald W. Rous se au Georgia In stitute of T ec hnolo gy Stanley I. Sandler University of D e lawar e Richard C. Seagrave I owa State Un i versity M. Sami Selim Co l orado School of M i nes Stewart Slater Rowan University Jam es E. Stice University of Texas a t Aust in Donald R. Woods McMaster University Sprin g 2000 Chemical Engineer Volume 34 Numbe 1 EDUCATOR 98 G V. ( Re x) Rekl a i t i s of "O ld Purdu e," Philli p C. Wankat, Frank S. Oreovicz DEPARTME NT 102 Un i vers it y of A l berta: Tradition and Innovation J Fra se r Forbes Sieghard E. Wanke SPECIAL SECTION: THE FUTURE OF ENGINEERING EDUCATION 108 Part 3 Dev e l oping Critical Sk ill s Donald R Woods, Ri cha rd M. Felder Armando Ru garc ia, J ames E Stice 118 Part 4 L ear ning H ow to Teach, James E. Stice, Ri chard M. Felder, Donald R Woods, Armando Rugar c ia AWARD LECTURE 128 Particle D y n am i cs in F luidi za tion a n d F lui d-Part i c l e Sy s tems: Part 2. T eac hing Examp l es, Liang-Shih Fan LE ARNING 138 T oward Technical U nder s t an ding : Part 5. General Hierarchy App li ed t o Eng in eering Education J.M. Hail e RA N DOM THOUGHTS 144 The Scholarship of Teaching, Ri chard M. Felder COMPUTI N G 146 Teaching PDE-B ased Modeling to ChE Undergraduates: Overcoming Conceptua l and Co m putatio n a l Barier s, Karsten E. Thompson CLASS AND HOME PROBLEMS 154 An Open-Ended Estimation" De s i g n Proje c t for Thermodynamic s S tud ents, Stephen J. Lombardo CLASSROOM 158 Low-Cost Mass Transfer Experiments: Part 6. Determination of Vapor Diffu s i on Coefficient, / Nirdosh L.J. Garred M.H .I. Baird 168 ls Matter Converted to Energy in Re ac tion s? P aul K. Andersen CURRICULUM 162 In corporati n g Mol ec ular Mode lin g into the C h E Curr i c ulum Rob ert M. Baldwin, Jam es F. Ely, J D ouglas Way, Stephen R Dani e l LABORATORY 172 A Laborator y for Gaseo u s Diffu s ion through P e rm eab l e So lid s: The Time Lag, Oli v i er Dufaud, Eric Favre, Louis Mari e Vincent 178 Use of a n Emiss i on A n a l yze r to Demon s trate Basic Prin ci pl es, K e ith B. L odge 167 Letter to the Editor ChE Division Program ASEE Convention In side Front Cover Position s Available I nside Ba ck Cover Call for Paper s Outside B ack Cover C H EM I CAL ENGINEE RI NG EDUCATION ( I SSN 0009-2479) is publi s hed quarterly by the C h e mical E 11 giueer in g Divi sio n America n Soc i e t y for E 11 gfoeeri11g Ed u cation a n d is edited at the U ni versity of Florida. Co rr espo 11 de 11 ce regarding e,lito ri a l m a tt e r ci r c ulatio11 and c han ges of address s hould be se nt to CEE, C h e mi ca l Eng in ee rin g Deparhne11t U ni ve r s i ty of Fl o rida Ga in esv ill e, FL 326 11-6005 Co p y ri g ht 2 000 by th e C h e mi ca l E n g in ee rin g Division A m e ri ca n Society for E n gi n ee rin g Educatio n. The stateme nt s and opi ni o n s ex pr essed in thi s p er iodical are th ose of th e writers and 1101 n ecessa ril y th ose of th e C h E Division ASEE, w hi c h body assumes 11 0 r es ponsibility for th e m D e fective co pi es replaced if 11 0/ified wit hin I 20 da ys of publi ca tio11. Write for illf o rm ation 011 s ubscription cos t s a nd for back co p y cos t s and avai labilit y. POSTMASTER: Send add r ess c h a n ges to CEE, C h e mical Enginee ring Department. U niv e r s ity of Florida Gai n esvi ll e, FL 326 11-600 5. P eriod icals Po st ag e Paid at Gain esv ill e, Florida. 97


(.3.-6_.._e_d_u_c_a_to_r ________ __,,) G. V. (Rex) Reklaitis of ''Old Purdue'' P HILLIP C. W ANKAT, FRANKS. 0 REOVICZ Purdue University West Lafayette, IN 47907 P rofessor Rex Reklaiti s' CV i s impressive: Head at Purdue for twelve years author or editor of six book s and editor-in chief of Computer s & Ch e mical Engineering, with over 110 refereed publications and numerous se minar and conference pre se ntation s; involvement in the organization of 42 conferences and symposia; director of AIChE, past president and current trustee of CACHE; NSF postdoctoral fellow and Senior F u lbright Fe ll ow; winner of the ASEE ChE Divi s ion lectureship award and ASEE ChE Division Corcoran award, Fellow of AIChE, winner of AIChE Computing in Chemical Engineering award ; re searc h advisor for 37 MS st udent s and 28 PhD st udent s, and involvement in grants for over five million dollar s for research and over ten million dollars for the School of Chemical Engineering Listening to thi s i s enough to make any good Boilermaker sing Hail, Hail to Old Purdu e But what is Rex Reklaiti s reaJl y like ? Out of the upheaval of World War II in E u rope in 1942, the wor l d gained one Reklaitis Rex was born while his parents were traveling through German-occupied Poznan Poland on October 20-and lost another two years later when his father bec a me a casualty of the war. From then on hi s uncle s, and later hi s stepfa ther filled that role as part of the extended family that helped mold Rex They lived in Bavaria until moving to the United State s when he was ten. ,.. '-:; ,, J,,~ w,l " .'-. I ,, f .,,. ": --~ ~' --' ,, r''' Professor-turned-chef grills 'e m up at a departmental picnic Arriving in a new country at such a tender age, Rex was able to learn Engli s h much more quickly than i s the experience of most adults-and without any trace of an accent. He was a lread y bilingual in Lithuanian a nd Ger man. Learning a third l ang ua ge also helped make him aware of the importance of grammar and struct ure in language and before too long h e was speaki ng as well as, if not better than most native Americans He and hi s mother became naturalized citizens in 1958. After gradu ating in 1961 from St. Rita high schoo l in Chicago, he enro ll ed at the Illinois Institute of Technology [also alma mater of one author (FS O) and of the other author's father]. He graduated from IIT with a BS in chemical engineering in 1965 (Back then gra duating in four years was common.) While growing up in Chicago Rex became interested in classical music especially opera and l ear ned to play Copyright Ch Div i s i on of ASEE 2000 98 Che mi ca l Engineering E du c ati o n


Rex with Ron Barile (left) and Lowell Koppel (right) at a faculty lunch in December of 1976. Rex and Janine hosts at a party in their home in 1994 sharing a smile with Ramki Ramkrishna and Linda Wang T the mandolin and the clarinet-the love of mu s ic has remained with him through the years. He ran track and played tennis in high school, and hi s uncle s introduced him to carpentry and sailing. Sailing and skiing (w hich he fust did in high sc hool on old-fashioned wooden skis) have remained hi s two major hobbie s. He met Janine when they were 15 but only s he real ized that it was a fateful meeting He "awo ke when they later became rea cq uainted throu g h s tudent organiza tion s at the end of their so phomore years in college. When Rex decided to go to graduate sc hool at Stanford in 1965 Janine fo ll owed him and enrolled at the Uni versity of California-Berkeley. They were married on Augu st 20, 1966. Rex and Janine se ttled down in San Franci sco and faced the daily routine of Rex taking the Southern Pacific to Palo Alto while Janine drove across the B ay Bridge to Berkeley. Commuting lost its charm after a while, so in 1968 Janine tran sfe rred to Stanford and they moved to Palo Alto. Rex worked wit h Profes s or Doug Wilde at Stanford and also had significant interaction s with Pro fessors Andy Acrivos and David Mason He took four years to earn hi s master 's degree and no time at all to earn his PhD since both degrees were awarded in 1969. Spring 2000 Rex presents an award to Kristi Anseth (now a professor at the University of Colorado Boulder) at the annual Razz banquet in 1991. ( He claims that a sec r e tary forgot to tell him to turn in a form to claim hi s master 's!) From 1969 to 1970 Rex was an NSF Po s tdoctoral Fellow at the In s titut fUr Operation s Re sea rch und Elektron i sch Datenverarbeitung in Zurich Switzerland. He worked with the Institute director s, Professor s Kiinzi and Krelle on nonlinear optimization. Janine continued working on her PhD while they were in Switzerland. In the summer of 1970, Rex started as an Assistant Professor at Purdue a position he had accepted before starti n g his postdoctoral year in Switzerland He and Janine se ttled in We s t Lafayette and Rex started the interesting and challenging life of an assi s tant professor while Janine finished her PhD (which she received from Stanford in 1972 ) She subsequent l y accepted a position as assis tant professor of linguistic s at the Univer s ity of Illinois at Chi cago, and once again the Reklaitis family became comm uter s as 99


they moved to Crown Point Indiana which is essentially halfway between Chicago and Lafayette. The eleven years in Crown Point were most productive for Rex. He drove the hour to Purdue three days a week on Interstate 1-65, without a doubt one of the flattest and most boring highways east of the Mississippi He survived the drive by listening to classical music and opera se lection s on the car stereo as he planned the day 's events in the morning and proce sse d them as he returned home in the evening. The other two days he spent working at home on hi s books and other research. With few interruptions at home he became very prolific especia ll y since he watches very littl e televi sion and only occasionally sees a movie Little did he know then that administrative duties would eventually red u ce his research output. Living near Lake Michigan also provided more time for sailing and put Rex and Janine close to the Chicago Lyric Opera The famjly still purchases season tickets for the op era. Shakespeare forms another of hi s abiding interests. JOO Rex's first love-competing in the Chicago-to-Mackinac race, with his uncle aboard in 1979 (top photo, left) and again in 1980 in much rougher waters (bottom photo, left) ... ... and his second love skiing, here with Janine on the slopes in Breckenridge, Colorado, in 1999. Rex and Janine's olde s t so n George was born in 1974 and their youngest, Victor, in 1979 At Crown Point, Janine's grandmother and Rex 's uncle lived with them and helped rai se the two boys giving the boy s the experience and ad vantages of an extended family. The family moved back to West Lafayette in 1983 At Purdue Rex progre sse d s teadily up the ladder, being promoted to associate professor in 1976 and full profes so r in I 980 while he was on sa bbatical as a Senior Fulbright Lec turer at Vilniu s State University and the Lithuaruan Acad emy of Science in Vilniu s, Lithuania (which was then an unwilling part of the USSR). Realizing Rex 's special quali tie s, in 1985, Dean Henry Yang persuaded him to become the Assistant Dean of Engineering at Purdue. In addition to his other duties Rex was appointed as Interim Head of the School of Chemical Engineering in August of 1987 when Professor Ron Andres s tepped down as Head. The School then prepared to start a new Purdue tradition-a long internal/external search for a new Head Chemical Eng in ee ring Edu c ation


After a few weeks, though, it became evident to the chemical engineer ing professors on the committee that Rex was the ideal candidate They were able to persuade the dean to forego the exter nal search since it was clearly unnecessary. Thus, late in 1987 Rex officially hecame Head of the School of Chemical Engineering, an office he has since fulfilled with distinction Rex has been very suc cessful in hiring outstand ing new faculty and help ing them obtain NSF CA REER and other presti gious grants. He has de veloped an active Indus trial Advisory Council consisting of about twenty companies and The Reklaitis family, at home for Christmas 1998. design course, but despite his heavy schedule, enjoys morning meetings with student groups. He pa tiently pushes students to produce just a little bit better" results Then, dur ing the spring semester he interviews most of the graduating seniors. These "exit interviews" provide very useful feedback to the school and some of the ideas generated by them have been adopted. Rex has also been a strong advocate for student groups in chemical engi neering It is not unusual for the Reklaitis to enter tain up to 40 students from various groups in their home before Christmas vacation. Rex does his share by doing clean-up duty. frequently personally hosts the board members. The council is currently working with the faculty on five projects: con tinual renovation of equipment in the undergraduate labora tory, purchase of new research equipment hosting young faculty at their research facilities developing reali s tic de sign problem s and developing a survey of graduates and employers for an ABET outcomes assessment. After years of effort Rex has also obtained approval from the admin istration to add a new wing to the chemical engineering building that will increase it s space by sixty percent. This will allow for a much-needed increase in the size of the faculty. All chemical engineering has to do now is raise the money for this purpose! Throughout all this time, Rex's teaching and research continued unabated. A dedicated teacher and educator who has taught many of the courses in the curriculum he revised the introductory course in mass and energy balances at Purdue and developed new courses on "Optimization and "Com puter-Aided Process Design." His efforts on two of these courses resulted in two textbooks Introduction to Material and Energy Balances (Wiley, 1983) and En g ineering Opti mi z ation with A Ravindran and K. Ragsdell (Wiley 1983) Both books have sold around 9500 copies-very respectable numbers The second edition of the optimization book is currently being prepared. Undergraduate students have always found Rex s office door open to them. He takes great care in teaching the senior Sprin g 2000 Working with a team assembled by Professor Bob Squires, Rex developed educational modules using videotape and computer simulation s to give students a feel for complex processe s. The modules were generously supported by in dustrial contributions and the completed modules are being distributed by CACHE Several articles on these modules have been published, the most famous of which, "Purdue Industry Computer Simulation Modules: The Amoco Resid Hydrotreater Process (Ch e m Eng. Ed. 32, 98, 1991) by R.G. Squires G.V Reklaitis N.C Yeh, J B Mosby, I.A. Karimi and P.K. Andersen won the ASEE ChE Division Corcoran award for best paper that year. As a culmination of his interest in the use of computers in engineering education, Rex now s erve s on the editorial board for the journal Com put e r Appli c ations in Engineering Education. Rex s research in computer applications in chemical engi neering and batch processing is well known He developed implemented and demonstrated computer-aided methodol ogy for the design scheduling, and operation of batch pro ces s es. This research involved development of a modularly structured dynamic/discrete process simulator that defined the structure of batch-processing networks and generated preliminary equipment sizes In particular, his research high lighted the importance of intermediate storage and devel oped new scheduling algorithms for multiproduct plants. Rex has coauthored well over 110 research papers and -------------Continu e d on pag e 153. IOI


f.,SJ departmen t ) ------------=------UNIVERSITY OF ALBERTA Tradition I I I I I and I I I I I I I I I I j 1 I I I I Innovation I I I I I j I I I I I I J. FRASER FORBE S, SIEGHARD E. WANKE I I I I I I I I I i I I I I I I I I t '-~ I I 'I I I I I j Chemical & Materials Engineering Building University of Alberta Edmonton Alberta Canada T6G 2G6 C hemical engineering ha s played a key role in the development of Canada 's oil and gas and associated petrochemical indu s trie s, and c hemical engineering at the University of Alberta (UofA) has been an integral part of the growth of the petrochemical industry in western Canada The UofA i s located in Edmonton, capital city of the Province of Alberta. The western part of the province is part of the majestic Canadian Rockie s The Continental Divide makes up a s ignificant part of the western border between Alberta and British Columbia The southeastern part of the province is part of the Canadian Prairies while the north is part of Canada 's extensive boreal forest. To the south, Alberta borders on Montana-Big Sky Country Alberta also has a l ot of "big sky"; there are more hours of sunshine per year than in any other part of Canada. Oil and natural gas fields are found throughout the great sedimentary basin from the Alberta foothills in the west to the prairies in the east. Heavy oil and oil sand deposit s, which contain more hydrocarbon s than the Middle East are located north and east of Edmonton. The discovery of oil in the 1940s and the beginning of large-scale commercial de ve lopm ent of the oil sands in the 1970s had major impact s on chemical engi ne ering education in Alberta. The evol uti on and current state of the chemical engineering program with s ome gazing into the future, make up the heart of this article. HISTORY The UofA opened for classes in 1908, about three years after the western part of the Northwest Territories of Canada became the Province of Alberta. The UofA campus is lo cated on the south bank of the North Saskatchewan River which flows through the center of Edmonton. The first An nual Calendar of the university described this site on the riverbank, 200 feet above the river as a beautiful wooded park, which lends itself s plendidly to an architectural scheme suitable for university purposes. Today the campus i s sur rounded by the c it y of 700 000 inhabitants. Despite the tre mendous growth of Edmonton the river valley that run s through the center of the city is largely an undeveloped beautiful park As the city has grown, so ha s UofA. The Copyright C hE Division of ASEE 2000 /0 2 Ch e mi ca l Engin ee rin g Educat i o n


univer sity's enrollment ha s grown from 45 st udent s in 1908 to s li g htl y over 30,000 in 1999. Size of the chemical engineering classes has also grown from the fir s t three graduates in 1928 (when chemical engineering was a spec i a l pro g ram in the Department of Chemistry) to between 65 and 70 Bachelor of Science de grees awarded annually for the last few years. Many other changes have occurred s ince 1928 In 1946 the department became one of the departm e nts in the Faculty of Engineering and in 1948 s hortl y after the di scove ry of a major oil field ju s t so uth of Edmonton, it was renamed the Department of Chemical a nd Petroleum Engineering. For the next 25 years, the department offered undergraduate and graduate programs in chemical and in petroleum engineering; it was also during this period that research began to be an increa si ngly important component of the department's activities. In 1973 the OPEC oil embargo precipitated a n oil crisis and the department again becam e the Department of Chemical Engineering when the petroleum engineering faculty member s joined another department. In the late 1970 s, a co-operative engineering program was introduced in which participants obtain 20 months of relevant industrial experience as part of the five-year undergraduate program The majority of our chemical engineering undergraduate s tudent s are currently enrolled in the co-op program. The com puter process control ( CPC) program was introduced as an undergraduate elective stream in 1986 and the first st udents graduated from this unique program in 1989. (A more detailed hi story of our department can be found in Wanke [ 1 J and Mather Y 1 ) The 1990s witnessed major changes The departm e nt had 18 academic faculty member s when the d eca de started but b y 1996 one-third of them h a d retired a nd been replaced and an additional two new position s had been created through expansion of the program. In addition nine materials faculty joined the depart ment resulting in the first Department of Chemical and Mat e rials Engineering in Canada. The influx of new staff brought with it excitement and new approaches to teaching and research in the department. The changes a nd grow th experienced in the 1990 s were accompanied by significant growth in the number s of undergraduate and graduate st udent s. Each undergraduate engineering program ha s a quota in the so phomor e year; the chemical engineering quota increa se d from 65 to 75 in 1996 to 90 in 1999 a nd will increase to 100 in 2001. About two-third s of the chemical engineering Spring 2000 Tina Barker in the departmental scanning electron microscopy laboratory Undergraduate student and instructor next to fluidized bed column. Application of surf ace-science principles to the process industries and the development of mathematical and analytical tools for improved process performance and materials c haracteri z ation will be the focus of our research in the next decade. 10 3


sophomores completed their first year of engineering at the UofA; the remaining one-third are transfer students from junior colleges and transfers from other programs. A total of 299 undergraduate students are registered in chemical engineering in the I 999-2000 acathe department and undergraduate students can choose from a variety of program and delivery options. Three degrees are offered: a BSc in Chemical Engineering, Chemical Engineering (Co mputer Process Control), and Ma terials Engineering-all of which can be com demic year. The number of graduate students has also increased significantly, from about 75 in 1996 to 110 in 1999; on the average, 40 to 45% of graduate students are PhD students and the remainder is enrolled in a variety of master's programs The department also hosts a large number of po s tdoctoral fellows, re search associates, and visiting faculty; cur rently 29 postdoctoral fellows, 22 research associates and 5 visiting professors reside in the department. Teaching chemical principles with applications to Alberta s pleted in the traditional mode (eight aca demic se mesters) or the co-operative educa tion program (eight academic semesters plu s twenty months of engineering work experi ence interspersed with the academic term s). Approximately one-third of the students are pursuing the regular route to their engineer ing degree and two-thirds are pur s uing the co-operative route. All the programs are ac credited by the Canadian Engineering Ac creditation Board. The current academic staff consists of 24 industries is oneofthe main functions of the department and Undergraduate students usually enter the chemical engineering programs after a com mon first year of study, summarized in Table I. First-year engineering students are placed in specific programs of study according to a quota for each program, based on their indi cated preferences and grades. In March of each year, the first-year engineering students select three programs of study in which they chemical engineering and IO materials engi neering faculty; 16 of these 34 have been hired within the past five years, and there are four academic vacancies to be filled within the next three years. The department also employs 20 permanent support staff: three machinists and two electronics technicians who run the de partmental machine and instrument shops and custom-build and repair equipment for underundergraduate students can choose from a v ariety of program and delivery options graduate and research laboratorie s; two computing and network specialists who keep the computing, data acquisition, and network sys tems functioning; four labo ratory technologists who assist with the undergraduate labo ratories and operate special facilities s uch as the departmen tal sca nning electron micro sco pe ; and the remaining support staff provide the clerical and administrative support nece sary for smooth operation of the department. The staff, graduate students, re searc hers, and some of the classrooms are housed in the Chemical and Materials Engi neering Building. This 8-story, l 84,000-ft 2 building was built in 1968 and currently over 80% of it s space is occupied by the Department of Chemical and Materials Engineering which will be the sole occupant of the building after an Electrical and Computer Engineering Research Facility is completed in 2001. Although the building is over thirty years old, it has been well maintained and the laboratory space is of excellent quality. Much of the large-scale separa tion equipment and high-pressure reactor facilities were con structed in the departmental machine and instrument shops. The se two shops, along with the interfacing expertise of our computer staff, have contributed sig nificantly to the success of our experimental research programs. PROGRAMS Teaching chemical engineering principles with applica tion s to Alberta's indu stries is one of the main functions of /04 are interested, ranking their choices in order of preference. Then, based on each student's GPA and the quota for the various programs the Faculty of Engineering assigns students to specific programs of study. The yearly entrance quota s for the chemical engineering programs currently total 90 s tudent s and will be expanding to 100 st udents within the next two years. Of the approxi mately 90 students per year that have been admitted to chemical engineering programs during the last severa l years, almost all have indicated chemical engineering as their "first" choice. As shown in Table 2, the traditional chemical engineering program is similar to many programs across North America The key exception is the emphasis placed on both chemical engineering laboratorie s a nd process de s ign. The chemical engineering laboratorie s consist of three se parate courses, which serve the dual purpo ses of providing the students with hands-on experience with pilotsc ale chemical engineering processes (e.g., heat exchangers, distillation columns, fluid ized bed systems, etc ) and report writing. The design stream includes courses in engineering economics and finance, as well as two single-semester process design courses. The engineering economics and first design course provide each s tudent with a so lid foundation to complete the seco nd de sig n course, which is project ba sed. In this final de s ign course, each st udent group, involving three or four students, chooses a se parate design project drawn from local industries. The CPC program i s a blend of chemical engineering Chem i ca l E11gi11eeri11g Education


.&. Undergraduate students doing the senio r distillation laboratory. fundamentals, specia lty process-control courses, and com ponents from electrical engineering and computing sc i ence. Wood 1 3 1 provides a hi storica l perspective on the program as well as detail s of the development and de s ign of the pro gram. In 1999 the CPC program was modified to create two streams: computing systems and signal processing The two streams are defined by technical-elective packages In the computing-systems stream, st udents take electrical engineer ing courses in digital logic and microprocessors, a seco nd computing science course, and a further technical elective. In the s ignal-proces s ing stream, students take an additional mathematics course in complex variab l e theory, the electrical engineering signal-processing course, a second computing scie n ce TABLE 1 Common First-Year Engineering Curriculum T Dr. Suzanne Kresta and stude nt s in a problem-solving lab. course, and a further technical elec tive Thus, the CPC program uniquely positions stude nt s to fill a nich e in the process industrie s. Subject Area Si 11gl e Semester Course E qu ivale 11t s Chemistry 2 Phy sics 3 Mathema ti cs 3 Computing Humanitie s I Oth er 0.67 TA B LE2 Chemical Engineering Curricula Beyond the First Year Subject Area ChE Chemistry 2 Mathematics, Stat i st ic s Numer i ca l Methods 5 Thermodynamics 3 Transport Ph enomena 4 Reactor De sign and Analysis I Proce ss De s ign Analysis, and Economics 4 Chemical Engineeri n g Laboratory 3 Pro cess Control Materials Science Electrical Engineering Fundamenta l s I Technical E l ectives 4 Communications (ora l an d written) 1.67 Hum a nities 3 Other 0.67 Chemical Engineering: figures given as si11gle-semester course equivalents C P C I 5 2 4 4 2 4 I 2 4 1.6 7 3 0.67 Computer Process Control: figures g iv en as single-se m este r co ur se equiva l ents. Spring 2000 At the graduate level Master of Science ( MSc ), Master of Engineer ing (ME ng ), and Doctor of Philoso phy (PhD) degrees in three disciplines (Chemical Engineering, Process Con trol, and Materials Engineer in g) are offered. The MSc and PhD degrees are re searc h-based and require that a student take six grad u atel eve l s inglese mester courses for the MSc degree and nine graduate -l eve l si ngle-semester courses for a PhD degree. Of these co ur ses, three mu s t be se lected from a core program, with the remain der being chosen to suit the student's interests Since these degree s are research-based each student must also complete a thesis At the PhD level, st udent s must comp l ete three written preliminary examinations taking three hours each (usually after the first eight month s of the program) and a candidacy examination of the proposed research and the student's capability to pursue the proposed research. The MEng degree is course-based a nd requires completion of ten graduate-level, single-semester courses as well as a project. The MEng is considered to be a terminal degree and not a route to the PhD program Recently at the graduate level a number of joint degree programs have been evo l ving. The two programs c urrently in place are Chemical Engineering and Medical Sciences and Chemical Engineering and Biological Sciences. Coursework includes core chemical engineering courses, courses taken from the partner discipline and courses that match the student's interests. These joint degree programs are new to the UofA, but student interest is driving their development. One of the strengths of the UofA programs is interaction with industry. This interaction occurs formally through some 105


uniqu e program s and r esearc h project s, and informally due to the diver se se t of indu s trie s in the Edmonton ar ea The program that ha s the lar gest si ngle impact on ensuring indu s trial interaction with the undergraduate s tudents i s the Stollery Executive-in-Re s idence program It i s u se d to bring practic ing engineers from a variety of industries into the depart ment to provide guidance to student gro up s working on their de s i g n project s; it u s u a ll y requires a co mmitment of ap proximately two week s of the practicing engineer 's time, spread over an aca demic term. Further the Stollery visitors u s ually bring de s ign project s from th e ir own companies with them. The visiting engineers are also invited to give s pecial lecture s in other co ur ses, where they pro vi de context for th e course s ubject matter. At the graduate level a large and g rowing number of re searc h project s enjoy the partner s hip of local companies. The se often require gra duate s tudent s to s pend extended s tretche s of time in co mpany lab s or at plant s ite s Thi s tr e nd toward industrial participation in the training of graduate students see m s to be of interest to local indu s trie s and is expected to expand. RESEARCH The seco nd main function of th e department is research, which ha s focused on adding to th e c h e mic a l e ngine e rin g knowledge base and addressing problem s of the proc ess industrie s, with particular emphasis o n We s tern Canadian industrie s. The oil and gas indu s trie s in We s tern Canada strongly influenced the areas of re sea rch in the department during the 1950s and 1960s E s tabli s hment of worlds cale petrochemical facilities in Alberta in the 1970s and 1980 s influenced the direction of applied re searc h and the increa ing importance of the sy nthetic crude production in the 1980 s and 1990 s opened up exciting new areas for application of chemical engineering fundament a l s. Con s truction of mod ern pulp mills a nd new s print facilities in the 1990 s added another dimen s ion to the department 's applied re searc h Funding for research co mes from external so urce s: in the 1998-99 fiscal year, funding obtained by the academic s taff excluding central overhead charges exceeded $4.5 million (Cdn); about 65 % of the fund s came from federal and pro vincial gove rnment agencies, a nd the r e maining 35 % came from industry. Five of the main areas of chemical e n g ine e ing re se arch are briefly de sc ribed below Catalytic Reaction Engineering Catalysis and r eac tion e n g neerin g research ha s b ee n d o n e in th e d e partm e nt s in ce th e 1950 s when investigation of se le c ti ve h ydrocarbo n oxidation a nd Claus catalysis were started. Claus catalysis for the conversion of hydro gen s ulfid e to elemental s ulfur i s s till of g reat importan ce s ince mo s t of Alberta 's natural gas contains hydrogen s ulfid e The se s tudi es not only re s ulted in significant improvem e nt s in s ulfur rec ove r y, but also resulted in the d eve l o pm e nt of te c hniqu es (s uch as infrar e d method s) for exa min a ti o n of fundamental processes 106 occ urrin g on the s urf aces of heterogeneous ca t a l ysts. Unders t a nd in g th e b e h av ior of these cata l ysts led t o th e a ppli ca ti o n of h ydro phobic s upp orted metal ca t a l ysts, whic h are finding ap pli cat i o n s in n ew proc esses s u c h as th e production of h ydroge n p erox id e and for th e r e moval of organic compounds from contaminated aqueous strea m s. Th ese ca t a l ysts are a l so well s uit e d fo r ca t a l y ti c di st ill ti o n fo r water-containing sys t e m s. Current ca t a l y ti c reaction e n g n eeri n g proj ec t s in c lud e: cata l ysts for e n v ir o nm e nt a l ap pli catio n s ( both Uquid a nd air), development of ca t a l y ti c distillation pro cesses h eavy-o il upgrading ca t a l ys t s, and o l efi n p o lymeri za ti o n cata l ys t s ( Zi eg l er-Natta a nd s in g l e-s it e catalysts) D e p art ment a l facilities for cata l yt i c st udi es includ e va ri o u s cata l yst c h aracte ri za tion e quipm e nt (c h e mi so rption and phy s i so rption x-ray diffrac tion sc annin g e l ectro n a nd a t omic fo r ce micro sco pi es, a nd infrared s p ec t rosco p y) as we ll as num ero u s reactors, includin g a co ntinuou s hi g h pr ess ure sys t e m for s tudying hydrocrackin g of bitumen and a reactor sys tem for cata l y ti c o l efi n polymerization in th e gas pha se Computer Process Control With the establishment of th e D ata Acquisition a nd Control a nd Simulation ( DA CS) Centre in the department in 1 968, research in comp ut er pro cess control b ecame one of th e m a in r esearc h areas in th e department. Eac h o n e of the m a n y workstations in the ce nt e r tod ay ha s mu c h m o r e computa ti o n a l power than the IBM 1 800 h o u se d in th e or i g inal DACS Centre but the ge n era l aim of tod ay's r esearc h i s th e sa me as that of three decades ago (i.e., development of techniques that a ll ow co mput ers t o be u sed for improving the effic i e n cy a nd reliability of indu s trial pr ocess o p era tion s) Re searc h areas hav e broadened over th e years from the m o r e traditional process co ntrol a nd identifica ti o n t o includ e process m o nit or in g a nd the a ppli catio n of multi var i ate stat i st i cs, co ntroll er-perfo rman ce assess m e nt art ifi c i a l intelli gence, proces sfault diagnosi s, a nd p rocess o ptimi zatio n The size of the pro cess-co ntrol re searc h gro up ha s grow n commensurate with the bro a denin g of th e re searc h sco pe. The computer proc ess control re searc h facilities include a n etwo rk of co mput e r s (U nix workstations and per so nal computers) and ex perim enta l e quipment includin g pilots cale reactors di st ill a ti o n columns a nd ot her s mall ex p er im e nt s. Fluid Mechanics and Transport Phenomena The s tud y of multiph ase flow a nd fl ow in p oro u s m e dia, with em ph as i s o n oi l pip e line s a nd c rud e oi l r eservo i rs, were maj o r topics of research in the d e partm e nt fro m the 1 940s to th e 1 970s. In th e 1 9 70 s a nd s ub se quent decades r esearc h s hift ed to ex p erime nt ation and mod e lin g of complex flows with a pplic a ti o n s to the tran spo rt pro cesses encountered in the proce ss in g of o il sa nd s. Th e work in c lud e d mod e lin g of co mpl ex pro cesses u se d in the ex tra ct ion of bitumen fr o m sa nd as well as ex p e rim e ntati o n n ecessa r y to in crease under s t a ndin g of the c h em i s tr y and phy s i cs that govern the processes in the lib erat ion of the bitumen from the s ur face of the sa nd a nd s ub se qu e nt processing of the bitum e n R esearc h in thi s area co tribut ed s i g n ifica ntl y t o th e impro veme nt of co mm erc i a l proc esses for th e eco nomic produ ctio n of sy nth et i c crude from th e Alberta oil sa nd s. Syncrude Canada Ltd ., th e l argest produ ce r of sy nth etic c rud e (225,000 bbl s per d ay in 1999 ) ha s r ecog ni zed th ese major co ntribution s a nd co-sponsors, with th e Natural Science a nd Engi n ee rin g Re searc h Council of Canada) two indu s trial research chairs in th e departm e nt. R esearc h in thi s area i s movin g from th e con tinuum to th e mol ec ular l eve l. Current proje cts include me as ur m e nt of interfacial propertie s of individual drops in e mul s i o n s, i nt eract i o n s between bitum en droplets u si n g a microcollider appaChemical Engineering Education


ratus and stabilization of bitumen-in-water emulsions by clay s Other active projects in this area include fluid dynamic s aspects of pulp and paper processing and computational (CFD) and experi mental studies in the area of mixing and separation equipment optimization Thermodynamics and Separation Technologies Thermody namics research largely dealing with vapor-liquid equilibrium of system s related to oil and natural gas started in the 1940 s. It culminated in the 1970s with the publication of the Peng-Robinson equation of state The separation research in these decades empha sized the measurement of data and modeling of s ystems used in the removal of hydrogen sulfide and carbon dioxide from natural gas The process models developed for the sour-ga s separation unit s and the vapor-liquid equilibrium data obtained for the amine-sour gas systems are still used today. In th e 1980s, separation research shifted to improving the energy efficiency of packed and trayed column s Fundamental work on factor s affecting the efficiency of separation columns, including studies in interfacial propertie s, re sulted in the development of column packings and column internals with improved efficiencie s These improvements are being used commercially. Current work is using the vastly increased power of computational fluid-dynamics (CFD) packages to model the influ ence of packing geometry on detailed flow patterns with the aim of improving separation efficiencie s. Distillation and packed tow ers, with one-foot diameters are used to validate the CFD predic tions Research projects in thermodynamics today deal with the application of statistical rate theory to interfa c ial and membrane transport experimental and molecular simulations of miscibility of polymer blends and measurement of hydrocarbon solubility in polymers. Advanced Materials This is the most recent general area of research in the department. It started in the 1980s with an indu s trial-sponsored project on gas-phase olefin polymerization ; the use of new catalysts to produce polyolefin s with novel molecular archi tecture continues to be an active research area. Inv es tigation into the thermorheological properties and microphase separation in block copolymers is ongoing. The preparation of magnetic microparticles with different surface functionalitie s is being studied ; such par ticles have wide application in removal of contaminants from in dustrial effluents, carriers for drug delivery and biological cell separation. The merger with materials engineering in 1996 brought many projects in advanced materials into our department including surface modifications for improved wear resistance preparation of electronic materials and sintering of cemented tungsten carbide Other Research Activities Academic staff participates in vari ous formal interdisciplinary projects beside the chemical-materials project s; these include projects with the Department s of Biological Sciences Chemistry Civil and Environmental Engineering Elec trical and Computer Engineering Mathematics and Mechanical Engineering. Graduate students who will receive double major degrees as described previously are involved in several of these interdisciplinary projects There are also joint research projects with other Canadian universities as well as with universities in China Germany, Great Britain, New Zealand Poland, Taiwan and Thailand THE FUTURE Chemical engineering continues to change, and the pro Sprin g 2000 grams at the UofA are no exception. The huge increases in computing and networking systems will affect the delivery of undergraduate and graduate programs and influence theo retical, modeling and experimental research Increasingly powerful and reliable de s ign packages will reduce the amount of instruction dealing with empirical information, e.g ., prop erty and transport correlations The efficiencies provided by improved software and computing systems will be used to include more instruction on interfacial and molecular pro cesses (e.g., molecular phenomena important in colloidal suspensions emulsions, and adsorption) This return to the fundamentals of chemical engineering and applied chemistry is needed if our students are to play a major role in the burgeoning oil-sands industry. The in creased reliance in Canada on synthetic crude oil, up to 50 % of Canada s oil use of 2010 will require chemical engineers with a sound knowledge base in interfacial science (e.g., the molecular processes involved in removal of high molar mass hydrocarbons from sand, the processes involved in the eco nomic removal of solids from "stable colloidal solids in process water suspensions, and the removal of corrosive materials present in submicron suspensions or emulsions). The revised curriculum s hould also reflect the changes in process control tools available today (e.g ., dynamic process simulation software computer-aided mathematics packages, robust-optimization packages etc ) The applied research in the department will continue to focus on the main industrial activity in Alberta; the oil-sands (synthetic crude oil) industry the petrochemical industries, and the pulp-and-paper industry These industries will con tinue to be the core of Alberta's process industry for many decade s, and the problem s to be solved will be a continuing challenge to the university. The increased capabilities of available analytical instrumental and computational tools will allow solution of previously intractable problems; how ever, the basic principles of science and mathematics are still applicable to these problems. Departmental research will concentrate on the application of fundamental science and mathematics to the solution of practical problems and the development of new tools and techniques to solve these problems. Application of surface-science principles to the process industries and the development of mathematical and analytical tools for improved process performance and materials characterization will be the focus of our research in the next decade. REFERENCES 1. Wanke S.E ., Chemical Engineering, in Sons of Martha: University of Alberta Faculty of Engineering 1913-1988, G. Ford, ed. University of Alberta Press ( 1988)\ 2 Mather A E ., History of the Department of Chemical Engi neering ," Colloquium University of Alberta 5 (1996 ) 3 Wood R.K. Computer Process Control Program, J. Eng. Ed ., 51, January ( 1995 ) 0 107


(.._s_,p_e_c_ia_l_F_e_a_tu_,,_e_S_e_c_ u o n ____ _________________________________ ) THE FUTURE OF ENGINEERING EDUCATION Part 3. DEVELOPING CRITICAL SKILLS Dona l d R. Woods McMaster University, Hamilton Ontario, Canada LBS 4L7 R ic hard M. F e lde r North Carolina State University, Raleigh, NC 27695 Armando Rugarcia Iberoamericana University, Puebla, Mexico Jame s E. S t i c e University of Texas, Austin, TX 78712 I n the first paper in this seriesl 1 J we proposed that our goals as engineering educators should include equip ping our students with problem-solving, communica tion teamwork, self-assessment, change management and lifelong learning skills. The se goals are consistent with ABET Engineering Criteria 2000, 1 2 1 currently a consider ation of great importance in the United States and (we pre dict) in other countries in the near future. In the second paperl 3 1 we described a variety of instructional methods that have been shown to improve student learning. In this paper we wi II consider the application of so me of those methods to the development of the desired ski ll s. Process skills are "soft" ski ll s used in the application of knowledge. The degree to which students develop these skills determines how they solve problems write report s, function in teams, self-assess and do performance review s of others, go about learning new knowledge and manage stress when they have to cope with change. Many instructors intu itively believe that process skills are important, but most are unaware of the fundamental research that provides a founda tion for developing the skills, so their efforts to help their students acquire the skills may consequently be less effec tive than they might wish. 1 4 51 Fostering the development of skills in students is chal lenging, to say the least. Process skills-which have to do with attitudes and values as much as knowledge-are par ticularly challenging in that they are hard to explicitly de fine, let alone to develop and assess We might sense that a team is not working well for example, but how do we make that intuitive judgment quantitative ? How might we provide feedback that is helpful to the team members? How can we develop our students' confidence in their teamwork skills? Research done over the past 30 years offers answers to these questions In this paper, we will suggest researchbacked method s that will help s tudent s to develop critical s kill s and the confidence to apply them. As was the case for the instructional method s di scusse d in Part 2, 1 31 a ll of the suggestions given in thi s paper are relevant to engineering education, can be implemented within the context of the ordinary engineering classroom use method s that most en gineering professors fee l comfortable with, are consistent with modem theorie s of learning and have been tried and found effective by more than one educator. Research s ugge s t s that the development of an y ski ll i s be st facilitated by giving st udent s practice and not b y si mply talking about or demon stra ting what to do. l4-oJ The instructor 's Do nal d R. Wo od s is a professor of chemical engineering at McMaster University He is a graduate of Queen s University and the University of Wisconsin He joined the faculty at McMaster University in 1964 after working in industry and has served as Department Chair and as Director of the Engineering and Management program there His teaching and re search interests are in surface phenomena plant design, cost estimation and developing problem-solving skills Ric har d M. F e l d er is Hoechst Celanese Professor (Emeritus) of Chemical Engineering at North Carolina State University He received his BChE from City College of New York and his PhD from Princeton He has presented courses on chemical engineering principles reactor design process opti mization and effective teaching to various American and foreign industries and institutions He is coauthor of the text E le mentary Principles of Chemi cal Processes ( Wiley 2000) James St ic e is Bob R Dorsey Professor of Engineering (Emeritus) at the University of Texas at Austin He received his BS degree from the Univer sity of Arkansas and his MS and PhD degrees from Illinois Institute of Technology, all in chemical engineering. He has taught chemical engineer ing for 44 years at the University of Arkansas Illinois Tech the University of Texas and the University of Wyoming At UT he was the director of the Bureau of Engineering Teaching Center and initiated the campus-wide Center for Teaching Effectiveness which he directed for 16 years Arman do Ru garcia graduated from the Universidad lberoamericana (UIA) in 1970 and went on to earn his MS in chemical engineering from the University of Wisconsin in 1973 and his Doctorate in Education from West Virginia University in 1985 He has been a full-time professor of engineer ing at UIA since 1974 and was chair of the Chemical Engineering Depart ment there from 1975 to 1980 He was also Director of the Center for Teaching Effectiveness at UIA from 1980 until 1986. He has written four books on education one on process engineering and more than 130 articles Copy r ig ht ChE Division of ASEE 20 00 10 8 Chemical Engineering Education


C Future of Engineering Education ) -----------------------------role is primarily that of a coach, encouraging the st udents to achieve the target attitudes and skills and providing con structive feedback on their efforts. A number of approaches to process ski ll development have been formulated and proven to be effective in scie n ce and engineeri n g education includ ing Guided Design 11 1 01 active/cooperative learning ap proaches ,13 161 Thinking-Aloud Pair s Problem Solving (TAPPS), 11 1 20 1 and the McMaster Problem Solving pro gram .14.1s 201 EIGHT BASIC ACTIVITIES TOPROMOTES~LLDEVELOPMENT The following activities promote estab l ishing an effective l earni n g enviro nm e nt for process ski ll development: I. Id entify the skills yo u w ant y our students to dev e lop, include them in the course syllabus and ( if department fac ulty agr ee) the university cata lo g, and comm uni cate their importan ce to the students. If developing problem-solving TABLE 1 Some Evi d e nc e -Ba se d Com pon e nt s of Ex pe rt Problem So l vi n g Prob l em so l ving i s the proce ss u s ed to effective l y and efficiently obtain the best va lu e of a n unkno w n or the be s t deci s ion for a g i ve n s et of constraints when the meth o d of so luti o n is not o b v i o u s b Evide n ce -ba sed targets I D escr ibe yo ur thou g ht s aloud as you so l ve problem s. 2 Occasiona ll y pause and r e flec1 about th e pro cess and what yo u hav e don e. 3 Do not expect your methods for s olving problem s to work e qually we ll for others. 4. Write thin gs down to h e lp overcom e the s tora ge limitations of s h ort-te rm m e mor y (w h ere probl em so lvin g take s pla ce). 5. Focu s o n acc ur acy and not on s p eed. 6. Int e ract with o ther s. 7_ Spend tim e reading the problem s t a t eme nt 8 Spend up to half the ava ilable tim e defining th e problem. 9 When d efi nin g prob l e m s, patientl y build up a clear pi c ture in your mind of the diff ere nt part s of the problem and the s ignificance of each p a rt. r I 0 Use different tactic s when so lvin g exe r c i ses a nd probl ems. 11 U s e an evidence-based sys tem atic s trategy (s u ch as read d efi n e the s l ated probl e m exp l ore t o id e ntif y th e r ea l problem, plan do it and look back ). Be flexible in yo ur app li catio n of the stra t egy. 12 Monitor your thought processes about o n ce per minute while so lving probl e m s_ Progress toward i11t e ma/i z i11 g th ese targets 20 % 40 % 60 % 80% 100 % b y D ona ld R. Woods, 1 998). S o m e of/he il e m s in 1his fabl e are d e ri ve d from m a l e ria/ in R efe r e n ce s 22 24. h Thi s pro cess i s in co nlra s/ 1 0 exercise so l v in g, w h e r e in 1h e sol uli on m e th o d s are quickly apparenl b ec aus e sim ilar probl e m s hav e b ee n so lv e d in /h e pas/. An imponanl tar ge t for l ea m probl e m so l ving. d Su ccess ful probl e m solvers may spend u p l o 1hr ee lim es lon ge r lhan un s u ccessf ul o n es in r e adin g probl e m s 1a1 e m e 111s Mos/ mislak es mad e b y un s u cce ssful probl e m so l ve r s are mad e in lh e d efi nili o n s /a ge. 1 Th e probl e m 1ha1 is sol ve d is no/ 1h e probl e m w rilt e n in 1h e / ex tbook l ns1ead ii i s yo ur m e ni a l in1 e 1pr e 1 a 1i o n of 1h a1 probl e m. Som e 1 ac 1i cs /hat ar e in effec tiv e in solving probl e m s in cl ud e ( 1 ) tr y in g 10 find an equa 1i o n 1h a1 includ e s pr ec i se l y all /h e va riabl es g i ve n in 1h e probl e m s/ale m e nl ins l e ad of fr yi n g 1 0 und e r s 1 a11d 1h ef undam e 111al s n ee d e d 1 0 solve 1h e probl e m ; (2) fr y in g 10 us e soht1ionsfrom pa s / pr o bl e m s eve n w h e n 1/i ey don 1 apply, and (3) /rial and e rror. Sprin g 2000 and teamwork skills are among your objectives for a course, include problem so lving a nd teamwork in the li st of course topics in the sy ll abus and university catalog and allocate time for activities that will provide practice in them. [7 2 11 Be sure the s tudent s understand the relevance of the sk ill s to their professional suc cess, and di sc uss the s kills with the same level of seriousness and enthus ia s m that you u se when presenting the technical co nt ent of the co ur se. 2. Use research not personal in tuition, to identify th e target skills, and s har e the res e arch with the stu d e nts Tabl e 1 s ummarizes evidence ba se d target problem-solving skills. A more complete compilation of novice versus expert evidence is given by Woods ,12 2 1 wi th more re cent evidence also available.r2 3 2 41 Target sk ill s have also been identi fied for comm uni cation,12 5 281 team work, 1 1 '16 2 9 3 61 assessment (including s elf-asses s ment) 129 37 3 8 1 1 ifelong learning 14 3 9 -4 71 and change manage ment. 14.4 8521 3 Make exp li cit the implicit be havior associated with successful appli c ation of the sk ills. Much pro cessing take s place s ubconsciously in the head of a skilled practitio ner. When asked H ow do you do that ?" the reply is ofte n "I don't know; it just happens Our task is to take the s kill and behavior apart, di scove r what is really important ( ba se d on re sea rch ), and commu nicate it to the st udents in easily digestible chunks. Illustrative ob jective s and assessment method s for most skills can be downloaded 10 9


(_s_p_e_c_ia_ _l_F_e_a_tu_re_S_e_c_u_ o_n ____________________________________ _,) from the World Wide Web. 1 7 5 1 1 4 Provide extensive practice in the application of the skills, using carefully structured activities, and provide prompt constructive feedback on the students' effo rts using There is a temptation for instructors to select their own terminolog y for problem-solving strategies in their courses This temptation should be resisted ... being exposed to different problem-solving terminology in different courses is a source of confusion to students. evidence based targets. People acq uire ski ll s most effec tively through practice and feedback. No matter h ow many time s st udent s may see a ski ll demon s trated th ey rarely ma ster it until they attempt it and receive g uidance in how to impro ve their performance after each attempt. 5. Encourage monitoring Monitoring is the metacognitiv e proce ss of keeping track of re g ulating and controlling a mental proce ss, co n sidering past, present and planned men tal ac tion s. As students are working ask them to pau se periodically and write responses to questions that force them to deepen their problemso l ving approach and improve their und ersta ndin g. For example, Why a m I doing thi s? What really i s the problem ? What are the constraints? If I was un s u ccessfu l w h at did I learn ? Arn I finished with this stage? What options do I h ave? Which i s most likely to s ucceed ? Can I write down these ideas? Can I u se c h arts, graphs or equations to represent the id eas? If I had a va lu e of ..... how would that help me in so lving the problem ? Can I check thi s re s ult ? Have I s pent enough tim e d efi nin g t h e problem ? What o th er kinds of problem s ca n I s olve now that I have so l ve d thi s one correctly? Schoenfeld 15 2 1 ha s s hown the importance of monitoring in the development of problem -so lvin g ski ll s. 6. Encourage reflection R eflectio n is the metacogniti ve process of thinking about past ac tion s. For each problem the st ud ents solve, eac h communication they write or team task they acco mpli s h, periodically ask them to write reflections /J O on how the y approached the task. For example, Kimbell, et al., 1231 report that occasionally asking students to sto p the problem-solving proce ss a nd de scr ibe w h at they are doin g improve s the problem so lving a nd the quality of the prod u ct. For example, st udent s may be asked to respond to que stio n s s uch as What did yo u do? and "W hat did yo u learn about problem solving ?" Schon ,15 31 Chamberlain ,15 41 Brookfield ,15 51 a nd Woods and Sheardown 1561 also highlight the importance of reflecting. 7. Grade the pro cess, not just the product For so me as sig nments grade only the problemso lving process the team process or the prewritin g proce ss Grade the reflections, u s ing the target s kills (e.g., tho se listed in Table 1 ) as the criteria Some s pecific exa mple s are avai l ab l e for problem so l ving 1571 and teamwork .l3 51 8. Use a standard ass ess ment and feedback form. Depart mental instructors s hould decide on criteria, and the same assessment and feedback forms s hould be u se d across the c urriculum. DEVELOPING PROBLEM-SOLVING SKILLS In addition to the eight ba sic activities, Use a standard r esearch-based problem-solving strategy across several ( and ideall y, all) courses in an instructional program. There i s a temptation for in s tructor s to se l ec t th e ir own terminology for problem-solving stra tegie s in t h eir courses This temptation s hould b e resisted Only a few of more than 150 publi s hed stra te gies are ba sed on re searc h and being exposed to di ffe rent problemso lving terminolo gy in different courses is a so urc e of confusion to s tud ents. Select an evidence-based stra te gy s uch as the six-stage McMa s ter Problem Solving Strategy: Engage, Define the Stated Problem Explore, Plan, Do It a nd Look Back. 1581 Solve so m e problems in depth If you would normally work through fo ur problem s in a given period of time take the same amount of time to so lve ju st o n e problem a nd hand out illu stra tive so lution s for the other three. Enrich the expe rie n ce for the s tudent s when you work out the problem : for exam ple purpo se l y mak e wrong assumptions so that they eventually realize that this i s not working out." Take time to explore questions lik e "What went wrong?" "What hav e we l earned?" Now what?" Ask the s tudent s to carry out so me of the problem-solving tasks individually or in s mall gro up s. Anonymously di s pl ay on transparencie s s tudent s' attempts to carry out specific s tep s s uch as identifying the sys t em, defining the problem drawing a diagram, and creat ing sy mbols for unknown s. H elp students mak e co nnections between the problem statement the identification of required technical knowl edge, and the problem solution For example, "We h ave just so lved probl em 5.6. Identify the key words in the problem Chemical Engineering Education


( Future of Engineering Education ) -----------------statement that helped you to identify the information needed to solve the problem. Which key words helped you identify the required simplifying ass umpti ons?" Explicitly making such connections helps build problem-solving expertise Y 1 59 1 Writing Skills In ad diti on to the eig ht basic activities, give assignments that require writing Long essays are not required: sing l e paragraphs can be effective at facilitating the development of writing skills and do not impose a heavy grading burden on the instructor. Brent and Felderl 60 1 offer suggestions for brief writing assignments that address a variety of different instructional objectives. In-class writing exercises are par ticularly val u able in that they provide snapshots of what the students act u ally do. Students can often be observed follow ing a typical pattern of unsuccessful writing : they sit with pen poised, stari n g at the paper a nd wait in g for in spiration to strike Encourage them to brainstorm ideas about the topic and abo ut the target audience and to try to find a match between the audience needs and the topic. Suggest that they free-write without critiquing themselves and then discard sections that don't work. Teamwork Skills Many instructors seem to believe that simply giving three or fo u r st ud ents somet hin g to do together-a laboratory experiment, for example, or a process design project-should somehow enable all of them to develop the skills of leader s hip time management communication and conflict resolu tion that characterize high-performance teams. Unfortunately, it is not that easy. What often happens und er suc h circum stances is that one or two st ud ents do most or all of the work and a ll students ge t the same grade. This promotes a great deal of resentment of both the slackers and the instructor. It does not promote development of teamwork skills. If promoting teamwork skills is an objective, use a struc tured approach to teamwork such as cooperative learn ingfl 1 1 3 1 5 l in addition to the basic eight activities. The team assignments shou ld be structured to assure positive interde pendence (that is, if anyone on the team does not fulfill hi s o r h er responsibilities, everyone is penalized in some manner) individual accountability for all the work done on the project face-to-face interaction (at least part of the time) develop ment and appropriate use of interpersonal ski ll s, a nd regular se lfassessment of team functioning. Part 2 of this series 13 1 offered suggestions for meeting the defining criteria of cooperative learning and for overcoming the resistance that some students initially feel toward the approach The fo ll owing procedures help make students aware of several of the requisites of good team functioning: Assign a chairperson/coordinator for every meeting Sprin g 2000 Research has shown that groups function better if a desig nated chairperson coordinates arrangements. The chair's ta sks are to sched ul e meetings, to make sure that all team mem bers know what they are supposed to do prior to each meet ing and to keep everyone on task. Research also s h ows that the chair s role differs from the ro l e of leader [ 33 l _someone who holds greater decision-making a uthorit y than the other team members-although servi n g as chairperson help s de velop l eadership skills. (We do not recommend including Many instructors seem to believe that simply giving three or four students something to do together-a laboratory experiment, for example, or a process design project-should somehow enable all of them to develop the skills of leadership, time management, communication, and conflict resolution that characterize high performance teams. Unfortunately, it is not that easy. the role of leader in team activities ) Require the c h airperson to prepare and circulate an agenda ahead of time and ask the group to give written feedback to the chair at the end of each meeting. The chairperson can use this input to reflect on hi s / her skill and to set targets for improvement. Ha v e the group hold a norms m e eting soon after they are form e d Ask the teams to hold a meeting at which they decide on group behavior norms, reaching consensus on specified questions such as "What is the role of the coordi nator ? " How will we handle missed meetings and late ness ?" How will we make de c isions ?" "How will we deal with team members who repeatedl y fail to meet their respon sibilities ?" "How will we deal with c onflicts that develop in the group ? The teams should summarize their norms on a sheet of paper, sign it and tum a copy in to the in structor. Several weeks later the instructor might return the copy and ask them to reflect on how well they are meeting the norms. A checklist of 17 items that should be addressed in establish ing norms is available .Pl Ask students to complete inventories such as the Myers Briggs Type lndicator, [ 6 1 J FIRO B, 1 3 1 621 Johnson's conflict inventory / 63 1 or the Index of Learning Styles. 1 64 1 Suggest that team members share their results and discuss the implica tions making sure they are aware that the most effective groups include people with different styles. Although differ ences might lead to apparent conflict, they can be used to bring a synergy to group activities that might otherwise be unattainable. Incorporate formal team-building exercises as part of 111


( Special Feature Section ) -_________________________ __, your implementation of cooperative leaming. 11 61 Self-Assessment Skills In addition to the basic eight activities, ave the students write resumes. Although workshop activities can develop self-assessment skills, a concrete ac tivity such as writing a resume is an excellent way to put the skill to practical use include self-assessment as part of what you do to help develop any other skill. Combine writing, reflection and se l f-assessme nt by requiring st ud ents to s ubmit their ana l sis of evidence of ski ll mastery gathered from classwork and other app li catio n s of the skills. Examples of such reports are avai l able on-line_l7 1 Data show that self-assessment tends to correlate with externa l assessments of skill mastery. 1 7 5 1. 651 Lifelong Learning Skills and Problem-Based Learning The learning process may be broken down into the following tasks : 1 6 61 Sense problem or need Identify learning issues Create learning goals and assessment criteria Select resources Carry out the l earning activities Design a process to assess the learning Do the assessment Reflect on the learning process In traditional instruction, the student is responsible only for the fifth of these tasks (learning activities), the last task (reflection) is usually omitted and the instructor takes re sponsibility for the remaining tasks lifelong learners, on the other hand take some responsibility for performing a ll of the tasks themselves One approach is to focus one of the eight tasks on li felong learning. For examp l e cooperative groups could be asked to "identify the learning issues" in a problem. Another more amb iti ous option is to convert "reporting back" to "teach ing." When students have completed an independent study or a research project, they typically report back by giving a speech The class listens with varying degrees of interest. The dynamics change i f the st ud ent teaches the material to a small group The audience li stens intently and asks ques tions, because now each of them is expected to learn the material being presented. The student speaker becomes the teacher. He/she learns and applies the ideas offered in Part 2 of this series 131 and receives the benefits of those that will be presented subsequent l y in Part 4 (how to train the teachers) Perhaps the most ambitious option for promoting the 112 development of skills in most of the tasks is called problem based learning (PBL). 147 66 67 1 Problems and projects can be incorporated into a course in a variety of ways. At one extreme is the traditional approach in which problems are given at the end of each text chapter and homework is assig ned after the professor h as lectured on the s ubj ect. The role of the problems is to help st udent s deepen their und er standing of previously-acquired knowledge. In contrast, when PBL is used the problem is posed before the students have acquired the knowledge needed to so l ve it. This inductive ordering simulates the research environment: the students be gi n with a problem and then proceed to figure o ut what they need to know, create hypoth eses, read the literature and/or searc h the Web, talk to experts with related knowl edge, acquire critica l information through modeling, experi menting a nd discovering and finally so l ve the problem. The approac h may be app li ed in any educational setting includ ing lecture classes, laboratory courses, and design courses. 1 68 1 Once a problem has been posed, different instructional methods may be used to facilitate the subsequent learning process-lecturing, instructor-facilitated discussion ,l5 11 guided decision-making, 1 8 101 or cooperative l earning 11 1 3 1 6 1 As part of the problem-solving process, st ud ent groups can be as sig n ed to comp l ete any of the l earning tasks li sted above, either in or out of class In the latter case, three approaches can be adopted to help the groups stay on track and to monitor their progress: (1) give the groups written feedback after each task (2) assign a tutor or teaching assistant to each group or (3) create fully autonomous self-assessed tutorless" gro up s Guided decision-making 1 8 10 1 is a model for the first op tion. The instructor anticipates how gro up s might handle each learning task and creates written feedback to guide the process. This approach was designed to a ll ow one instructor to manage many groups at a time. It has been u sed s u ccess fully in the teaching of engineering design at the University of West Virginia 168 1 and of pharmacy at Purdue Univer sity .169 1 Option #2-assigning a tutor to each group of four to seven students-has been used extensively in the health sciences. 14 3 .4 61 Option #3 is used when a tutor cannot be provided for each group (a common situation in e n gineering) and/or when the goal is to move students away from dependence on the instructor toward independence and interdependence Each group i s trained and empowered w ith process skills (de scribed previously in this paper); the groups monitor and self-assess their work; and the instructor establishes condi tions to aid the groups in self-management 701 The instruc tor should select a technical topic that would normally be l ectured" on for abo ut three weeks, and use PBL to address it instead The instructor's role is to create the environment, monitor the students' progress and help them reflect on the Ch e mi ca l Engin ee rin g Edu c ation


( Future of Engineering Education ) .__ ________________ _:__....:;___..:....._ ___,_ lifelong learning skills they are developing. Illustrative stu dent reflections and self-assessments are available ,l7l as are more examples of how to move gradually into a full PBL modeJ_f5 1 J This approach has been used succes s fully in engi neering, science and pharmacy education_[ 4 7 1 73 l environment. According to William Perry, a Harvard psy chologist college students progress through s ome or (in rare cases) all of the following stage s of development. Level 2 (Dualism). Every point of view is either right or wrong. All knowledge i s known a nd obtainable from in s tructors and text s, and the s tudent s task is to ab s orb what the instructor presents and Extensive evaluation of small-group, s elf directed self-assessed, interdependent ( cooperative) problem-based learning has been re ported for medical schools [ 44 --4 6 J National Board Medical Examination scores earned by students in such programs were compared with scores earned by students in conven tional programs. The experimental group scored low er on the exams testing basic sci ence, while the opposite result was observed for the exams testing medical problem solv ing The differences were statistically signifi cant. The students who participated in a PBL program exhibited a greater tendency to adopt a deep (as opposed to surface or rote) ap proach to learning, 1 7 477 1 a greater mastery of interpersonal and lifelong learning skills and greater satisfaction with the learning experi ence. Positive program evaluation s of the McMaster Problem Solving program in engi neering[ 4 l and of the Guided Decision-Mak ing Model 1 l0l have also been reported ; how ever, the role of PBL in attai nin g these out comes could not be easily determined be cause the programs studied involved multi faceted skill development efforts. Change-Management Skills People inevitably encounter unexpected and stressful changes in their lives, but successful people are able to cope with the changes in such a way that they emerge with renewed or even greater strength in performance self confidence, and interpersonal relationships even if they initially experience l osses in these In contrast, when PBL is used, the problem is posed before the students have acquired the knowledge needed to solve it. This inductive ordering simulates the research environment: the students begin with a problem and then proceed to figure out what they need to know create hypotheses, read the literature and/or search the Web, talk to experts with related knowledge acquire critical information through modeling, experimenting and discovering, and finally solve the problem. demonstrate having done so by repeating it back. Confusion occurs if the text and the instructor do not agree. Dualist s want facts and formula s and don t like theories or ab s tract models open-ended questions or ac tive or cooperative learning Level 3 (Multiplicity). Most information i s known but there are some fuzzy areas with question s that have no an s wers yet but eventually will. The instructor 's dual role is to convey the known answers and to teach students how to obtain the others Students s tart using supporting evidence to resolve i s sues rather than relying completely on what authorities s ay but they count preconcep tion s and prejudices as acceptable evidence, and once they have reached a solution they have little inclination to examine alterna tives. Open-ended question s and coopera tive learning are s till resented e s pecially if they have too much of an effect on grades. Level 4 (Transition to relativism) Some knowledge i s known but some is not and probabl y never will be. Student s feel that a lmo s t ever y thing i s a matter of opinion and that their a n s wer s are a s good as the instructor s The instructor s task is to present known information and to s erve as a role model that can be discounted Independent thought is valued, even if it is not substanti ated by evidence and good grades should be given to s tudent s who think for themselves, even if they are wrong. domains Stressful changes that students might experience include leaving home for the first time, being expo s ed to unaccustomed intellectual challenges, being thrust into a student-centered learning environment in which the instruc tor can no longer be counted on to s upply all required knowl edge, and making the transition from an academic world to the professional world. Level 5 ( Relativi s m ) Knowledge and values depend on context and individual per s pective rather than being exter nally and objectively based a s Level 2-4 students believe them to be. Using real evidence to reach and support conclu sions becomes habitual and not just something professors want them to do. Different knowledge i s needed and differ ent answer s are correct in different contexts; there is no absolute truth. The student 's ta s k is to identify the context and to choose the best answers for that context, with the instructor serving as a resource. Students at this level are comfortable with corrective feedback. Perr y' s Model of Intellectual Developmenr 40 6 7 7 8 791 (or an equivalent model such as King and Kitchener's Model of Reflective Judgmenil 8 01 ) provides a good framework for help ing students cope with the expectations of the new learning Sprin g 2000 11 3


TA B LE 2: R e tl ec ti o n o n a nd Se l f -R a t ing of S kill D eve l o pm e n t S t ra t eg i es R e tl ec ti o n: P r oblem so l vi11g skill Value the skill: make it an explicit outcome of your course Hand out research evidence for the skill Make implicit behavior exp l icit: li st goals and criteria Use student reflection and monitoring Grade (mark) the problem-solving process Use standard assessment and feedback forms Solve some problems in depth Use a common strategy for problem solving Other Co mmu11 ic a tio11 sk ill Value the skill: make it an explicit outcome of your course Hand out research evidence for the ski II Make implicit behavior explicit: list goals and criteria Use student reflection and monitoring Grade the communication process Use standard assessment a n d feedback forms Require in-class writing Other Tea m s k i ll Value the skill: make it an explicit outcome of your course Hand out research evidence for the skill Make imp l icit behav i or explici t: l ist goa l s and criteria Use stude n t reflec t ion and monitoring Grade the teamwork process Use standard assessment and feedback forms Assign a chairperson for every meeting Start with a norms" meeting Other Se lfassess m e 11t s kill Value the skill : make it an explicit outcome of your course Hand out research evidence for the skill Make implicit behavior explicit: list goals and criteria Use stude n t reflection and monitoring Grade the self-assessment process Use sta n dard assessment and feedback forms Require resume writing Other Life l o11g l earn i11 g skill Value the skill: make it an explicit outcome of your course Hand out research evidence for the skill Make imp l icit behavior explicit: list goals and criteria Use student reflection and monitoring Grade t h e process Use standard assessment and feedback forms Use structured cooperative learning groups Use guided decision-making Use small group, se l f-directed, se l f-assessed PBL Other C ha11 ge ma11ag e m e 11 t s k i ll Value the skill: make i t an exp l icit outcome of your course Hand out research evidence for the skill Make implicit behavior explicit: list goals and criteria Use student reflection and monitoring Use the grieving-process model Use the Perry inventory to guide students List new opportu n ities afforded by t h e change Run a change management workshop Other 114 Already Should Might Not My Do this Work Work Style 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Chemical E11gineering Education


( Special Feature Section ) ........ -------------------------Higher level s (6-9) involve the development of commitment to an internally-based system of values. Most entering college st udents can be found on Levels 2 or 3, and relatively few attain Level 5 by the time they gra duate. Students being asked to function at a level higher than their current level are likely to be under a great deal of stress, especially if the two level s are not adjacent. Their reactions to this s tress account for much of the resistance and occa sional hostility instructors often encounter when they be gi n to u se student-centered teaching method s like cooperative and problem-based learning 166 8 1 If st udent s learn stra te gies for managing the stress associated with the transition to student-centered instruction, they may be better able to deal with the s tre ss ful profe ssio nal and personal s ituation s they will inevitably encounter later in their live s. Strong justification for helping students learn to cope with change is the all-too-common situation wherein a well-in tentioned faculty member hears about problem-ba se d or co operative learning and simply launche s students into it, with little or no explanation or preparation. The outcomes of s uch experiments often include s tudent anger and frustration, pe tition s to the department head and terrible student ratings. One can hardly blame instructors in this s itu atio n for going back to more conventional teachin g, to the ultimate detri ment of their students. When students are helped to prepare for change, it may not eliminate th eir unhappines s about it but they are likely to tolerate it long enough to begin to see the benefits. The first six of the eight basic activities de scri bed pre vi ously apply to the development of change-management ski ll s. In addition, In class or in your office, tell students about the stages of reaction to stressful change. People who find themselves in highly s tressful s ituation s may go through so me or all of the stages that have been associated with the grieving proce ss: s hock denial stro ng emotions, resistance and withdrawal, acceptance, struggle better under sta nding and integra tion. 166 8 1 1 Student s undergoing this process may find it help ful to know how the proce ss works, and more to the point that it eventually ends. You might also take a few minute s to elaborate on how the students can u se the same stage model to help them manage other s tressful situations s uch as death of a friend or relation or the loss of a job Doing so is another way to demonstrate concern about their careers and lives beyond the confines of the classroom which is one of the hallmark s of effective teaching .l3 1 When using s tudent-centered instruction acknowledge to the students that it may be stressful to some of them but make it clear that you are doing it for good reason s. If possible get them to come up with benefits themselve s. For example, Sprin g 2000 I n this course we will be u sing extensive cooperative learning, following the rul es and procedures in the syllabus that we ju s t outlined. Hundr e ds of resear c h studies have shown that this approach leads to some real benefits for students. Form groups of three and make a list of what those benefits might be. Then I'll tell y ou what the resear c h shows and we' fl see how man y of them you get Run a workshop on the management of change. 14 71 SUMMARY Tran smitting knowledge is the easiest part of teaching ; far more challenging is the task of equipping s tudents with the critical ski lls they will need to s ucceed as professionals and responsible members of soc iety The following strate gies have been recommended to help achieve this goal: 1. Identify the skills you wish you r students to develop and communicate their importance to the students 2. Use research, not personal intuition, to identify the target skills. Share some of the research wit h the students. 3. Make explicit th e implicit behavior associated with successful application of the ski lls 4. Provide extensive practice in the application of the skills, using carefully structured activities Provid e prompt cons tru ctive feedback on the students efforts. 5. Encourage monitoring. 6 Encourage reflection 7. Grade the pro cess, not just the product. 8. Use a standard assessment and feedback form. Additional suggestions have also been given that apply specifically to the developm en t of problem-solving writing, teamwork, se lf-a ssessme nt lifelong learning, and change management ski ll s. IF YOU GET ONE IDEA FROM THIS PAPER Focusing lectures, assignments, and tests entirely on tech nical course content a nd expecting students to develop criti cal proce ss ski ll s automatically i s an ineffective strategy In s tructor s who wish to help s tudents develop problem-solv ing, communication, teamwork, self-assessment, and other process ski ll s s hould explicitly identify their target skills and adopt proven instructional strategies that promote those skills. We suggest that you reflect on the strategies listed in Table 2 and rate their potential applicability to your teaching. ACKNOWLEDGMENTS We are grateful to Heather Sheardown (McMaster Univer sity), Antonio Rocha ( lnstituto Technol6gico de Celaya) Robert Hudgins (U niversity of Waterloo), Inder Nirdosh (Lakehead University), John O'Connell (University of Vir// 5


( S n ec ial Feature Section ) .._:_ r _______________________ __, ginia), Tom Regan (University of Mary l and), and Wallace W h iting (University of Nevada-Reno) for helpfu l reviews of t h is paper. In addition, we thank Suzanne Kresta (University of Alberta) and Inder Nirdosh (Lakehead University) for their helpful reviews of Part 2 of the series, and we apologize for inadvertently omitting their names from the acknowledg men s in that paper. REFERENCES 1. Rugarcia A., R.M Feld er, D R. Woods, and J E Stice, The Future of Engineering Education. I. A Vision for a New Century," Chem Eng Ed 34 ( 1 ), 16 (2 000 ) 2 Details about EC 2000 are provided on the ABET Web site: < http: / /> See also R.M. Felder, ABET Cri teria 2000: An Exercise in Engineering Prob l em Solving ," Chem. Eng. Ed, 3 2 (2 ) 126 ( 1998 ) 3. Felder R.M. D R. Woods J E Stice, and A. Rugarcia, "The Future of Engineering Education. II. Teaching Methods that Work, Chem. Eng. Ed. 34 ( 1 ), 26 ( 2000 ) 4. Woods D R. A.N Hrymak R R. Marshall, P E. Wood C M Crowe, T.W Hoffman J D Wright P.A. Taylor, KA Woodhouse and C.G.K Bouchard Developing Problem Solving Skills: The McMaster Problem Solving Program ," J. Eng. Ed 86 ( 2), 75 ( 1997 ) See for additional details 5. Woods D.R., Problem Solving-What Doesn 't Seem to Work ," J College Sci. T each., 23, 1, 57 (1993) 6. Bandura A., "Self-Efficacy Mechanism in Human Agency ," Amer. Psychologist, 3 7 122 ( 1982 ) 7 Woods, D R. Probl emBa sed Learning: Re sou r ces to Gain th e Most from PBL. Woods, Waterdown, ON, distributed b y McMaster University Bookstore, Hamilton ON 1997 Avail able on-line at 8. Wales, C.E., R.A. Stager, and T R. Long Guided Engine er ing D es ign West Publishing Co., St. Paul, MN ( 1974 ) 9. Wales, C.E., A. Nardi and R.A. Stager, Thinking Skills : Making a Choice, West Virginia University Morgantown, WV ( 1987 ) 10. Wales, C.E., Does How We Teach Make a Difference? Eng. Ed., 69 394 ( 1979 ) 11. Johnson D.W., R.T. Johnson and KA. Smith,Active L earn ing : Cooperation in the College Classroom, 2 nd ed., Interac tion Book Co Edina, MN ( 1999 ) 12 Springer L. M.E Stanne, and S. Donovan, Eff ects of Small Group Learning on Und e rgraduat es in Science, Math e mat ics, Engin ee ring and Technology: A Meta-Analysis. Madi son, WI, National Institute for Higher Education ( 1997 ) Available on-line at . 13. Felder, R.M ., "A Longitudinal Study of Engineering Stu dent Performance and Retention. IV : Instr u ctional Methods and Student Responses to Them," J. Eng. Ed. 84 (4), 361 ( 1995 ) Available on-line at 14 Felder, R.M ., G.N. Felder and E J. Dietz "A Longitudinal Study of Engineering Student Performance and Retention. V : Comparisons with Traditionally-Taught Students," J. Eng Ed. 87 (4), 469 ( 1998 ). Available on-line at . 15 Heller P. R. Keith, and S. Anderson, "Teaching Problem 116 Solving through Cooperative Grouping. Part 1 : Group ver sus Individual Problem Solving ," Am J Ph ys ics 6 0 (7), 627 ( 1992 ) 16 McNeill E.W. and L Bellamy, Engin ee ring Core Workbook for Active Learning, Assessment, and T eam Trainin g Ari zona State University, Tempe AZ, Available on-line at 17. Whimbey, A. and J Lochhead Problem Solving and Com pr e hen sio n, Franklin Institute Press, Philadelphia ( 1984 ) 18 Stice, J.E. ed., Dev e loping Critical Thinking and Problem Solving Abilities. New Directions in Learning and Teach ing No 30 Jossey-Bass San Francisco CA ( 1987 ) 19 Ko, E.I., and J .R. Ha yes, "Teac hing Awareness of Problem Solving Skills to Engineering Students." J. Eng Ed., 8 3 ( 4 ) 331 ( 1994 ) 20. Lewis, R.B. "Creative Teaching and Learning in a Statics Class," Eng. Ed., 81 ( 1 ), 15 ( 1991 ) 21. Woods D R., Thr ee Trends in Teaching and Learning Chem. Eng. Ed. 3 2 (4), 296 ( 1998) 22. Woods D.R. Novic e vs. Expert Research Suggests Ideas for Implementation ." J Coll. Sci T e aching 18 66 ; 77 ; 138 ; 193 ( 1988 ) 23. Kimbell R., K Stables T Wheeler A. Wosniak and V Kelly Th e Assessm e nt of P erfo rmanc e in Design and T ec h nology. School Examinations and Assessment Council, Lon don UK ( 1991 ) 24. Leifer, L., "Design Team Performance : Metrics and the Im pact of Technology, in Evaluating Organizational Train ing: Models and Is sues S.M. Brown and C Seidner eds. Kluwer Academic Publishers ( 1997 ) 25. Flower L., Probl e m-Solving Strategies for Writing, 2 nd ed Harcourt Brace and Jovanovich, New York NY ( 1985 ) 26. Reiff, J .D., "In-Course Writing Workshop in a Program of Writing Across the Curriculum ," J. of Basi c Writing In structional Resource Center New York, Spring / Summer ( 1980 ), p. 53, and Workshop on Improving Writing Skills in Humanities ," York University, Downsview ON June ( 1987 ) 27. Hayes J.R. "A New Model of Cognition and Affect in Writ ing," in The Science of Writing, C.M. Levy and S. Ransdell, eds., Lawrence Erlbaum Associates, Hillsdale NJ ( 1996 ) 28. Evidence-Based Target Skills for Communication." McMaster University. Available for viewing and download ing at 29. Locke, E.N., "Goa l Setting and Task Perform ance, 19691980 ," P syc hologi cal Bulletin 90a 125 ( 1981 ) 30 Fisher B.A ., Small-Group D ecision Making, 2 nd ed McGraw-Hill, New York, NY ( 1980 ) 31. Whetten D.A., and KS Cameron, D evelo pin g Management Skills, Scott, Foresman and Co ., Glenview, IL ( 1984 ) 32. Reddy W.B., Intervention Skills: Proce ss Consultation for Small Groups and T eams, Pfeiffer and Co., San Diego, CA ( 1994 ) 33. Hoffman R.L. E. Harburg and N.R.F Maier, Difference s and Disagreement as Factors in Creative Group Problem solving. J Abnormal and Social Psychology, 64, 206 ( 1962 ) 34 Boulding, E. "Further Reflections on Conflict Management, in Pow e r and Conf7,ict in Organizations R.L. Kahn and E. Boulding eds Basic Books New York NY (1964) 35 Wood, D R., S Taylor and S. Jaffer, Assessment of Team work. Part 1: The Group s Performance," Chemical Engi neering Department McMaster University (1999 ) 36. Evidence-Based Target Skills for Team Skills, McMaster University. Available for viewing and downloading at 37. Baud D ., Enhancing Learning through Self-Assessment Ch e mi c al Engine e ring Edu c ation


c Kogan Page, London ( 1995 ) 38. Evidenc e -Based Target Skills for Self-Assessment," McMaster University. Available for viewing and download ing at . 39. Longworth, N ., and W.K. Davi es, L ifelong Learning, Kogan Page,London ( 1993 ) 40. Perry, Jr ., W.G ., Form s of Int ellectual and Ethical D evelop ment in the College Y ea rs, Holt Rinehart and Winston New York, NY ( 1968 ) 41. Knowles M., Self-Directed L earni ng Follett Publishing Co., Chicago, IL ( 1975 ) 42 Tough A.M ., Major Learning Efforts : Recent Research and Future Directions ," Adult Ed. 28 (4), 250 ( 1978 ) 43. Barrows, H.S ., and R. Tamblyn Probl emBas ed L earning: An Approa c h to Medical Education, Springer New York NY (1980) 44. Albanese M A and S. Mitchell "Problem-Based Learning : A Review of the Literature on its Outcomes and Implem en tation Issues ," Academic M edicine 6 8 52 ( 1993 ) 45. Vernon D.T.A., and R.L. Blake Does Problem-Based Learn ing Work? A Meta-Analysis of Evaluative Research ," Aca demic M e dicine, 68, 550 ( 1993 ) 46. Nooman Z.G. H .G. Schmidt and E.S Ezzat Innovation s in Medical Edu cation: An E valuation of its Pr esent Status, Chapters 1 to 7 Springer New York NY ( 1990 ) 4 7. "Evidence-Based Target Skills for Lifetime Learning Skills McMaster University. Avail a ble for viewing and download ing at 48. Westberg G E. Good Grief, Fortress Press, Philadelphia PA ( 1971 ) 49. Bridges W ., Managing Transitions Addison Wesley, Read ing MA ( 1991 ) 50. "Evidence-Based Target Skill s for Change-Management ," McMaster University. Available for viewing and download ing at 51. Woods D R. Probl emBa sed L earning: H elping Your Stu dents Gain the Most from PEL D R. Woods, Waterdown, ON ( 1995 ). Distributed by McMaster University Bookstore Hamilton ON. Additional details are available at 52. Schoenfeld, A.H. Episodes and Executive Decisions in Math ematics Problem Solving," in Acquisition of Mathematical Concepts and Pro cesses, R. L es h and M. Landau eds Aca demic Press New York NY ( 1983 ) 53 Schon D A., Educating the R e fl, ective Practitioner: Toward a New D es i gn for Teaching and Learning in the Prof essions, Jossey-Bas s, San Francisco, CA ( 1987 ) 54. Chamberlain, J.M., Eliminat e Your Self-Defeating B ehav iors, Brigham Young University, Provo UT, (1978 ) 55. Brookfield, S.D., The Skillful Instructor Jossey-Bass San Francisco CA ( 1990 ) 56. Woods D R., and H Sheardown Reflecting and Journal Writing, Chemical Engineering Department McMaster Uni versity ( 1999 ) 57. Woods D R. P.E. Wood, H. Sheardown and T Kourti "As sessing Problem Solving Skills, Chemical Engineering De partment, McMaster University ( 1999 ) 58. Woods D.R. "An Evidence-Based Strategy for Problem Solving, J. En g Ed., accepted for publication 59. Woods D.R. "Problem Solving, Patt ern R ecognitio n and the Resulting Connections in Knowledg e Structure," Chemi cal Engineering Department McMaster University ( 1999 ) 60. Brent, R., and R.M Felder, Writing Assignments PathSpring 2000 F u tu r e o f E ngineering Education ) ways to Connections, Clarity, Creativity," College T eac hing, 40 (3 ), 43 ( 1992 ) 61. Myer s -Briggs Type Indicator ( MBTI) Available to certified administrators from Consulting Psychologists Press, 3803 E. Bayshore Rd Palo Alto CA 94303. An abbreviated vers ion of the MBTI is the K e irse y Temperament Sorter, available at 62 Schutz, W.C. FIRO: A Three-Dimensional Theory of Int e p e rsonal D eve lopm ent, Holt, Rinehart and Winston, New York NY ( 1958 ) 63. Johnson D W ., R eac hin g Out Prentice Hall, Englewood Cliffs, NJ ( 1986 ) 64. Felder R.M. and B.A. Soloman, Ind ex of Learning Styles. Available on-line at < http: / / e rs I users I f I {elder I public I !LS pag e. html> 65. Woods, D.R., R.R. Marshall and A.N. Hrymak "Se lf-As sessment in the Context of the McMaster Problem Solving Program ," Eval. and Assess. in Hi gher Ed. 12 (2), 107 ( 1988 ) 66 Woods, D.R. Probl em -Based Learning : How to Gain the Most from PBL ," D.R. Woods, Wat e rdown, ON ( 1994 ). Dis tributed by McMaster University Bookstore, Hamilton ON 67. Chapman, N .S., Th e Rough Guid e to Probl emBased L ea rn ing in Engineering, Oxford Brookes University ( 1996 ) 68. Bailie R.C., J.A. Shaeiwitz, and W.B. Whiting An Inte grated Design Sequence: Sophomore and Junior Years," Chem. Eng. Ed. 28 (1 ) 52 ( 1994 ) 69. Popovich, N School of Pharmac y, Purdue University. Per so nal communication ( 1997 ) 70. Woods, D .R., W Duncan-H ewitt, F. Hall C. Eyles and A N Hrymak, Tutored versus Tutorless Groups in PBL ," Am. J. Pharmaceutical Ed. 60 231 ( 1996 ) 71. Winslad e, N. Large-Group PBL: A Revision from Tradi tional to Pharmaceutical Care-Based Therapeutics ." Am J. Pharma ceu ti ca l Ed., 58 64 ( 1994 ) 72. Wilkerson, L ., and W H Gijselaers, Brin ging Probl emBased Learning to High er Education: Theory and Pra ctice, New Dir ections for T eac hing and Learning No. 68, Jos sey -Bass San Francisco, CA ( 1996 ) 73. PBL Insight Samford University Birmingham, AL. 74 Entwistle N and P. Ramsden, Understanding Student L earning, Croom Helm London ( 1983 ) 75 Biggs, J.B ., Individual Differences in Study Processes and the Quality of L earning Outcomes, High. Ed. 8, 381 ( 1979 ) 76. Maron, F., and R. Saljo, "On Qualitative Differences in L ear ning : I: Outcome s and Proces s, British J. of Ed Psych. 46 4 ( 1976 ) 77. Felder, R.M. Meet Your Students: 3. Michelle Rob and Art ," Chem. Eng. Ed., 24 ( 3 ), 130 ( 1990 ). Available for view ing and downloading at . 78. Felder, R.M ., "Meet Your Students: 7. Dave Martha and Roberto, Chem Eng. Ed. 31(3 ) 106 ( 1997 ). Available for viewing and downloading at 79 Woods D R. Models for Learning and How They 're Con nected-Relating Bloom Jung and Perry ," J Coll. Sci. T eaching, 22, 250 ( 1993 ) 80. King P.M and KS. Kitchener "Developing Reflective Judg ment: Understanding and Promoting Intellectual Growth and Critical Thinking in Adolescents and Adults ," Jossey Bass, San Francisco, CA ( 1994 ) 81. Felder RM., and R. Br e nt "Navigating The Bumpy Road to Student-Centered Instruction ," College T eac hing 44 (3), 43 ( 1996 ) 0 Ill


( S pecial Feature Section ) ----------------------THE FUTURE OF ENGINEERING EDUCATION Part 4. LEARNING HOW TO TEACH J ames E Stice University of Texas Austin, TX 78712 R ichard M. Fe/,de r North Carolina State University, Raleigh NC 27695 Donald R. Woods McMaster University, Hamilton Ontario Canada LBS 4L7 Armando R ugarcia lberoam ericana University, Pu ebla, Mexico I n the next few years a large number of college teachers will retire to enjoy their golden years far from the so und of class bells, the demands of lectures, exams, grades, research and committee work and all the other joys and vexations of their busy lives as educators. Hired in the ex pansionist 1950 's and 1960 's, they have fought the good fight and have found sa tisfaction in their work, their col league s, and their students. Their departure is watched with intere s t by the young PhDs who would like to have their jobs It will be the bigge s t turnover in faculty s ince the University of Bologna was founded a thousand years ago! And what of the qualifications of these hordes of would-be teacher s? Do they know anything about course design, learn ing p syc hology, classroom dynamics student learning styles, testing, grading, analysis, synthesis, cooperative learning problem-based learning, leading di sc ussions .. ? Do they know anything about teaching ? You can bet they probably know very little, but know not that they know not. Teaching ? "Why, I'll just teach the way I was taught. By the way, can I borrow your lecture note s?" Then there are the in-betweener s, who have been teaching for three to fifteen years. Obtaining funding for their re search is the biggest roadblock on their path to tenure. 111 They spend inordinate amounts of time writing grant pro posal s and keeping their research programs together in the face of ever-shrinking funding pool s. It 's ridicu l ous-the two-inch thick proposal has to be so detailed that I almost have to do the research before I can write it ," one grumbles in frustration. When the y aren't writing proposal s and culti vating funding agencies, they visit companies to forge con nection s that may lead to industrial s upport If they venture into the education literature to try to see what they might do more effectively in the classroom, they soon encounter a language that 's foreign to them, with term s like epistemol ogy, Bloom 's taxonomy, Jun gian typo lo g i es, and Perry lev e ls Deciding that they don t have time to decipher all that gibberish, they give up and just go on lecturing Still other faculty member s may have given up on the c ha se for research fund s to focus on teaching concentrating on writing clear sets of note s and de s igning and preparin g good overhead tran s par e ncie s. They may try so me experi ment s in the classroom s u c h as putting st udent s in team s to work on problem s, but find that their ratings drop, and de cide to "Forget that!" They s ub seq uently focus on pla yi n g it safe -avoiding rocking the educational boat becau se st dent ratings are their bread and butter. Many of the problems faced by these diver se souls-the wannabe faculty member s in PhD or postdoctoral program s, the new or well-established profe sso rs who s u s pect there are more effective ways to do things but don t exac tl y know how and tho se who have little time to spare from their never-ending quest for research dollar s-s tem from a s in g le cause. With rare exceptions no one teache s college teachers to teach! They receive training as researchers, join faculties, and enter their clas sroo m s without so much as five seconds of instruction on what to do there A few of them seem to have an innate ability to motivate students and facilitate l earning and high-level s kill development and so me acquire thi s ability through years of experience. Man y never acquire it however and in the absence of any pedagogical training they teach the way their teachers (w ho also n eve r received any training ) taught them. Thi s i s a questionable way to run a profe ss ion, but it 's been done thi s way for centuries. The first paper in thi s se rie s 121 established the need to Copy ri gh t ChE Division of ASEE 2000 11 8 Chemical Eng in ee ring Education


( change the traditional way of delivering engineering educa tion in order to respond to rapidly changing conditions in technology and society, and the second two papers [ 3 .4l ex plained some of the education jargon and offered ideas for improving teaching effectiveness and personal sa tisfaction with teaching We now come to the question of how engi neering faculty members can best learn their craft and con tinue to keep up with emerging developments in ed u cational methods. In this paper, we will describe a variety of faculty preparation programs and offer suggestions for self-study. GRADUATE COURSES ON TEACHING Every skilled craft provides formal instruction and/or mentorship for its new practitioners ... except college teach ing, which expects its newcomers to learn everything them selves by trial-and-error. While there is something to be said for trial-and-error learning, requiring it for a craft as com plex as teaching is absurd. If the learning occurs at all, it normally takes years, and the ones who pay the penalty for the errors are not the ones who make them. Much of the knowledge and many of the skills college teachers need to be effective can be taught. Good courses on college teaching are offered on a few-perhaps several dozen-campuses but our applause at their existence should be somewhat restrained. Why don 't we see s uch courses at every college that offers the doctoral degree ? What hap pened to good old academic entrepreneurship? A change is long overdue Graduate courses on teaching offer several benefits: Teaching Assistants (T As) s upport their graduate studies by providing formal or informal in s truction to undergraduate s in lecture and lab courses. The students whom the T As are assisting de se rve good teaching, which is what they (a nd their parent s and, for public universitie s the s tate legisla ture) think they are paying tuition to get. If we can impro ve the skills of the TAs by as little as five percent in a teaching course, the cost of the course would be a bargain. There i s no way a te ac hing course could fail to lead to at least that much improvement. A well-designed teaching course gives st udent s considering academic careers a much better picture of the profe ss ion than they could ever get in the normal course of a graduate program. Their career deci s ions are much better informed after they take the course, and if they eventually take teaching positions their profe ss ional learning curves could be shortened b y years. Moreover if the course is taught well some s tudents leaning toward indu stria l careers might be motivated to go into teaching helping to meet the challenge of filling all the faculty vacancies predicted to occur in the coming decade. Students who take teaching courses recei ve training in effective pre se ntation teamwork assessment of learning time management dealing with student-related problems, and other important topics that are not part of normal Spring 2000 Future of Engineering Education ) graduate training o ut side sc hool s of education. The r esu ltin g knowledge and ski lls are u sefu l and marketable whether the graduate joins a faculty or goes into an industrial or government career. There are severa l reasons why such courses are not com monplace their benefit s notwithstanding. First most faculty do not see a need for courses on teaching, believing that the knowledge and skills required to teach effectively can ju st as well be picked up on the job. (If they think about some of their colleagues or their own teachers, they will quickly see the fallacy of this reasoning. We never see our own short comings in our mental telescope, of course ) In addition, Every skilled craft provides formal instruction and/or mentorship for its new practitioners ... except college teaching, which expects its newcomers to learn everything themselves b y trial-and-error. While there is something to be said for trial-and-error learning, requiring it for a craft as complex as teaching is absurd. many di sser tation advisors actively discourage their gradu ate students from taking courses that are not required to pass the qualifying exams and take time away from research. Finally, most engineering faculty do not feel prepared by their own education or experience to teach courses on teach ing Thi s fact in itself is a criticism of our system, which allows u s to practice in a profession whose skills we are not equipped to pass on to others. The time has come to change the way we think about preparation for college teaching. In the first three papers in this series, we propo se d viewing undergraduate education less as amassing of information and more as learning how to think how to create, and how to develop the motivation and ski ll to be a lifelong learner and problem solver. In this paper we argue that a graduate education should be viewed in a similar way Learning how to do research is an impor tant component of a PhD program, but it s hould be exactly that-a component. All engineering PhDs do not go into research as soon as they graduate, and very few of those who do s pend their entire careers there. They may become design engineers, middle and upper-level corporate managers, con sulting engineers, faculty member s, department head s, deans provosts, and chancellors, and a wide variety of other things that do not involve research. Part of our responsibility to our graduate students is to equip them with some of the commu nication and interper so nal skills they will need to succeed in those positions. Providing training in teaching is a good step in this direction. Following are some examples of how it might be done. Since 1972 Jim Stice has offered a course at the Univerll9


C S ,p ecia l Featur e Sec tion ) ..__:_ -----------------------sity of Texas at Au s tin on improving teaching skills. 151 The following topics are covered in a typical one-semester offer ing: 1 Intr oductio n and ove r v i ew ( 1 p e ri od) 2. Th e K o lb L eam in g Styl e In ventory (2 periods) 3. Th e M yers -Bri ggs T y p e Indi ca t o r (2 p e ri ods) 4. In st ru c tional design ( 1 p e ri od) 5. Writin g in str u c ti o obj ec tives (2 p e ri o ds ) 6. P rod u ction of over h ead transpar e n c i es ( 1 period) 7. Mi c roteaching : s h ort v id eotaped pr ese ntations b y clas s m em b e rs (4 to 5 p e riods for a class of about 15 ) 8. T es tin g and g radin g (3 periods) 9. Student c hara cte ri st i cs ( I peri od) JO. T eac hin g b y l ec tur e ( 1 period) I 1. T eac hin g b y discu ss ion ( 1 period) 1 2 Learning th eo r y (2 p e riods) 1 3 Mi c rotea c hin g ll (4 to 5 periods) 14. Th eo ri es of J ean Pia ge t ( I period) 15 Indi vid uali z ed instru c t ion (2 p er i ods) 16 T eac hing probl e m so l v ing: anal y ti ca l t hinkin g (3 periods) 1 7 Tea c hin g problem solv in g: c reati v i ty ( I p e ri od) 1 8. Wh e r e t h e tea c hin g jobs are ( 1 p e riod ) 19 Summary of the co ur se; eva luat ion(] period) Stice s uggests that anyone who ha s the interest and seve ral years of teaching experience can t eac h s uch a course. Those who fee l a pprehen s ive about the fir s t few offerings can team up with so meone from the Colle ge of Education-which is what Stice did-or th e Instructional Development Center. Teachin g the course will provide a real learning experience for the in s tructor s (as teaching courses always doe s). After a few se me s ter s the engineering profe sso r s hould be a bl e to go it on hi s /her own although if the course i s going well there i s a lot to b e sa id for continuing to pr ese nt as a team. The prime recommendation is to keep the cl ass s mall-say below 15primarily bec a u se the microteachin g exercises ( topi cs 7 and 13 ) tak e too long when th e clas s s i ze increa ses Sinc e th e early 1980 s, Phil Wankat and Frank Oreovicz hav e offered a 2to 3-credit course on college teaching in the School of Engineering at Purdue Univers i ty .l 61 Their general outline follows: Part I. Methods and procedures 1 20 1 What wo rk s 2. Effi c i e nc y and effec ti ve ness 3 Taxonom y and ob j ec tives 4. ABET and accreditation 5. Pr ob l em so l ving and c reativi ty 6 Obtainin g an academ i c p os iti o n 7. T eac hin g m e th o ds: l ec tur e, coo p era ti ve groups, di scuss ion t eac hing wit h t ec hn o lo gy, maste, y and P e r so nali ze d System of In struction ( PSI ), laboratori es, d es i gn 8. Graduat e mentorin g 9 T esting and g rading c h eating and dis c iplin e 10 Evaluation of t eaching Pa r t II. The student 1 Pia ge t Jun g, and P e r ry 2. H ow peop l e l earn 3. Moti va tion Part Ill. Redesign of engineering education 1 W eb page proje c t 2. Cas e study: ideal g radu ate program 3. Proj ect: id ea l un dergrad uat e program At McMa s ter Univer s ity North Carolina State Univer sity and other campuses, the campus In s tructional Devel opment Center offers courses to graduate s tudent s thinking about going into teaching and to intere s ted faculty Attendance by engineering graduate s tudent s i s ge nerally low unle ss so me one in the sc hool of engineering vigorously c hampion s the courses and encourages graduate students to attend them WORKSHOPS AND SEMINARS Workshop s and se minars la s ting anywhere from an hour to a week are far more common than academic courses as vehicles for teaching about teaching. The se program s may be external to any campus (e g., profes s ional soc i e ty confer ence workshops), campus-wide engineerings pecific or de partmental. The National Science Foundation s pon s or s the Engineer ing Edu ca tion S c holars Program s ( EESP), 1 11 week-long s um mer workshops at Carnegie Mellon Univer s ity, Stanford University, and the University of Wi sco n s in that examine all facets of academic careers The EESPs are for e n g ineerin g g raduate students a nd relativ e ly new faculty member s, with 30-40 applicants acce pted for each offering. N a tionally known engineering educators give pre se ntation s, and the program at Carnegie Mellon University u ses the excellent book b y Davidson and Ambro se 181 as a required text. T a ble 1 s umma rize s the topical outlines of recent offerings. In the s ummer of 1999 the University of Wi sc on s in presented the S cie n ce and Engin eering Education S c holars Pro g ram to new fac ulty member s and graduate s tudent s in sc ienc e. The National Effective Teaching In stitu t e [NETI] is a three day workshop g iven to faculty member s in engineering and engineering te c hnology under th e auspices of the American Society for Engineering Education ( ASEE ), with so me fund ing from indu st ry. 19 101 Beginning in 1991 the NETI ha s been given every ye ar immediately preceding the annual ASEE Meeting in June. The topic s include learnin g s tyle s and teaching s tyle s, pl a nning a course (including writing in s truc tional objectives) and getting it off to a good s tart effec ti ve lecturing active a nd cooperative learning te s tin g and grad in g, helping s tudent s d eve lop problemso l v in g a nd critical a nd creative thinking s kill s, dealin g with s tudent problem s C h em i ca l Engineering Ed u ca ti o n


( Future of Engineering Education ) -----------------and problem st udent s, a nd managing the stresses associated with academic careers. Deans of e n gineering and engineer ing technology are invited every January to nominate up to two of their faculty members to attend the NETI. Nomina tions are accepted on a first-come-first-served basis and the enrollment is closed at 50. Since 1991 472 faculty members from 157 different institutions have participated the Chemical Engineering Division of the ASEE sponsors a week-long "s ummer sc hool for chemical engineering fac ulty that offers a rich se t of workshops on effective teaching in general and on teaching s pecific topic s One-day and hal f-day workshop s on education-related topics are generally offered befor e, during, or after the an nual ASEE meeting in June 1 111 and the Frontiers in Education (FIE) Conference in October or November. 11 21 Educational works hop s are also offered by the American Institute of Chemical Engineers, the Mexican In s titute of Chemical En gineers, and other professional societies. Every five years The Educational Research and Method s Division of the ASEE has compiled a list of its member s who present work shops on campuses around the country. 11 31 The list includes workshop topics and fees. TABLE 1 Many universities offer workshops and seminars on dif ferent aspects of academic careers or specifically on teach ing Some are open to all faculty members and some are designed specifically for new faculty members and/or grad ate st udents The paragraph s that follow describe severa l programs of this type. At the University of Texas, a unique approach for new faculty is the three-day Engineering Education Scholars Program Workshops* Summer Seminar, initiated in August 1980 by Jim Stice and his colleag u es in the campus-wide Center for Teaching Effectiveness. 114 1 51 All new hires are in vited to attend by their department chairs and the Provost. The presenters-all UT faculty member s and administrators (i cluding the president)-discuss a variety of topics including learning and teach ing styles, instructional objectives, writ ing a syllabus, testing and grading, stu dent characteristics, important university rule s and regulation s, research act i vities and resource s, and what to do on the first day/week of class. Attendance ranges Sunday Monday Tuesday University of Wisconsin-Madiso11 Get ac qu ai nt ed Goals ; retention ; options for t eac hin g; improvin g t eac hin g; collaborative learnin g Academic ca reers; finding mentors ; seeking tenure ; w ritin g grant pr oposa l s; climbing th e acade mic l adder ; departmental tour s Wedne s day Course d es i g n ; assessing student performance Thursday Inno vative t eac hin g options ( overview and par a llel workshops ); panel -yo ung faculty reflect from th e trenches Friday NSF pro g rams ; parti cipant exc hange of material s developed ; activity-participants share materi als Saturday Diver s ity ; workshop eva lu ation Camegie Mello11 University Prior knowledge assessment; introdu c ti ons; c h allenge of cha n ge Goals ; retention ; und e r s tandin g s tud en t n eeds; di versity; proces s es in learning Systematic course design ; problem-based learnin g; balancing t eac hin g, re sea rch and ad mini stra ti on; assess ment of l earn in g; active learnin g; accounting for st ud e nt wor kl oads Cour se design; m e nt o rin g a nd s up e r v i s ing graduate st ud e nt s; v ideot ape participant pr ese ntati o n s w ith feedback Getting research fu ndin g; NSF programs; the future of e n gi neering e du cation; i n st ruction al technolog y Ethics; workshop eval u at ion Stanford U niv ersity Goals ; problem-based learning activity; act i v i ty ; Why am I a profe sso r ? academic roles-teaching research administration Trends in engineering ed uc a tion ; how s tud e nt s learn-

( S p e cial Feature S ec tion ) --------------------------ing like a baby again. I am very much in your debt." It is worth noting t h at the new faculty members who atte n ded the S u mmer Seminar from 1980 throug h 1984 re ceived a n extra week's salary. This st i pend was a powerfu l incentive to participate. Many said somet h i n g like, "I came for the money-and I'm glad I did!" In 1985 the prices of oil and beef went down the drain and the economy of the State of Texas suffered to an extent that the Unical engineering s tudent s at Carnegie Mellon University i s described by Ko 1181 versity was no lo n ger able to provide the extra sa l ary money Attendance at later Sum mer Seminars suffered but by this time the administrators had heard a lot of positive things from those of their faculty who had attended, and so the Seminar's good reputa tion was established. It is still held each sum mer and department chairs and deans still recommend that their new people attend. Several years after the inception of the Sum mer Seminar members of the regular faculty began to ask if they could attend it. This was not felt to be a good idea so instead a second program called the Seminar for Experienced Faculty was initiated at a more sophisticated leveJ. l 1 6 l It was held in January during the week before registration for the spring se mester, and also lasted three days. The first year the number of participants was modest but those attending were very enthusiastic. In the third year, 170 experienced faculty mem bers attended. At North Carolina State University, three day faculty workshops have been offered an nually in the College of Engineering since 1986 by Richard Felder and faculty col leagues. The workshop content is similar to that described previously for the National EfMENTORSHIPS In most skilled professions, novices are mentored by expe rienced practitioners who provide guidance and constructive feedback on the novices initial efforts This process can cut With rare exceptions no one teaches college teachers to teach! A few of them seem to have an innate ability to motivate students and facilitate learning and high level skill development .. [but] many never acquire it, ... and in the absence of any pedagogical training they teach the way their teachers (who also never received any training) taught them. This is a questionable way to run a profession .. years off the learning curve normally required for unmentored novices to reach an accept able level of effectiveness at a skilled profes sion. Doctor s, psychologists lawyers, pre-col lege teacher s, and practitioners of every type of craft are routinely inducted into their pro fessions with the aid of such guidance. As noted previously the on l y skilled profession that does not routinely provide mentoring is college teaching. Felder 1 1 91 describes a mentoring program in the Chemical Engineering Department at North Carolina State University where each new fac ulty member is assigned a research mentor and a teaching mentor. The teaching men tor-who shou l d be a n excellent teacher with a desire to serve in that capacity-and the new professor co-teach a course in the latter 's first semester. The mentor initially takes most of the respon s ibility for planning lectures, clas s activities, ass ignments test s, and conducting classes ; the mentee observes and takes notes ; and the two discus s the class at a weekly debriefing meeting As the semester progresses the mentee gradually takes more responsibility for the instruction and the men tor becomes more of an observer, refraining from intervening in class if the mentee gets fective Teaching Institute. Felder et at.,! 1 7 1 describe the workshop and offer tips for getting engineering faculty to attend such workshops and making them effective. The sug gestions include having both engineering expertise and peda gogical expertise on the presenting staff (sometimes the same individuals can fill both roles but this situation is rare), emphasizing practical applications and putting learni n g theory and research in a supporting role and drawing examp l es primarily from engineering courses. into difficulty and troubleshooting the prob lem at the next debriefing Next semester, the mentee teaches a course and the mentor functions only as an occasional observer in class and consultant at periodic (but not nece sarily weekly) debriefing s. The mentor also makes an effort to introduce the mentee to faculty colleagues with related interests, both locally and at professional conferences. After the first year, the formal mentorship terminates and the mentee joins the normal teachi n g rotation. A similar mentoring approach called peer co unseling was pioneered by Roger Beck of the University of Alberta and has spread to campuses throughout Canada Still another approach to teaching improvement involves partnerships in which two faculty co ll eagues visit each ot h er's classrooms and offer feedback and suggestions .r 20 2 11 At the Universidad Iberoamericana in Puebla, Mexico an eight-hour teaching workshop is presented to all beginning professors the School of Engineering offers workshops on teaching development in a year l y summer program and the D epartment of Teaching Development offers courses and workshops for Mexican and Latin American instit u tions A series of seven seminars on academic careers give n to c h emi122 Some institutions have program s wherein faculty mem bers provide mentoring in teaching to graduate s tudents con Chemical Engine e ring Education


( Future of Engineering Education ) ---------------------templating academic careers. The University of Colorado ha s an advanced TA requirement for all PhD students that is typically fulfilled in the third year of graduate study Y 21 Typically, the student prepares and presents several video taped lectures prepares and grades homework and test que tion s, holds office hours and teache s a recitation section if one i s offered for the course a nd the in str uctor provide s feedback and guidance at weekly meeting s. In the Prepar ing the Profe ssor iate program at North Carolina State Uni versity, a faculty mentor and graduate student mentee ma y work together on a course (as at Colorado) or on a cla ssroo m research study r 23 241 An important requirement for a mentorship program is for department heads and deans to recognize that effective mentoring take s a certain amount of s kill and a grea t deal of time Several hour s of mentor training s hould be pro vi ded by campus instructional development staff or experienced mentors and all mentor s should be compensated in some manner for their efforts. NETWORKING The most common-and arguably the mo s t effective way for new members of a professional organization to learn the ropes and adapt to the local culture is informal network ing with experienced colleagues. Unfortunately man y ne w faculty member s are introverted and wait in their offices for their more experienced colleagues to come to them. It doe s not always happen and it i s lea s t likely to happen to women and minority facu l ty in engineering, who may have the great est need for such s upport. In Th e New Faculty Member ,l1 51 Robert Boice report s on s tudies he ha s conducted of the early careers of man y profe s so r s Boice found that about 13 % of his subjects were qui c k s tarters who reached high level s of research productivity and teaching effectiveness in their first 1-2 years on facul tie s, as opposed to the 4-5 years required b y mo s t new faculty member s. Prominent a mong the factors that differen tiated quick s tarter s from their more numerou s counterparts was that the quick starters spent between two and four hour s per week networking with faculty colleagues-going to lunch or for a cup of coffee with them or visiting them in their offices-and talking about re searc h and teaching. Boice s trongly recommends that new faculty member s force them se lves to engage in s uch activitie s and that department admin istrators and senior faculty member s frequently initiate conver sa tion s with new colleagues in their first year. Other vehicles for teaching-related networking are cam pu s learning com muniti es, l 26 27 1 in which groups of facult y members meet periodically to talk about teaching-related topics or to read and discuss se lected reference s on teaching and meeting s of profes s ional socie ties like the American Society for Engineering Education. Other organizations that Spring 2000 sponsor conferences on teaching and learning include the American Educational Re searc h Association, International Society for Exploring Teaching Alternatives, National Sci ence Teacher 's Association and the Canadian organization called Society for Teaching and Learning in Higher Educa tion Woods and Ormerod r 281 offer additional ideas about networking and its importanc e. CONSULTATIONS WITH CAMPUS TEACHING EXPERTS Analyzing a v ideotape of a lectur e with the help of a teaching consultant i s an effec tive (al beit so metimes hum bling) first ste p toward tea c hing improvement. The Clinic to Improve University Teaching of the University of Massa chusetts developed the following structured approach to class room vi deot api n g that ca n b e implemented either with a consultant or a lone .r2 91 B efore the cla ss, make a list of six question s yo u ha ve a bout yo ur lecturing and write down your guesses at the answers Have the class videotaped and ask the class member s to complete a traditional student evaluation form. Complete the s ame form yourself twice once ba se d on how yo u felt the class went a nd once based on how you gue ss the students rated the experience. Then watch a replay of th e videotape and analyze it in the context of your s ix que st ion s. Compare the st udent evaluations with your two sets of responses a nd identify five s trengths and two areas to work on Thi s proce ss works be s t if yo u go through the proce ss with a consultant but it is s till useful if you do it alone. Much can be learned eve n without the videotaping RESOURCES FOR SELF-STUDY Books McKeachie 's T e aching Tips 1 301 is probably the best known ge n era l reference on college teaching Now in its 10 th edition, it offers s ugge s tions on every aspect of teaching and cites research supporting the s uggestions An excellent reference that applies specifica lly to technical discipline s i s a nd Oreo vicz's Teaching Engineering, 1 3 11 which re cently became available on the World Wide Web ; other book s di sc u ss the attributes of effective college teaching and teacher s irre s pective of di sc ipline Y 2 39 J Some references sur vey the th eory and practice of the in s tructional models dis cussed in Reference s 3 and 4 that have repeatedly been s hown to promote learnin g and skill development. John so n et al. 1401 do this for cooperative learning and Woods l 4 11 does it for problem-ba se d learning For Mexican and Latin Ameri can educators Rugarcia s book La F o rmaci6n de In ge ni eros, 1 421 i s recommended. Several references are written specifically for faculty mem bers new to the profession including books by Davidson and Ambrose ,f 8 I Schoenfeld and Magnan,f 431 Whicker, et al.,l 44I Gmelch ,r 45 l and the previou s ly mentioned work of B oice l 251 on the characteristics of quick starters." The last reference ma y be particularly u sefu l to department head s and se nior 1 23


( Special Feature Section ) -------------------------------' faculty members serv in g as m e ntor s to their junior colleagues. Electronic and Videotape Resources A s ub s tantial and rapidly grow ing collection of resource s for in s tructor s ca n be found on the World Wide W e b. Table 2 li s t s s ite s particularly relevant to engineering education. The sites co ntain cla ss m ateria l s (i n c ludin g multimedi a re so urce s), teac hing and assess ment g uid es, h a ndout s for s tudent s, and link s to sti ll m ore sites. Assessment of l earning is becoming an increasingly im portant topic in engineering education as the day approaches when the outcomes-based Engineering Criteria 2000 be comes th e accreditation s tandard for all U.S engineering department s l 461 Be s ide s th e u s ual midterm and final exa mi nation s, classroom assessment te c hniqu es ( CAT s) ca n b e u se d to monitor what s tudent s are learning and what con fu ses them. Angelo and Cross 14 71 de sc rib e a variety of CAT s that ca n be used to assess l ea rnin g and s tudent attitude s, and Boud 1481 offers ideas for h e lp A g rowing number of li s t se rver s provid e rich opportuni ties for interaction with co lle ag ue s see king to improv e their teaching. Table 3 li s t s severa l of th e m ing s tudent s to se lf-a ssess their own learning. Journals and Newsletters ASEE Pri s m is the n ews journal and Th e J ournal of Engineering Education ( JEE ) the re searc h jour nal of the American Society for En g ineerin g Education Pri sm con tain s W as hington update s, fea tur e articles on current issues and r cent de ve lopments in engineering education, and a colum n on teach ing methods written by Wank a t and Oreovicz the author s of T eac hin g Engineering P 11 JEE contains ar ticle s on instructional meth o d s and program s as well as review s of r e cent book s of intere s t to engineer in g pro fessors Both journal s co m e with member s hip in the ASEE. Oth e r journals containing u sefu l article s for college teacher s includ e Th e Journal of Colle ge Science T eac hin g, College T eaching, Change, J ou rnal on Excellence in College Teaching, the AAHE Bul l et in (published by the American Association for Higher Educ a tion ), and Studies in Hi g h er Education. Several e ducation journals s uch as Chemical Engineering Education and, in Spanish Edu caci6n Qu{mi ca and Revista del IM/Q fo cu s on i ss ue s related to s pecific branche s of engineering New s letters that offer teachin g tip s and s ummaries of recent book s includ e Th e T eac hin g Pr ofessor, 1491 Th e National T eaching and L earn ing F orum,15 1 and Cooperative L earning and College T eac hin g 1511 1 24 TABLE2 Useful Web Sites for Engineering Educators Wor ld Lecture H a ll NEEDS-Na tional Eng ineerin g Education D elivery System < www .n> Resources in Engineering and Science Education ( Richard Felder's Web s it e) Deliberations on Teaching and Learning in Higher Ed ucation (Lo nd on Gui l dhall U ni versity) Co ll aborative Learning Website (Natio nal Institute of Sc i ence Educat i on) Fie ld-T es ted Learning Assessment G uide (Na tional Institute of Scie n ce Educa ti on) Co mputer-Ba sed Teaching and Learning Links (U niversity of Newc as tle ) < http://lorien.n c l.a c. u k/1ning/ R eso ur ces/ca //CAL.htm> Taking Your Course On-Line (No rth Caro lin a State University) Problem-Based Learning and the McMa s ter Problem Solving Program ( McMaster University) For Yo ur Consideration (U niver s ity of North Carolina) Mount A lli so n University Teaching and Learning Page Links to a Better E duc atio n U ni vers it y of Guelph University of Technology Sydney (Australia) Comments Lecture note s and multimed i a resour ses for co u rses in many fie ld s, including engi n eeri n g. Multimedia resources for a vast co ll ection of topi cs Articles, Random Thoughts columns from Chemical Engineering Education st udent handout s, sof tware tut orials. M a t eria l on co ll aborative l ea rnin g, assessment of learnin g a nd teachin g, and engineer in g ed ucati o n. Practi ca l s ugge stio n s anecdotes, research cita ti ons and a n ex t ensive annota t ed biblio grap h y on cooperat i ve learning Techniques resources and references on assessment of learning in science, mathem a ti cs, engineering and technology Large co llection of links to s ite s that deal with a pplied and theor et i ca l as pects of in struc tion a l technolo gy. S u ggest i o n s and r eso ur ces for co u rse delivery via the World Wide Web a nd ot h er e l ectronic media Tec hn iq u es and in s tructional resources for bot h programs. Short monograp h s on topics such as active learning writing to l earn, teaching large lecture classes and assessme nt of teachin g and learning. Listin gs of ed ucati on -related conferences and link s to ot h er si t es sor t ed by t op i c (co ll aborative l ear nin g, l earning s t y l es, technology e t c.). H andouts fo r st ud e nt s o n learning a n d problem so l ving sk ill s taking tests critica l thinking technical writing, time management teamwork, learning s tyl es, creativity and many other topics S u ggestions, onlin e assessment t ools and link s to sites that dea l with l earn in g sty l es teaching portfolio s, copyright laws, and course de sign Survival guide for new teacher s evaluating teaching and courses teaching portfolios C h e mi ca l E n g in eeri n g Educatio n


( Future of E ngin e ering Education ) .__ _________________ ......:,__~----The National Technological University regularly offers seminars on education-related topics over satellite links to campuses around the country, and also makes available vid eotapes of past programs _l5 2 l Some topics that have been presented include cooperative learning (Karl Smith and Ri chard Felder) programs for minorities ( Ray Landis) learning styles (Felder), and women in engineering (Eleanor Baum). A MODEL ENGINEERING FACUL T Y DEVELOPMENT PROGRAM Beginning in 2001, engineering departments seeking ac creditation will have to s how that they are equipping their graduate s with a specified array of s kills and that they have established a program to asse s s the levels of these skills a n d remedy any deficiencies revealed by the assessment. 146 1 Quali tative changes in the content and delivery of engineering cour s e s along the lines outlined in the first and second pa per s of thi s s eries will be required to attain the desired learning outcomes. To implement these changes, most engi neering professors will have to be educated in the new in s tructional methods as oppo s ed to the relative few who have been motivated to learn about them in the past. The University of Technology at Sydney make s available excellent videotapes on four topic s Lectures ," "Tutori als," "Practicals," and Assessment -and a support text called Survival Guide for New T e a c h e r s. 15 3 1 The University of Victoria offers a series of Critical Incident Videotapes ," brief scenarios of typical class problem s that provide focal points for discussion Y 41 For example the ten critical inci dents on Tape 1 include one that deals with a student at one level of intellectual development trying to write an essay that calls for thinking at a higher level and another that involves students complaining to th e in s tructor that the clas s lacks structure. Woods has produced a videotape on s elf directed learning that can be obtained by request. f 55 l Margarita Sanchez of the lnstituto Tecnol6gico y E s tudio s Superiore s de Monterrey (ITESM) offers s at e llite-linked course s on problem solving A model engineering faculty development program is be in g developed by the Southeastern University and College Coalition for Engineering Education (SUCCEED) and imple mented at the eight Coalition campu s e sY 61 The core of the program i s a broad variety of learning opportunities and re s ource s for faculty members and graduate students. Op portunitie s may include cour s es on teaching workshops and s eminar s, mentorships and partnership s, learning communi ties, and individual consulting with instructional developO rga n iza ti o n o r Th e m e A lt e rn a t ive a nd co ll a b ora ti ve l earn in g Alt e rn a ti ve l ea rnin g a ppro ac h es A ssoc iati o n for Hi g h e r Edu ca ti o n Adult e du c ati o n network A ss n for th e Stud y of Hi g h e r Edu ca ti o n Probl e m so lvin g a nd c re a ti v it y l e arnin g Coo p e rati ve L ear nin g L e arnin g St y l es Hi g h e r Edu ca ti o n Proce sses Problem ba se d L ea rnin g Cent e r for F ac ulty D e velopm e nt E x pl o rin g th e way we e du ca t e P rof a nd Or ga n iza t io n al D eve l o pm e nt S oc i e t y fo r T eac hin g a nd L e amin g in Hi g h e r Edu ca ti o n (Ca nad a) Co ntinu o u s Qualit y Impro ve m e nt Sprin g 2000 TABLE3 Educat i on-Re l ated Listservers / 11for111a t io11 Email Addr e ss aa h esg it @ li s t. c r e n n e t < LISTPRO C @ li s t. c r e n.n e t > aedn e t @ pul sa r. aca s t.n ov a .e du Th e a pp roac h of T o n y Bu za n t o
  • l ea rnin g m e m o r y a nd c r ea ti v it y C L @ jar in g m y < LISTS E R VE R @ jaring. m y> Sub sc ri be CL fi r s tn a m e l as tn a m e < h e p rocL @ I i s t s e r v .a m e ri ca n e du > < im sac pbl L @ im s a.e du > PBL-LIST M o n as h U ni ve r s it y < LISTSERV @ e n g. m o n as h. e du. a u > Au s t ra li a S U B PBL LIST yf n a m e y ln a m e U ni ve r s it y o f A ri zo n a < n ewed u L @ uh cc v m. uh cc h awai i .e du > < p o d @ li s t s .ac s ohios tate edu> s tl h e -L @ unb c a
  • s ub STLHE L y fn a m e y ln a m e CQI L @ mr.n e t < li s t e r v @ m r.n e t > ment personnel. Some of these program s are open to all faculty members and others are de s igned s pecifically for faculty members in their first two years of teaching Programs for gradu ate s tudent s include course s on teaching, workshops and semi nars and mentorships. Some of the graduate student programs are designed for teaching assis tants and others for students con templating academic careers. Resource s for self-study are also provided as part of the program, including books, journals, vid eotape s and guides to useful Web sites Program facilitators should collectively have expertise in both pedagogy and engineering. An essential compo n ent of a successful faculty development program is strong institutional s upport. An adequate budget is of cour s e a necessary condition Beyond that academic admin i s trator s s hould convey a clear expectation that the faculty will be good teachers, good teaching 125

    PAGE 32

    ( S p ecia l Feature Section will be rewarded in tangible ways, and inadequate tea ch ing will be penalized. We will return to this point in th e last paper in this series. SUMMARY Few engineering schools explicitly prepare their grad ate students to teach and new professors consequently join faculties equipped with a PhD in their discipline but no background in pedagogy. Also, most colleges and universities have few criteria to screen prospective can didates for their teaching ability; much of the emphasis in hiring is on perceived potential as a researcher. Candi dates often give seminars on their research, and if they can give a passable performance and can answer a few questions without complete intellectual collapse, then their teaching skills are judged "good enough." As time passes some of those hired become good teachers by instinct and others learn their craft by years of trial-and error effort, but some never rise above mediocrity or worse. Teaching is a complex craft, but the skills required to do it effectively can be taught. In this paper we have outlined the elements of an effective engineering faculty development program To recapitulate, we advocate a program that includes a subset of teaching improvement workshops, courses, seminars mentorship s and partner s hip s, learning communities and consultation with cam pus teaching experts. Graduate courses in college teach ing should be provided for tho se students who think they might be interested in academic careers. The faculty development coordinator s hould maintain resources for self-study, including books journals, multimedia re sources and guides to useful Web sites. Such a program should enable a far greater percentage of new faculty hires to become highly effective in 1-2 years-i.e., to become what Robert Boice has termed "q uick start ers"-instead of the 4-5 years required by most of the new faculty members Boice studied. Table 4 invites reflections on the options for teaching improvement presented in thi s paper. IF YOU GET ONE IDEA FROM THIS PAPER We have described many options for new in s tructors to learn the craft of teaching including courses, work s h ops and seminars on teaching professional society con ferences, mentorships and working with teaching consult ants. Faculty members should take advantage of as many of these opportunities as pos s ible rather than relying on trial-and-error for mastering the craft of teaching ACKNOWLEDGEMENTS W e thank Chris Knapper Queen 's University, 126 ) TABLE4 Reflection and Self-Rating Rate t h e ideas Already Wou ld Mig ht Not my do work work style Draw on others Take a co ur se on effec ti v e teaching 0 0 0 0 Attend workshops o n teaching 0 0 0 0 Ask for a mentor 0 0 0 0 Partner w ith a co ll eague t o improve teaching 0 0 0 0 Work w ith a teaching co n s ultant 0 0 0 0 Be videotaped in clas s 0 0 0 0 Other 0 0 0 0 Self study Read book s abo ut effective tea c hin g 0 0 0 0 Read articles in education journals 0 0 0 0 Watch videotapes about effec ti ve teaching 0 0 0 0 Browse education-related Web s ites 0 0 0 0 Other 0 0 0 0 Keep up to date Join the American Society for Engineering Education 0 0 0 0 Read a t l east one ed ucation journal each month 0 0 0 0 Subscribe to an educatio n-r e lat ed li s tserver Attend an educa ti on co nfe re nc e 0 0 0 0 Other 0 0 0 0 Pass on yo ur knowledge Give a wo rk s h op or seminar on effective teaching 0 0 0 0 Serve as a mentor to a n ew instructor or graduate s tud e nt 0 0 0 0 G i ve a confe r e n ce presentation and/or write a paper about a teaching method yo u h ave tried 0 0 0 0 Teach a course on effective teaching 0 0 0 0 Kingston ; Alan Blizzard and Dale Roy McMaster Univer s ity ; Susan Ambrose, Carnegie Mellon University; Rich Noble, Univer sity of Colorado ; and Phil Wankat Purdue University, for their comments and s u ggestions. REFERENCES 1. Bert R., What Do Assistant Professor s Want ?" ASEE Pri sm, pp. 24 27, May-June ( 1999 ) 2 Rugarcia, A., R.M Felder, D.R. Woods, and J E. Stice, The Future of Engin ee ring Education. I. A Vision for a New Century," Chem En g Ed ., 34 ( 1 ), 16 (200 0 ) 3. Felder R.M. D R. Woods, J.E. Stice, and A. Rugarci a, The Future of Engineering Education II. Teaching Methods that Work," Chem. Eng Ed ., 34 ( 1 ), 26 (2000) 4. Wood s, D R. R.M. Felder A Rugarcia and J.E. Stice The Future of Engineering Education. III. Developing Critical Skills," Chem Eng. Ed., 34 (2), 108 (20 00 ) 5. Hereford S.M., and J E Stice, "A Course in College Teaching in Engineering and Science," Annual Conference of ASEE Iowa State University, Ame s June 25-28 ( 197 3) 6. Wankat P.C., and F .S. Oreovicz Teaching Prospective Faculty Members About T eac hing : A Graduate Engineering Co urse ," Eng. Ed. p 85 Nov. ( 1984 ) 7. Information about the Engineering Education Scholars Workshops is available on -li ne at 8. Davidson C.I., and S.A. Ambrose Th e New Prof essor s Handbook: A Chemical Engineering Edu ca tion

    PAGE 33

    ( Futu r e of En g i ne e ring Education ) ...__ ________________ __,_ Guid e to T eac hing and R ese ar c h in Engine eri n g and S c i ence, Anker Publishing Co. Boston MA ( 1994 ) 9 "NET! Always Draws A Crowd," ASEE Prism p. 36, D ec ( 1997 ) 10. More information about the National Effective Teaching In stitute can be obtained at 11. Information about ASEE meetings and conferences can b e obtained at 12. Information about FIE conferences can be obtained at < http: I I fairway. ec n.purdu e.e du I -fie / > 13. Culver, R.S. The ERM Workshop Catalog: Faculty Dev e lop ment Workshops for Engineering / Technology Facult y," Binghamton University Box 6000 Binghamton NY 139026000 . 14 Stice J E. "A Model for Teaching New Teachers How to Teach," Eng. Ed. 75 ( 2 ), 83 ( 1984 ) 15. Lewis KG., M D Svinicki and J.E Stice Filling the Gap: Introducing New Facult y to the Ba sics of Teaching ," J. Staff, Prag & Org. D ev., 31 16 ( 1985 ) 16. Lewis KG M.D. Svinicki and J.E. Stice A Conference on Teaching for Experienced Facult y ," J. Staff, Pr ag & Or g. D ev 35 137 ( 1989 ) 17. F e lder R.M., R. Leonard and R.L Porter Oh God Not Another Teaching Workshop ," Eng Ed., 79 ( 6 ), 622 ( 1989 ) 18. Ko E.I. A Seminar Series on Ac a demic Careers for Chemi cal Engineering Students ," Chem. Eng. Ed. 2 9 ( 4 ) 230 ( 1995 ) 19. F e lder, R.M ., Teaching Teacher s to Teach: The Case for Mentoring Chem Eng. Ed ., 27 ( 3 ), 176 ( 1993 ). Available on line at 20. K a t z, J. and M Henry Turning Pr ofesso r s into T eac h ers: A N ew Approach to Facult y D eve lopm ent and Student L e arn ing American Council on Education New York ( 1988 ) 21. American Association for Higher Education Peer Revi ew of Teaching Project ," 22. Noble, R.D., personal communication ( 1999 ) 23. Beaudoin S.P. and R.M Felder Preparing the Profe ssor ate : A Study in Mentorship J. Grad T c hng A sst D ev., 4 ( 3 ) 87 ( 1997 ) 24. Kaufman D.B. R.M. Felder and H. Fuller Accounting for Individual Effort in Cooperative L ea rning T ea m s," J. En gr. Ed ., in pres s 25. Boice R. Th e New Fa c ult y M e mb e r. Jos sey -Ba ss, San Fran cisco CA ( 1992 ) 26 Brent, R. R.M Felder D Hirt D Switz er, and S. Holzer "A Model Program for Promoting Effective Teaching in Colleges of Engineering ," 1999 ASEE Annual Conferenc e Proce e din gs, ASEE, June ( 1999 ) 27. Boyer, E.L. S c holarship R ec onsidered: Prioriti es o f th e Pr f esso r ia te Carnegie Foundation for the Advanc e ment of Teach ing Princeton NJ ( 1990 ) 28. Woods D R. and S.D. Ormerod N e tworking : H ow to Enri ch Your Life and G et Thing s Don e, P fe iff er and Co. San Di ego, CA ( 1993 ) 29 University of Massachusetts The Clinic s Teaching Improve ment Process : Some Working Materials and "C linic to Im prove Teaching : Second Annual Report Th e Clinic to Im prove University Teaching School of Education Univ ers it y of Massachusetts Amherst MA 01002 ( 1974 ) 30. McKeachie, W J ., T e aching Tips : Strat eg ies R esea r c h and Th eo ry for College and Univ e rsity T eac h e r s, 10 t h ed. Houghton Millin Boston MA ( 1999 ) 31. Wankat P ., a nd F S Oreovic z, T e aching En ginee ri ng, McGraw Hill New York NY ( 199 3 ) Available on-line at Spring 2000 . 32. Brown G., and M Atkins, Eff ective T eac hing in Higher Edu ca tion, Methuen London UK ( 1988 ) 33. Ramsd e n P. L ea rning to T eac h in Higher Education Routledg e, London UK ( 1992 ) 34. Newbi e, D and R. Cannon A Handbook for T eac h e rs in Universiti es and Colleges, 3rd e d. Kogan Page, London UK ( 1995 ) 35. Reis, R. Tomorr ow s Prof essor, IEEE Press Piscataway NJ ( 1997 ) 36. Eble K Th e Craft of T eac hin g, 2nd ed. Josse y -Bass San Francisco CA ( 1994 ) 37. Elbow P ., Embracin g Contraries, Oxford University Press New York NY ( 1987 ) 38. Lowman J Ma steri ng the T ec hniqu es of T e aching, 2 n d ed. Jossey-Ba ss, San Francisco CA ( 1995 ) 39. Palmer P J. Th e Courage to T e a c h : E x ploring th e Inner Landscap e of a T eac h e r 's Lif e Jossey Bass San Francisco, CA ( 199 8 ) 40. John so n D W ., R.T Johnson and KA Smith, Activ e L ea rn ing: Coop e ration in the Colleg e Classroom, 2 nd ed. Interaction Book Co. Edina MN ( 1998 ) 41. Wood s, D R. Probl e m-Bas e d L earning: How To Gain th e Most from PBL Woods Waterdown ( 1994 ). Distributed by McMaster Univer s ity Bookstor e, Hamilton, ON, along with the companion book s, Probl em -Ba sed L earni ng: H e lping Your Students Gain th e Mo st from PBL and Probl e m-Bas e d L ea rn ing: R esou r ce s to Gain the Mo st from PBL Chapters from the latter two book s a re available on-line at < http: I I che m eng.mcmaste r .ca / inno v l htm > 42. Rugarcia A ., La F ormaci6n d e In genie ro s. Universidad Iberoam e ricano Puebla Mexico ( 1997 ) 43. Schoenfeld A.C. a nd R. Magnan M e ntor in a Manual: Climb ing the Acad e mi c L adder to T e nure, 2 nd ed., Magna Publica tions, M a di so n WI ( 1994 ) 44. Whick er, M.L. J J Kronenfeld and R.A. Strickland Getting T e nur e Sage Public at ions N e wbury Park, CA ( 1996 ) 45. Gmelch W.H Coping with Fa culty Str ess, Sage Publica tions, London UK (1993 ) 46. Detail s a bout EC 2000 are provided on the ABET Web s ite: . See a lso R.M F e lder "ABET Crite ria 2000 : An Exerci se in Engin ee ring Problem Solving ." Chem. En g. Ed ., 32 ( 2 ), 126 ( 1998 ) 47. Angelo T A ., and KP. Cross Classroom Assessm e nt T ec h niques: A Handb ook for Coll ege T eachi n g, 2nd ed. Jossey Bas s, San Franci sco, CA ( 199 3) 48 Boud D ., Enhan cing L earning Throu g h Self-a ssess m e nt Kogan-Page London UK ( 1995 ) 49. W e im e r M E. ed. Th e T eac hing Prof es sor, Magna Publica tion s. For information about subscribing, see < http: I I www magnapub m>. 50. Rh em, J. e d. National T eaching and Learning Forum Oryx Pres s. F or information about subscribing see . 51. Millis B. ed., Coop e rativ e L e arning and College T e aching, N ew Forum s Pres s, P O Box 8 76 Stillwater OK 74076 52. N a tion a l T ec hnological University NTU National Engineer ing Facult y Forum Series < http : / /> 53. T e aching Matt e r s ( videotape s) and Survival Guid e for N ew T e ach ers at UTS The University of T ec hnology, Sydney Austr a li a, < http : / / au> 54. Critical In c id e nt Videotapes. Learning and Teaching C e ntre The University of Victoria Victoria BC Canada 55. Woods D R. Th e MPS SDL Pr og ram ( videotape ) Chemical Engin eering D e p art m e nt McMaster University, Hamilton ON Canada ( 199 3 ) 0 1 27

    PAGE 34

    Award Lecture _____________ Chemical Engineering Division, ASEE 1 99 9 U nion C a r bide Aw ard Lec tu re PARTICLE DYNAMICS IN FLUIDIZATION AND FLUID-PARTICLE SYSTEMS Part 2. Teaching Examples LIANG-SHIH FAN The Ohio State University Columbus, OH 43210 In Part 1 of this lecture, I discussed the general educational issues concerning particle technology with specific emphasis onfluidization andfluid particle systems. In this part I will discuss some pertinent materials pertaining to fluidization andfluid particle systems that could readily be integrated into existing required c hemical engineering course materials. These materials, eac h introduced to students for a specific purpose, cover both gas-solid and gas liquid-solid fluidization systems. In addition, I will discuss relevant com mer cia l codes that are available for students to learn about the computation of fluid particle systems. Some representative results marking the state-of-the-art efforts in com putational fluid dynamics of fluidization will also be given. L.-S. Fan i s Di s tinguished University Professor and Chairman of the Department of Chemical Engineering at The Ohio State University. Hi s expertise i s in fluidi zatio n and multipha se flow, powder technolo gy and particulates reaction engineering. Professor Fan is the U.S. editor of Powder Technolog y and a consulting editor of the AJChE Journal and the Int ernatio na l J ournal of Multiphase Flow. He has authored or co authored three books, includin g Principles of Gas-Solid Flows (w ith Chao Zhu; Cambridge University Press, 1998 ) Prof esso r Fan i s the principal inventor (w ith R. Agnihotri) of a patented process OSCAR ," for flue gas cleaning in coal combustion and is the Project Director for the OSCAR commercial demonstration funded at $8.5 million as Ohio Clean Coal Technology, currently taking place at Ohio McCracken power plant on The Ohio State University campus. H e h as se rved as thesis advisor for two BS, twenty-nine MS, and forty-two PhD s tudent s at Ohio State and is a Fellow of the American Association for the Advancement of Science and the AIChE. '' Part 1 of this Award Lecture appeared in the Winter 00 issue of Chemical Engineering Education (C EE 34(1), p. 40, 2000). 1 28 SAMPLE SUBJECTS OF PERTINENCE TO CHEMICAL ENG I NEERS It is ideal to encompass both the two-phase and three phase systems in the teaching of tluidization as these two systems behave significantly differently Salient subjects con cerning gas-solid fluidization and gas-liquid-solid fluidiza tion are given below. Flow surrounding a bubble. two-phase theory. and flow segregation introduced so students will be familiar with the use of proper assumptions for developing theories that capture the dominant behavior features. Gas-solid fluidization phenomena are strongly dependent on the physical properties of the solid particles employed. Therefore, it would be appropriate to introduce the classifi cation of fluidized particles to the student. Particles are classified into four groups (i.e., Groups A, B, C, and D) based on their fluidization behavior. This classification known as Geldart's classification,1 1 1 is shown in Figure 1, where particles are classified in terms of the density differ ence between the particles and the gas, ( Pp p ), and the average particle diameter dP. Figure 1 was obtained empiri cally and has been widely adopted in the fundamental re search and design of gas-solid fluidized beds. Group C comprises small cohesive particles (dP < 20 Group A particles, with a typical size range of 30 to 100 m, are readily fluidized. For Group B particle fluidization, there exists no maximum stable bubble size. Group D com prises coarse particles (dP > 1 mm) which are commonly Chemical Engineering Education

    PAGE 35

    pro cesse d b y s poutin g W e would then demon s trate to the s tud e nt the rise of a bubbl e or s lug in a den se gas-so lid sus p e n s ion u si ng a s imple known experiment that involves plac ing fine particle s ( FCC Group A particle s) in a sea led 6000 3 4 000 "If 2000 e A ci. l000 it, Aeratable 5 00 ;ii-, C C oh es iv e ..... \' 1 00 I 10 ]0 4 ]O l d P, m Fig u re 1 Ge l dart's classification of flui diz ed particles .' 11 tub e. The tube is 58 c m lon g and 87 % full of particles. As s ho w n in Figure 2 w h e n the tube is flipped up side d ow n a s lu g rises in the tub e (F i g ur e 2a). When th e sl u g ex it s to the top of the tube it l eaves behind a d e n se particle bed (Fig ur e 2b) that h as a h e i g ht high e r th a n the p acke d co ndition of th e particl es (Fig ur e 2c) The phy s ical implication from com parin g Figures 2b and 2c i s th at a bed of fine particles ca n be expa nd e d b y gas to an ex t e nd ed h eig ht without th e formation of bubbles. This wo uld l ead to a discussion of the o n se t of bubblin g. Bubbl es are formed as a result of the inherent instability of gas solid systems. The instability of a ga s -solid fluidized bed is charac terized by fast growt h in l ocal voidage in response to a syste m p e rturbati on B eca u se of the instability in the bed the local vo id age u s uall y grows rapidly into a s hap e r ese mblin g a bubbl e Although it i s not a lw ays true the initi a tion of th e in s tabilit y i s usually p e ceived as the onset of bubblin g, which mark s th e transition from particul a te fluidi za tion to bubblin g fluidization The theoretic a l ex plan a tion of the ph ysica l origin a nd prediction of the onset of the instability of gas-so lid fluidized b eds h as b ee n a tt emp ted .l2 1 Efforts h ave foc u s ed o n the primary forces be hind the stabi lit y among interparticle co tact forces particle-fluid inte rac tion forces a nd particle-particle interaction via par ticle ve locit y fluctuation Fluidiz a ti o n of fine particl es ( Group A particle s) without the for mati o n of bubble s is known t o b e in the particulate fluidiza tion regime. For lar ge a nd/ or hea vy par ticle s (i. e Group B or D particles) par ti culate fluidization does n o t exist. Th at is the onset of bubbling co in ci d es with that of minimum fluidization of th e packed b ed. Figure 2. Simple flu idiz a tion experiments : (a) slugging regime (b) parti c ulat e fluidization regime and (c) packed-bed regime Most bubbl es in gas-so lid fluidized bed s are of s pheri cal ca p or ellipsoidal cap shape Configurations of two b asic typ es cf bubbl es fast bubbl e (clo ud e d bubbl e) a nd s lo w bubble (clo udl ess bubbl e)l 3 1 are sc h matically d epicte d in Figure 3. Th e c l o ud is the region establis h ed b y the gas that circ ul ates in a closed l oop bet wee n the bubble and it s s urroundin g s Th e cloud ph ase ca n be v i s uali ze d with the aid of a color tracer gas bubbl e For exa mple when a dark brown NO 2 bubble i s injected into the bed (see Fi g ure 4), the light brown color s urroundin g the bubbl e repre se nt s the cloud region. 141 When the bubble-ri se ve lo ci t y is higher than the interstitial-gas ve lo city a clouded bubbl e for m s in w hich th e circ ula tory flow of gas takes pla ce b tw ee n the bubbl e and the cloud, as s h ow n in Figure 3a. Th e c loud s ize decrea ses as the bubble-ri se velocity increases. As the a ----------------b----------~ . : C loud Gas through fl ow Crr culanng clo\14 ---::;_ -, / ~ yt = ; ~I "' ~ ) } 1 tK : ; 1 \ : t;: ,, ;: ,.:i,fjJ t i Wake Fig u re 3. Bubbl e configura tion s and gas-flow patterns around a bubbl e in g as -so lid fluidized beds. ( a) Fast bubble ( cloude d bubble) U b > U mf / amf; (b) Slow bubble (cloudless bubbl e) Ub < Umf I amf. Spring 2000 1 29

    PAGE 36

    bubble ri se velocity become s sig nificantly higher than th e interstitial-ga s velocity (i.e., U mf /
    PAGE 40

    ri ;:,0 5 "' ~0 2 3 4 2.. g, I() .P, 2 3 4 "" ;; "' "' 0 J z 2 g: b= 4 2cm 4 U 8 =32.3cm/s Reb 1 36 x 10' ~6 ::, Fig ur e 11 Entrainm e nt of particl es b y bubbl e wake and drift indu ced bya rising bubble (from T s u c hiya e t a/. 1 14 1 reproduced with p e rmi ss i o n of th e American In s titut e of C h em i cal Engineers 1992 A!ChEJ. Figure 12. Pr eci pitation of CaCO 3 par ticles from a rising CO 2 bubbl e (from Tsutsumi, e t al./ 1 reproduced with permission). "3 92 x1 D-' cm/s at 21 rc 5 6 7 8 9 y/b -2 1 0 2 0 2 -..., Staying in near wake 4 6 U 1 4.3 cmls Figure 14. Instantaneous local liquid-solid mass transf er coeff i c i e nt s in the wake region and the effects of vor ti ces on the primary wake (from Arters e t al./ 1 reproduced wit h permission). x vertical downward di s tan ce from th e bubbl e base y horizontal right-hand distance from th e ce ter of th e bubbl e bas e b bubbl e breadth U 8 absolute bubbl e rise veloc it y R eb bubble R eynolds number based on the bubble breadth I= 0.13 sec .._ Figure 13. 0 3 bubble rising in a KI-star c h so lu tion and 0.46-mm g l ass-bead I OlE OO l~Jl:: 0 1 l!(IF..{I J 14:lE-0I l.10C O I 11011:,-01 lll!OC-ll l BCQE.01 @ (i:-{/1 IJOC-0 1 ,_,,. '"""' 76,~0 1 :I' ..(! 1 '''"'" '""" ' .. EEX --0 1 !.t:E-01 61~ 1 6[l{E 4] 1 ''''''" 6 4 QE.-O l ~JO[ 0 1 50CC-0 1 .,.., &IE -Cl 44'~ 0 1 4 ~ -il l fluidized bed FLUIDIZED BED TWO-FLUID APPROACH Au Vokme Fraction Max= 1 OOOE+OO Min= '4 000E-01 T Figure 15. Simulation results of bubble formation in a gas-so li d fluidized bed usin g Fluent

    PAGE 41

    [] Liquid-Solid Mass Transfer Mass transfer from the liquid to the surface of the solid, and hence the reaction rate, are governed, apart from the activity of the so lid, by the local flow patterns of the liquid relative to the so lid. Due to the unique flow structures associated with the liquid in the wake and the presence of s olid particle s in thi s region, it i s of interest to examine the interaction between a solid particle and a bubble wake and its effect on liquidso lid ma ss tran s fer. The instantaneous value of the ma ss tran sfe r coefficient, k for a s ingle particle in a two-dimensional liquid-solid fluidized bed s ubjected to the di st urbance of a s ingle rising gas bubble. can be measured by an electrochemical method using tethered particles The method mea s ures the limiting current and thereby allows evaluation of k Visualization techniques can be employed to track the particle in relation to the bubble and bubble wake. Synchronization of the mass tran sfe r data acquisition with the video record allows a hi tory of the local ma ss transfer to be a nalyzed. 11 61 Figure 14 s how s the liquidso lid ma sstransfer behavior interactions of a particle with the bubble and the primary wake. The axes in the figure s are linked s uch that informa tion regarding the mass tran sfe r coefficient, event time and particle position with respect to dimensionless bubble coor dinate s can be cross-referenced. The ma ss tran sfe r coeffi cient is expressed in terms of k/k 0 where ko i s the liquid solid ma ss tran sfe r coefficient under liquidso lid fluidization conditions at the same liquid velocity. The most salient feature i s that the interaction of the particle with the wake region produces s ub s tantial increa ses in the mas s tran sfer coefficient. It can be seen in th e figure th a t a twofold in crease in mass transfer result s when a particle traveling directly underneath and along with the bubble i s ejec ted from the primary wake through the cross flow. The free shear layer formed at the bubble edge i s also found to pro duce significant increases in ma ss tran sfe r. Lesser increases are seen when the particle is exposed to a s hear layer not strong enough to pull the particle into the flow. COMPUTATIONAL FLUID DYNAMICS OF PARTICULATE SYSTEMS Computation is an area of great importance. Students should be kept abreast of the current approach in computation for particulate systems. In the following, a general background on the basic methods of particulate flow computation is introduced and is folJowed by an example of a s tate-of-the-art computational problem that my research group is tackling. General Background The computational fluid dynam ics approach has provided considerable in s ight into the dy namic behavior of multiphase sys tem s. The Euler-Euler/ 1 7 1 Euler Lagrange 1 1 81 and direct 1 1 91 numerical s imulation s are three widely u se d approaches for particulatesys tem compu tation. In the Euler-Euler method the individual pha ses are treated as p se udo-continuous fluid s, each b e ing governed by Spring 2000 the conservation laws expressed in terms of volume/time or ensemble-averaged propertie s The conservation equations are clo se d by constitutive relations that could be obtained from empirical relationship s, or theorie s. The dynamic mo tion of solid particles, especially for collision-dominated s hear flow s of s olid particle s, i s often simulated using ki netic theory 120 1 in which theoretical analogies between the gas molecule and solid particle s are applied. In the Lagrangian approach, the discrete particle s are treated as a group of point ma sses with their position velocity, and other quanti tie s being tracked ba s ed on the motion equation of indi vidual particles. The dispersed phase can exchange momen tum mass, and energy with the fluid phase In the dispersed phase particle-particle collision dynamics characterize the particle-particle interactions. In direct numerical simulation, the fluid flow could be solved by using finite difference/ volume/element di sc retization of the Navier-Stokes equa tion s or the lattice-Boltzm a nn or Lagrangian multiplier method Direct numerical s imulation s require no empirical constitutive equations and could provide detailed informa tion about flow surrounding individual particles The Euler-Euler and Euler-Lagrange approaches have been incorporated in many commercial software packages. Fluent ( by Fluent, Inc ), CFX (by AEA Technology), Flow3D (by Flow Science Inc.), and CFDLIB ( by Lo s Alamos National Lab ) are so me of the common packages used in academia and indu s try for chemical proces s applications. The results for a s imulation of the bubble formation process in a gas so lid fluidized bed using Fluent 4.47 are s hown in Figure 15 In thi s example, the Euler-Euler two-fluid approach is used to so lve the gas and solid flow in a fluidized bed. The rectan g ular domain i s 0.4 m wide by 0 6 m high and i s filled h a lfway with a fluidized bed. The parti c le diameter used is 0.5 mm with a density of 2610 kg/m 3 Air is used as the gas pha se, which has a density of 2.3 kg/m 3 and a viscosity of 1 7 x 10 5 kg/ms Initially, the bed ha s a uniform vertical air flow of 0 .28 4 mis introduced from the lower boundary When a s imulation is started, a vertical air jet is injected from the lower center of the fludized bed. The orifice width of the air jet is 0 03 m. The bubble s ize i s seen to increase s ignificantl y with time. Similar result s were presented ear lier b y Sinclair 12 11 u s ing Fluent 4.32. Examples of State Of-The-Art Computation My re s earch group has been engaged in computation code devel opment for simulation of the gas-liquid-solid fluidization systems.'2 2 1 The di sc rete-phase approach is employed with the volume-averaged method the discrete-particle method (DPM), and the volume-of-fluid method (VOF) used to ac count for the flow of liquid so lid and gas phases respec ti ve l y A bubble-induced force model ( BIF ) a continuum s urface force model ( CSF ), and Newton's third law are a pplied to account for the couplings of particle-bubble (gas), gas-liquid, a nd particle-liquid interaction s, respectively. A 1 35

    PAGE 42

    clo sedi s t ance interaction model ( CDI ) i s in cluded in the particl eparticle collision a naly s i s, which co nsider s th e liquid int e r s titi a l ef fects b e tween collid in g particle s. The fo ll ow ing pre se nt s representative result s for a single bubble rising and p ar ticle entrainment b y a s in g le bubble in a liquidso lid fluidized b ed und e r ambient conditions and multi-bubble s ri s in g in a liquidso lid fluidized bed under hi gh-press ur e conditions The b e h avior of a s in g le bubble ri s ing in a liquidso lid fluidized bed s u s pen s ion under ambient co ndition s is s imulat e d in Figure 1 6 One thou sa nd particl es with a d e n s ity of 2,500 kg/m 3 and a diameter of 1 0 mm are u se d as the so lid pha se. An aq u eo u s glycerin so lution (8 0 wt %) i s u se d as th e liquid pha se. The s uperfici a l liquid velocity is 5 mm/ s, y ieldin g a solids holdup of 0.44. It can be see n that the bubble i s of s pherical-cap s h ape rising recti linearl y. Also shown in the figure are photo graphs of a s ingle bubbl e rising in a liquidso lid fluidized b e d ob t ai n ed ex p e rim e ntall y under the sa me operating conditions as tho se of the s imu l ation As s hown in th e figure, the s imulat e d and experimenta l re s ult s of the bubble ri se ve locit y and the bubble s h ape generally agree Figure 17 s how s th e s imulat e d results of particl e e ntrainment b y a bubble from the b e d 0 0 0 0 0 0 , Figure 16. Simulation and experi m e ntal resu lt s of a single bubble rising in a liquid -so lid fluidized bed. Figure 17. Simulation results of a bubble e m erg in g fro m th e surface of a liquidso lid fluidized bed. 0 a (a) I = 1 0 {b)1=1 0 +0.2s -0 () 0 0 0 0 0 ,, ,___ _ _____ ( c ) I = 1 0 + 0.4 s (d)1=1 0 +0 6s Figure 18. Simulation results of mu l ti-bubble r i s in g in a liquid-solid fluidized bed at a pressure of 17 .3 MPa 5 =0.17; d p =0 .5 mm ; Pp=1,500 kg / m 3 ; db=4. 0mm ) 1 36 10.00 ,---------~ frame 1 10.00 ,---------, frame2 : 8 \ :;;:\. :;.~i: '. ..... f. _; :._:;. ~~: 6 00 0 00 6 0 (a) I= 1 0 10 00 ,--------~ frame3 C) .. -::. -: .. :.:,:: ... t .. . :,,.x : : ., . :, 6 00 0 00 6 0 1 0 00 ,--------~ frame 4 6 00 0 00 .~:. . -~l :-t. ;-t ~. ~.-( . r.: : _; 6 0 (b) t = 1 0 + 0 2s (d) I= 1 0 + 0.6 s Figure 19. Simulated ve lo city vector fie ld s of f luid s for the cond i tions g i ven in Figure 18 Chemical Engineering Educat i on

    PAGE 43

    surface. As seen in the figure, particle s are drawn from the upper s urface of the s uspension into the freeboard of the bed through the wake behind the bubble ; and particle-containing vortices are s hed from the wake in the freeboard, consistent with the experimental observation s hown in Figure 11. Fig ure 18 s how s the rising of four bubbl es in a liquidso lid fluidized bed with a so lids holdup E 5 of 0.17 and a pre ssure of 17 .3 MPa The corresponding ve locit y vec tor fields of fluids are shown in Figure 19 As can b e seen from the figure, the four bubbles are not rising at the sa m e ve locit y even though their initial conditions are the sa me Clearly complex interaction s among gas bubble s, liquid, a nd soli d particle s result in nonuniforrnit y of the flow field s hown in Figure 19 yielding uneven rise characteristics of th e bubbl es. CONCLUDING REMARKS I would like to conclude my lecture with the following thought s: Multiphase fluidization is a subject of importance to chemica l eng in eering education as it encompasses the fundamental ph ys ics that govern multiphase fluid and particle mechanics and their interactions. Furthermore, interest in the subject is heightened because of its significant indu strial applications. For gas so lidfluidi za tion, topics of most relevanc e include, for low-velocity fluidi z ation particle and bubble d y namics, bed stability, bubble-phase and emu lsion-phas e interactio n and two-phase theory. For high-velocity gas-solidfluidization, the co re topi cs are particle segregation and clust e ring. For gas -liquidsolid fluidi z ation, the bubble-wak e d y namics is ke y to the fundamental characterization of transport phenomena. In gas -solid or gas -liquid-s olid fluidization th e particle property, which is an important operating v ariable affects the fluidi z ation r egimes and their transitions. Th e com putational fluid dynamics approach has provided a viab l e means for flow system and chemical rea c tor c hara cter i zat ion. Although a v ailable commer cia l c odes may not always y i e ld accurate predi c tions the familiarity of students with these computational tools wou ld fortify their capability of understanding co mpl ex multiphase fluidization systems. ACKNOWLEDGMENTS This lecture is dedicated to the memory of Profe ssor Shao Lee Soo of the University of Illinoi s, Urbana. I benefited from Prof. John Da vi d so n 's lecture on fluidization seve n years ago when I was on sa bbatical at Cambridge Univer s ity in which he demon s trated the s imple s lu g -ri s ing device that I duplicated s hown in Figure 2 I am indebted to Prof. Jack Zakin and my re searc h group member s, Dr. Jianping Zhang Mr D.-J. Lee Mr. Brian McLain, Mr. Will Peng and Mr. Guoqiang Yang who have provided constructive Spring 2000 feedback in the prepar a tion of thi s lecture mat er ial. REFERENCES 1. Geldart D ., "Types of Gas Fluidization, Powd e r T ec h. 7 285 ( 19 73 ) 2. Anderson T.B ., and R. J ac k so n "A Fluid Mechanical De scription of Fluidized B e d s: Stability of the State of Uniform Fluidization ," I&E C Fund ., 7 12 ( 196 8) 3. Kunii D. and 0 L eve nspiel Fluidi za tion Engin eer ing, 2n d ed. Butterworth-Heinemann Boston MA ( 1991 ) 4. Rowe P N. Experimental Properties of Bubbles ," in Flu idization, J.F. Davison an d D H arr i son, e d s., Academic Pr ess, New York NY ( 1971 ) 5. Davison J F ., and D. Harri s on Fluidi ze d Particl es, Cam bridge Universit y Pres s, Cambri dg e, UK ( 1963 ) 6 Reuter H. Druckverteilung um Blasen im Gas-Feststoff Fliefibett ," Chem-lng.-Tech ., 35 98 ( 1963 ) 7. Reuter H ., Mechani s mu s d e r Bla sen im Gas-Feststoff Fli e fibett ," Chem-lng.-Tech., 35 219 ( 1963 ) 8. Stewart P.S.B. I so lat ed Bubbles in Fluidized B e d s: Theory and Experiments ," Tran s lnstn. Chem. Engrs. 46 T60 ( 1968 ) 9. Toom ey, R.D and H .F Johnstone "Gaseo u s Fluidization of Solid Particles ," Chem. Eng. Prag ., 48 220 ( 1952 ) 10 Fan, L.-S. and K. T s uchi y a Bubbl e Wak e Dy namics in Liquids and Liquid-S ol id Suspensions, Butterworth Heinemann Bo s ton MA ( 1990 ) 11. T s uchi ya, K. and L.-S. Fan, Near-Wak e Structure of a Single Gas Bubble in a Two-Dimensional Liquid-Solid Flu idized B e d : Vortex Sh e dding an d Wake Size Variation ," Chem. En g. Sci., 43 1167 ( 19 88 ) 12 Massimilla L. N Majuri and P. Signorini Sull assorbimento di Gas in Sistema: Solido-Liquido Fluidizzato ," La Ri ce r ca S c i e ntifi,ca, 29 19 34 ( 1959 ) 1 3. Tsuchiya K. T Miyahara and L.-S Fan 'Visualization of Bubble-Wake Interactions for a Stream of Bubbles in a T wo -Dim e n s ional Liquid-Solid Fluidiz e d Bed ," Int J Multipha se Flo w, 15 35 (1989 ) 1 4. Tsuchiya K. G. H Song W.-T Tang and L.-S. Fan Par ticle Drift Induced by a Bubble in a Liquid-Solid Fluidi zed Bed with Low-Density Particles ," AIChE J ., 38 1847 ( 1992 ) 1 5. Tsutsumi A. J .Y. Nieh and L -S Fan Rol e of the Bubble Wak e in Fine Particle Production of Calcium Carbonate in Bubble Column Systems ," I&E C R es., 30 2328 ( 1991 ) 16. Arters D. C., K. Tsuchiya an d L.-S. Fan, "So lid-Liquid Mass Transfer in the Wake Region Behind a Single Bubble in a Liqu id-So lid Fluidized Bed," in Fluidization VI, J R. Grace, L W Shemilt and M A. Bergougnou eds., Engineering Foun d ation, pp. 507-514 ( 19 89 ) 1 7. Sinclair J L. and R. Jack s on "Gas -Particl e Flow in a Verti cal Pipe with Particl e -P article Int eractions," AIChE J. 35 14 73 ( 19 89 ) 18 Hoomans B P B. J.A.M Kuip ers, W.J Bri e l s, an d W.P.M. van Swaaij "Discrete Particle Simulation of Bubbl e a nd Slug Formation in a Two-Dimensional Gas-Fluidized Bed: A H ard Sphere Approach ," Ch e m. Eng. Sci. 51 99 ( 1996 ) 19 Jos e ph D D. "Inter ro gatio n of Num erica l Simulation for Modeling of Flow Induc e d Microstructure ," ASME FED 189 31 ( 1994 ) 20 Jenkins, J T. and S.B. Savage A Theor y for the Rapid Flow of Id entical, Smooth, Nearly Elastic Spherical Par ticl es," J. Fluid M ec h. 130 187 ( 1983 ) 21. Sinclair J L ., "C FD Case Studies in Fluid-Particle Flow ," Ch em. Eng. Ed. 31 108 ( 1998 ) 22. Li Y. J Zhan g, an d L -S. Fan, Numerical Simulation of Gas-Liquid-Solid Fluidization Systems Using a Combined CFD-DPM-VOF Method : Bubbl e Wake B e havior ," Ch e m. En g. Sci. 54 5101 ( 1999 ) 0 1 37

    PAGE 44

    .,a ... fi 11111 3.._1e a rn in g :::..__ ____ __ __ __,) TOWARD TECHNICAL UNDERSTANDING Part 5. General Hierarchy Applied to Engineering Education* J.M. HAILE Clemson University Clemson, SC 29634-0909 I n the first papers in this series, I pre sente d a special hi erarc h y of technjcal und erstan din gs 1 1 3 J based o n my expe ri e nc e in trying to h e lp st ud e nt s l ear n and inform ed b y our curre nt knowledge of the struct ur e a nd function of th e human brain. In the previou s p aper, 1 4 1 I s howed ho w the s pecial hierarchy is related to a more general hie rarc hy developed by Donald 151 a nd independently, by Egan. 1 6 1 In di sc u ss ing the ge neral hjerarchy I adopted Egan 's nom e clature, whjch identifie s five le ve l s of human understand ing s: so mati c, mythic romantic, philosophic, a nd ironic Each lev e l co rre s pond s to a s p ec ifi c mode for ge tting thought s out of th e mind and into forms b y which they can b e di s sec ted analyzed, and r easse mbl e d To recapitulate, the so mati c l eve l includes tactile learnin g, mythic corresponds to oral l ear ning, romantic involv es gra phic s and written learn ing, philo so phic refers to learning by formal reasoning and the ironic level encompasses exce ption s, limit a tion s, and learnin g by modeling. It i s the philo so phic level that encompasses the basi c cog nitive ski ll s required of e ngin ee r s; these include use of for mal lo g ic mathematical reasoning, critical thinking, and problem so l v ing But the s pecial and general hierarchical model s are both integrative; that is pro gressio n to a hi g her level requires the individual to ma ster ski ll s and reor ga niz e knowl e dge gained at lower level s Consequently, students cannot develop facility with philo sop hic activities until they have mastered lower-level cognitive s kill s. In thi s paper we illustrate how the five cognitive level s can be u se d to guide teaching and learning activities appropriate for engineering students To do so, we apply each level to Par t 1 Bra in Structur e and Function," CEE 31 (3), 152 (1997 ); Part 2 Elementary Levels ," CEE 31 ( 4 ), 214 (1997 ) ; P a rt 3, Advanced Levels ," CEE, 32 (1) 30 (1998) ; Part 4, '.A General Hi erarch y Ba sed on t h e Evolution of Cognition CEE 34 ( 1), 48 ( 2000). the concept of energy. A s noted pr evio u s l y ,1 41 energy is already a highl y abstract co nc ep t characterist i c of th ose em ployed at a philo sop hi c l eve l of und ersta ndin g; however, the word e nerg y is common in daily discourse and therefore, it is familiar to s tudent s. Nevertheless, freshman and sop ho more engineering st udent s ge nerall y h ave only vag u e no tions of the concept, a nd often confuse energy with force and pr ess ure. For the se rea so n s, e nergy is a good concept for s howin g how the hierarchy could be applied We emp ha size that th e sugges tion s here are fragmentary and s up erfic i al; they are intended only to offer a flavor of the kinds of ac ti vities th at cou ld b e pursued. Note that o ur goa l s are not so much to develop, say, somatic and mythic mode s of technical under stan din g, but rather to appeal to suc h mode s for under sta nding a particular concept. SOMATIC U N DERSTANDING At th e mo st basic l eve l our objectives are to help st ud e nt s obtain a phy s ical feel" for kinds an d quantities of e ner gy For example we might have st ud ents try to increase the temperature of water in a bowl b y u si ng a hand-driv e n egg beater. Or we might ha ve them manually compress air in a piston-cylinder device s u c h as a lar ge medical syri n ge. To test whether energy is extensive, s tudent s could measure the time required for a 500-watt microwave oven to brin g a cup of water to boil; then they could repeat the he at in g u s ing two cups. More elaborately, we could invert a bicycle attach a fric tion-driven electric ge ne rator to the rear wheel a nd run an J.M. Haile Professor of Chemical Eng ineeri ng at Clemson Uni versity is the author of Molecular Dynam ics Simulation published by John Wiley & Sons in 1992 and is the 1998 recipient of the Corcoran Award for the Chemical Eng i neering Di visio n of ASEE. Co p yrig ht C h E D i vis i on of AS E E 2000 1 38 Ch e mi c al En g in ee rin g Edu c ati o n

    PAGE 45

    electric circuit from the generator to a light bulb .(7 1 Students would then be asked to keep the bulb burning by cranking the pedals by hand. If we add a voltmeter and ammeter to the circuit students could determine the amount of power they generate. Such exercises need be only semi-quantitative, for the intent is to help students connect physical effort to measurable changes in temperature, volume, and current flow. served. Other common misconceptions surround the distinc tions between quantity of heat and the intensity of heat; thus, it is a difference in intensity (temperature), not quantity, that drives heat transfer. More subtle confusions are attached to the possibilities of changing temperature without heat transfer and transferring heat without a tempera ture difference. At the most basic somatic level students confront physical situations and devices; at a higher level, we try to appeal to their somatic experiences without fu1ther direct contact. Such attempts might take the form of simple questions requiring modest computations. For example, if the cost of electricity is $0.1 per kilowatt-hour how much does it cost to burn a 100-watt light bulb for one hour ? The en ergy density of a typical gasoline is about 45 MJ/kg; if your car gets 25 mpg, estimate the amount of energy (kJ) your car uses per mile. The energy density of ethanol is 30 MJ/kg; estimate the amount of energy (kJ) in a 750m] bottle of white wine that is 12 % alcohol by volume. 1 8 1 The key here is to contrive ques tions that make contact with situations that are familiar to students, else the so matic ad vantage is lost. ... the special and general hierarchical models are both integrative ; that A s another exercise of the oral compo nent, each student could be asked to give a three-minute presentation on the origin ety mology and historical significance of one piece of energy-related jargon Appropriate words could include energy itself, horse power Btu, watt, Joule, kinetic, potential efficiency, and friction. is progression to a higher le v el requires the indi v idual to ROMANTIC UNDERSTANDING master skills and r eorganize knowledge gained at lower levels Consequentl y, students cannot de v elop facili ty wi th philosophic To identify the principal features on the energy landscape we can have students list various forms of energy: kinetic potential, chemical, nuclear radiant, electrical, mag netic heat work, etc Can these be distrib uted among certain categories? Students should also list kinds of molecular energies: kinetic, potential from intermolecular forces, electronic and nuclear. MYTHIC UNDERSTANDING An important binary alternative that is fun damental to any study of energy is this : Does energy come in only one form, or are there many forms? If there are many, can we conacti v ities until the y ha v e mastered lower le v el cognitive skills To exercise the narrative component of romantic understanding students could be asked to contrive a chain of conversions; for example living plants convert radiant en ergy to chemical, people eat plants to convert among them? Can the students cite examples of conversions in both directions between two forms? For example, electric motors convert electrical energy to mechanical, while electric generators convert mechanical energy to electrical. Similarly solar cells convert radiant energy to electrical, while light bulbs convert electrical energy to radiant. Are some conversions between energy forms easier than others ? Do some conversions occur naturally? Are some conversions undesirable so that we seek to prevent or restrict them? Are some forms primarily for energy stor age? These can lead to such questions as: What common devices are used to store energy? What is the defining characteristic of a machine? I s there a distinction be tween a motor and an engine? One way to exercise the oral and narrative components of mythic understanding is to discu ss with students old miscon ceptions about energy and forms of energy. Examples in clude the ancient idea that fire is an element, or, in an updated version, that heat is a thing ("caloric") that is conSprin g 2000 vert chemical energy to other forms of chemical energy human muscles convert the stored chemical energy to mechanic energy, the muscles might crank a hand generator that converts me chanical energy to electrical, and the generator might be wired to a light, which converts electrical energy back to radiant. To identify extremes, we would offer students numerical examples of situations involving large amounts of energy: the potential energy behind the Hoover Dam, the energy required to launch a Saturn V rocket the energy consumed by all automobiles in the U.S. in one year. At the other extreme, we might cite the energy required by one light emitting diode (LED), the amount to depress one key on a keyboard or the amount used by a hummingbird during five minutes of flight. 181 To appeal to human interests and motivations we could start by working out an estimate of the energy-hence man years of effort-required to construct one of the Great Pyra mids of Egypt. 191 Then we could note that the desire to replace man-power with machine-power motivated the in 139

    PAGE 46

    dustrial revolution. This leads to a description of socio economic conditions prevailing in Europe in the early l800s and, in particular, to a discussion of Joule's careful, system atic, extended, experimental studies of the relations between heat and work. We could describe Joule's paddle-wheel experiments, which illustrated the equivalence of heat and work and led to an identification of internal energy. We would emphasize that these crucial experiments discredited the caloric theory of heat and laid the foundations for articu l ation of the principal of conservation of energy. or apartments etc. The educational advantage here comes when students can see and touch objects, and they attempt to represent relations among those objects abstractly on paper. PHILOSOPHIC UNDERSTANDING At the philosophic level our first goal is to find those unifying generalizations that connect the things and con cepts encountered at the somatic, mythic, and romantic lev els: the stories, the devices and equipment the many concepts, the transformations among concepts, the ex Another instructive story is that of the Haber-Bosch process for the catalytic for mation of ammonia from its elements under high tem peratures and pressures. That process was first used to make nitric acid for ex plosives and thereby en hanced Germany s ability to prosecute World War I, but after WWII, it made pos sible large-scale production of ferti l izers that sustain the world s growing popula tions. Thus, we have an ex ample of the common di lemma of technology being used and misused. But in the context of energy usage, this story illustrates one way in which technologies evolve: the fundamentals of the Haber-Bosch process are unchanged, but improve ments have reduced the en ergy costs of the process by more than an order of mag nitude-from 380 MJ/kg of NH 3 in I 930 to 35 MJ/kg in 1990.L 8 1 Many important Abstraction tremes, etc The cognitive hierarchy guides us in how this is to be done. We em phasize that we do not at this point, confront students with the answer-the gen eralized energy balance. Rather, we proceed system atically from concrete situ ation to abstract generaliza tion, following the left leg I Ge neraliz a ti o n Engineering Education I Tr ansfe r en ce I Co7tua l izati o n Co ncr e te S it u a tio n Engineering Practice Ex t e n s io n C o ncrete Situ a ti o n in Figure 1. Our second goal is to help students develop the ability to use the gener alized energy balance which is represented by the right leg in the Figure Thus our pedagogical goal is dis tinct from the practical one Figure 1. Understandings of abstractions dev e lop in a bot tom-up strategy from concrete situations to abstract con cepts ; thus, in helping students learn new c onc e pts we should start with concrete and specific examples and move toward abstract generalizations We apply abstractions in a top-down fashion, however, from abstract notion to concrete situation. Thus in helping students learn to solve problems we should teach them to identify the generalized concept that applies and then to proceed deductively to th e ir par ticular situation We might start a philo sophic discussion of energy with equations that define individual energy forms such as mechanical work, electrical work, and changes in kinetic and potential en ergies. Then students would exercise those definitions by applying them to relatively chemicals have histories that can be exploited to appeal to students' romantic understandings; another example is the story of the Leblanc soda process, nicely told by Cook. 1101 Still another aspect of romantic understanding is embed ded in the graphical representations of physical objects and processes-plots schematic diagrams, and flowsheets. An effective initial exposure to these tools is to confront stu dents with objects and have them create schematics: cooling cycles in refrigerators or room air conditioners, the cooling water cycle on an automotive engine, the steam cycle at a power generating plant, the water lines through their houses 140 simple situations : a ) esti mate the speed of a crescent wrench as it hits the ground after a free fall from the top of a 30-foot distillation tower; b) estimate the work performed by an adiabatic air compressor; c) estimate the heat required to raise the temperature of 1 kg of water from 20 C to 100 C. (a) Co n c r ete S i tuation To start the progression on the left in Figure 1 we choose one of the concrete situations that the students have already encountered; a possibility is the compression of a gas in an insulated piston-cylinder appa ratus. Many choices are legitimate here, so long as the one chosen arises from a situation for which students have strong visual images. Ch e mi c al En g in ee rin g Edu c ation

    PAGE 47

    (b) Conceptualization We now lead the students to identify the concept associated with the concrete example We ask, what is it we, as engineers are likely to want to know about the compression? Presumably the amount of effort required That effort is conceptualized by a particular form of energy-the work; here it is adiabatic work, because the apparatus is insulated. Note that because of the ground work laid by the earlier somatic, mythic, and romantic exer cises the students should be able to participate actively in this discussion. Inversely, if those lower-level understand ings have been ignored and instruction starts here at the philosophic level, then many students will be immediately overwhelmed. Once students have recognized work as the appropriate concept, we th en have them calc ul ate va lu es for the adiabatic work under various sets of parameters applied to the piston-cylinder device (c) Transference At this stage we want students to apply the concept of adiabatic work to situations other than the piston-cylinder apparatus For example, we could pose problems involving adiabatic compressors adiabatic turbines, and adiabatic pumps. The objective is for students to recognize that all such problems belong to the same conceptual class. ( d ) Generalization Now we come to the difficult s tage at which we generalize away from the special case of adia batic work processes. Thus, we first relax the adiabatic con straint and consider workfree heat transfer situations; then we introduce processes invo l ving both work and heat trans fer. We emphasize the extent to which these situations are conceptually the same as, but practically different than, the adiabatic work processes considered earlier. Then we con sider steady-flow processes, with the introduction of flow work and the possibilities of changes in kinetic and potential energy. Finally, we end with a comp l etely abstract conse quence: the general energy balance which applies to any process This establishes the important connection among the various forms of energy; that is, this step relates the principal features of the energy landscape as identified at the romantic level. Our seco nd goa l is to help st ud ents learn how to use the general energy balance ; our strategy is now top-down as on the right in Figure l. Thus, we want students to appreciate that any situation they enco unt er is a special case, but we attack that special case by starting with the comp l etely gen eral energy balance and identifying the assumptions that are appropriate to the situation at hand Thus, we wou ld exercise the general energy balance applied to such situations as adiabatic processes on closed systems, to workfree processes on closed systems and to steady-state processes on open systems. The latter would include illustrations of the specia l forms known as the mechanical energy balance and Bernoulli's eq u ation The concrete applications wou ld in clude he at duties for h eat exchangers, sizing of pumps, tur bine s a nd co mpre ssors, analyses for thermal efficiencies, Sprin g 2000 etc. These kinds of activities are addressed in modern text books and many current learning strategies so they need little attention here IRONIC UNDERSTANDING To develop ironic understanding of energy, we would revisit the assumptions and limitations that pertain to the equations used at the philosophic level. For examp l e, most calculations of mechanical work can be done on l y for ideal ized processes in which the driving forces are differential. For real processes, in which the driving forces are finite, we need an efficiency obtained either from measurement or by estimation To calculate changes in internal energy and en thalpy, we often need an equation of sta t e that models the PVT behavior of the working fluids. Many texts restrict such calculations to the ideal-gas model but students must be introduced to more realistic models, and they must be in structed in the engineering task of selecting a model that is appropriate to their problem. Thus we must confront issues associated with model processes and model substa n ces For example we may be able to perform an exact analytic calcu lation of the required heat duty for a heat exchanger design, under the presumptions of particular model processes a nd model substances. However such exact calculations are still approximate to the degree that the assumed models fai l to represent the real situation. Students often have difficulty in reconciling how an approximate answer can be obtained from an exact calculation NONLINEAR INSTRUCTION For purposes of clarity the suggestions in the foregoing sections were presented in a linear progression that builds from somatic to ironic In practice, however, instructors of college students need not-indeed should not-proceed in such a linear fashion. Of course, somatic activities should generally be performed well before philosophic activities, but this does not mean we should avoid somatic and mythic digressions in an otherwise largely philosophic lecture. For example continuing with energy as the foca l point, the list ing of types of energy (romantic) could be done as soon as students acknowledge that energy comes in many forms (response to the mythic binary). Calc ul ation of the velocity of the falling crescent wrench (philosophic) cou ld be embel lished with the observation that the answer is independent of mass, so the velocity would be the same for a manhole cover or a pocket watch; this harks back to the tale of Galileo and the Leaning Tower of Pisa (a romantic reference). The dis cussion could be further extended by noting that the terminal velocity is independent of mass only when the air resistance is negligible; thus, we have done a model calcu l a ti on that yields an approximate answer (ironic). It is appropriate and beneficial to include somatic, mythic, and romantic a llu sions in a lar ge ly philosophic presentation; 141

    PAGE 48

    One of an instructor's goals is to find the level of understanding at which students are balanced between perplexity and c o nfidence; at that point of cre a tive tension, teaching is m o st effective and learning most rapid. however the inverse procedure i s counterproductive and s hould be avoided. That i s, we accomplish little when we introduc e abstract generalizations (such as the generalized energy balance ) and models (s uch as Raoult' s law) before so matic mythic, and romantic contexts have been estab lished for those generalizations and models. At this point, kind Reader, you might indulge in the fol lowing reflective exercise. Pick a course you have taught recently ; can you identify the cognitive level at which you did most of the teaching? For example, if most of the instruc tion took the form of anecdotes based on your industrial experience, then you were functioning at the mythic level with appeals to the somatic. If the instruction depended heavily on students reading the text technical report s, re sear ch journals, and on their making and interpreting plot s, schematic diagrams and flow s heet s, then you were working at the romantic level. If the instruction emphasized deriva tions, problem solving, and calculations, then you were at the philosophic level. If the instruction involved liberal doses of all of these, plus efforts to sensitize students to the uses and limitations of models then you were teaching at the ironic level. Any of these approaches may be right or wrong effective or not depending on the situation-that is depending on what your students needed at the time. So now ask yourself, why did you choose to instruct at the level you did? Was the choice made implicitly for your own convenience and com fort, or was it made explicitly to address the needs of the students? What was the outcome of your work? If the s tu dents were generally frustrated then your teaching level failed to match their needs. If the students were generally happy comfortable, and secure, then your efforts probably were limited to reinforcing their current levels of under standing. If the students were apprehensive but stimulated, then they were probably growing toward higher levels ( If only the reality were as simple and clear-cut as these ideal ized comments imply.) COMMENTS As an individual grows through levels of understanding lower-level understandings are not lost or displaced; rather they are reorganized and subsumed into high levels. Never thele ss, there is a loss associated with each transition; 1 61 for example, the admonition to be objective means to s trip away mythic and romantic associations such as emotion and anecdotal evidence, and rea so n logicaJly 151 But if the basic somatic and mythic understandings are not Jost, what i s? Part of the process of solidifying understandings at one level 142 includes the creating of mental scaffolding that will support the tran s ition to the next level. 1 3 1 Once the transition is com plete the scaffolding collapses. But some people become attached to the scaffo ldin g and experience a sense of los s when it collapses. Such is the nature of mental growth. How much under s tanding does an individual need at one level before he can move to the next ? The answer must be that it depends on the individual and the extent of his earlier experience with understandings at that level. For example, we conjecture that an individual who has developed somatic understanding s of some concepts will find it easier to de velop somatic understanding s of other concepts. The situa tion mu s t be much like the learning of a foreign language (or a computer language), which is m a de easier if the indi vidual has already learned another. In higher education, we appeal to so matic m y thic and romantic mode s of thinking to solidify the foundations for philosophic and ironic understandings Successful generalization (con crete to abstract) and extension (abstract to concrete) depend on facility with manipulating objects and con cepts at somatic and mythic levels. To illustrate let u s consider the value of somatic thinking. Marvin Minsky 1111 asks why we in s i s t on thingifying abstrac tion s. It can only be becau se thingifications help our think ing. Thus we think about energy as a thing, even though most forms of energy are abstract mathematical function s and are not objects at all. We do this so we can draw fruitful analogies between energy and mass: mass can flow into and out of systems, so can the energy-thing; mass is conserved, so is the energy-thing; ma ss is a resource whose use incurs cost, so i s the energy-thing. The power of such analogies is so well accepted that we take it for granted. But our familiar ity with s uch analogies mu s t not blind us to the significance of the achievement nor to the difficulty students have in accepting such analogies and using them In recent years engineering educators have renewed em phasis on the development of oral (mythic) and written (ro mantic) communication skills. But according to the cogni tive hierarchy these skills are valuable not merely for com munication ; rather they are important because they support subsequent development of understandings at the philosophic level. Further, the hierarchy asserts that oral skills develop before written skills; this reverse s the order employed at many institutions where oral skills are addressed late in curricula and after written s kills have been exercised. Students come to u s at many different levels of under standing, and our obligation is to help them grow to higher Chemical Eng in eering Education

    PAGE 49

    levels. It may be that many people in our society cannot become philosophic thinkers in any mathematical sense. Such people cannot become engineers, and their talents should be developed and applied in other ways Further, some people can function at the philosophic level but they may be more effective at some other level. These people can fulfill important and even creative roles in engineering; how ever, they cannot make informed judgment s about the best use of their talents until they have acquired some skill in philosophic thinking. Good teaching meets students at their current levels of understanding and attempts to push them to higher levels. This requires that instructors be able to cross readily among levels of understanding: this is an attribute of the ironic thinker. An obvious general rule-of-thumb is : If the student seems perplexed or confused, the instructor should push the discussion to a lower level of under standing. Equally important, but often overlooked is the inverse: If students seem confident and secure, then the instructor should push the discussion to a higher level of understanding One of an instructor s goals is to find the level of understanding at which students are balanced between perplexity and confidence; at that point of cre ative tension teaching is mo st effective and learning most rapid. This goal is relatively easy to achieve for a single student (a graduate student), but exceedingly diffi cult to achieve for a group of heterogeneous talents and personalities (an undergraduate class) Between say, 1950 and about 1990 engineering educa tion developed along ever-increasing theoretical, mathemati cal, and abstract lines; that is, engineering education came to be practiced almost completely at the philosophic level. Such is the natural progression that mirrors cognitive evolu tion. But in recent years we have come to realize that solely philosophic modes of instruction fail to help today 's stu dents. The typical reaction has been to dilute philosophic instruction in various ways For example, some chemical engineering departments have reduced the philosophic con tent of the curriculum by removing physical chemistry, quan tum mechanics, transport phenomena or computer program ming. In the courses that remain, the philosophic content has, perhaps, been diluted by over-emphasis on practical applications and tlowsheet design But too much attention to applications produces a catalog of special cases, when the objective should be development of organizing principles that generalize across individual situations. Further today s tlowsheet design tends to be accomplished with the aid of process-simulation programs; but without sufficient com mand of philosophic and ironic thinking, students can only allow such exercises to devolve to syntheses of black boxes with issues of engineering judgment relegated to default settings of the software Such dilutions of philosophic in struction actually make matters worse : 161 not only do they Spring2000 fail to develop philosophic thinking but they also leave students with confused and useless somatic, mythic and romantic understandings of technical material. Rather than dilute the philosophic content of engineering curricula, we should be moving in the other direction. As a rough generalization, our goals might be to use early engi neering courses to solidify understandings at the somatic, mythic and romantic level s. But though the levels would be emphasized, higher mode s would not be neglected; some foreshadowing of philosophic and ironic thinking must also be done. Then once s tudents begin the transition to philosophic thinking, the curriculum s hould develop that thinking by being more abstract and theoretical, not less This is the direction of growth for individuals cultures, and even engineering. Finally we ask, is there any level of understanding beyond ironic ? I think the only proper answer i s, we don t know Donald notes that each level of understanding incorporates a particular mechanism for off-line processing-for getting thoughts out of the mind so they can be more readily ma nipulated dissected and reassembled Y 1 At the somatic level, the off-line processor is the human body; at the mythic level, it is speech; at the romantic level it is graphics and writing; at the level of ( technical) philo so phic and ironic thinking, it i s mathematics and written chains of logic. So the question is can we find another mechani s m for out-of-mind process ing? Can the computer fulfill this role? I think we can only wait and see. ACKNOWLEDGMENTS It is a pleasure to thank Professor J.P. O Connell of the University of Virginia and Professor K. Egan of Simon Fraser University for offering constructive criticism on an early draft of thi s paper. REFERENCES 1. Haile, J.M ., Toward Technical Understanding: 1. Brain Structure and Function, Chem. Eng. Ed ., 31, 152 ( 1997 ) 2 Haile J.M "Toward Technical Understanding: 2. Elemen tary Levels ," Chem. Eng. Ed., 31, 214 ( 1997 ) 3. Haile J.M., "Toward Technical Understanding: 3 Advanced Level s," Ch e m. Eng. Ed. 32 30 ( 1998 ) 4 Haile J.M. "Toward T ec hnical Understanding: 4. A Gen eral Hierarchy Based on the Evolution of Cognition, Chem. Eng. Ed., 34 48 ( 2000 ) 5 Donald M. Origins of th e Modern Mind, Harvard Univer sity Press Cambridge MA ( 1991 ) 6. Egan, K. The Educat ed Mind, University of Chicago Press, Chicago, IL ( 1997 ) 7. O'Connell, J.P ., and T C Scott, private communication ( 1998) 8. Smil V ., En e rgies The MIT Press, Cambridge, MA ( 1999 ) 9. Stewart, I. Counting the Pyramid Builders," Sci. Am., 279 ( 3 ), 98 ( 1998 ) 10. Cook, M. The Leblanc Soda Process ," Chem Eng. Ed 32 132 ( 1998 ) 11. Minsky, M., Th e Society of Mind Simon and Schuster, New York NY ( 1986 ) 0 143

    PAGE 50

    Random Thoughts ... THE SCHOLARSHIP OF TEACHING RICHARD M. FELDER North Carolina State University Raleigh, NC 27695 I n hi s landmark 1990 monograph, Scholarship R econsid ered,f' i Ernest Boyer observed that the work of the pro fessoriate involves four different functions: discovery (adva n cement of the frontier of knowledge in a discipline) integration (putting re searc h discoverie s in broader con texts making connections across disciplines), application (a pplyin g the outcomes of discovery and integration to so cially consequential problem s), and teaching ( helping st dent s to acquire s pecified knowledge and develop s p ecifie d skills and attitudes). Boyer argued that these four activities are equally vital to the academic mission and that the acad emy s hould therefore recognize and reward scholarship equally in each of them. The sc holar s hip of di scovery-fro ntier research-is what most faculty members think of as academic sc holarship and while the sc holarship of integration and the scholarship of application may not occupy the same honored position in the faculty incentive and reward syste m most professors would at lea st agree that they exjst in principle. It's a different story with the scholarship of teaching Administrators and faculty members traditionally put teaching and scholarship in non-overlapping categories: some argue that "sc holar s hip of teaching is a contradiction in term s and many who concede it s theoretical po ss ibility question whether it can be validly assessed. What is the scholarship of teaching? According to Boyer, the elements that define teachin g as a sc holarly activity are ma s t ery of the s ubject being taught, knowledge of pedagogical methods that have been proven effective at promoting learning and sk ill development, and commitment to continuing personal growth as an educator. To thi s li st might be added involvement in educational r e searc h and development--de signi ng implementing, assess ing, and di ssemi nating innovative instructional method s and materials Research in education-related di sci plin es has a long-e s tab li s hed tradition. When done right, it adheres to the same s tandard s of sc holar s hip that characterize goo d engineering 144 research These sta ndard s have not been routinely observed in engineering education, howe ve r and until relatively re cently most of the literature ha s consisted of variat ion s on the theme, We tried thi s method and liked it and so did the s tudent s." This s ituation ha s begun to change in the pa st decade largely du e to the efforts of the National Science Foundation Divi s ion of Undergraduate Education and the Engineering Education Coalitions, a nd a growi n g percentage of the engi neering profe ssoriate i s now engaging in serio u s e ducational research and development. It i s no longer enough to say that everyone liked a method and the st udents performed well when it was u se d. The NSF project monitor and the J ournal of Engineering Education reviewers will inevitably re s pond with questions such as What learnin g objectives were you trying to achieve?" How well were tho se objective s met?" and How do you know-what were your assessment mea s ure s yo ur control populations, your statistica l analysis pro cedures, yo ur eva luation criteria?" How should the scholarship of teaching be assessed? Boyer propo ses making the sc holar s hip of teaching a le git imate basi s for awarding tenure and promotion to fac ulty member s who choose to make education a major focus of their careers. (Not all faculty member s s hould be expected to do so ) This propo sa l-which ha s predictably encoun tered considerable skepticism and so me outright ho s tility from administrators and profe sso r s-w ill gain widespread acceptance only if criteria for evaluating the sc holar s hip of teaching are established and generally agreed-upon. I pro pose that the evaluation shou ld consist of answering three que s tion s: Richard M. Felder is Hoechst Celanese Professor Emeritus of Chemical Engineering at North Carolina State University. He received his BChE from City College of CUNY and his PhD from Princeton He has presented courses on chemical engineering principles reactor design process opti mization and effective teaching to various American and foreign industries and institutions He is coauthor of the text Elem entary Principle s of Chemi cal Processes (Wiley 2000) Copyri g ht ChE Division of ASEE 2000 Che mi ca l Engineering Education

    PAGE 51

    1 To what extent did th e t eac hing qualif y as a scholarly activity? An swe ring this que s tion requires evaluating the faculty member 's subject knowledge, pedagogical knowl edge, commitment to growth as an educator and involve ment in educational research and development. 2. How effective was the teaching ? How well has the faculty member's teaching motivated s tudent s to learn and promoted their acquisition of desired knowledge skills, and attitudes? 3. How effective was the educational r esearch and dev e opment? How well were the faculty member 's educational innovations designed implemented assessed and evaluated, and disseminated ? What has been their impact on engineer ing education ? The data that can be used to answer these questions fall into four categories : archival data (lists of courses devel oped and taught, representative instructional materials and student products; numbers of undergraduate and graduate students advised and faculty colleagues mentored ; disciplin ary and education-related conferences and workshops at tended ; journals subscribed to; conference pre se ntations seminars, and workshops given; articles, book s, and courseware published); learning outcomes assessment data (test results, evaluations of written and oral project reports TABLE 1 and other s tudent product s, student self-assessments); sub jective evaluations b y others (student end-of-course ratings retrospective s tudent and alumni ratings, peer ratings, awards and recognition received reference letter s); and self-assess ment data (s tatement of teaching philosophy and goals, self evaluation of progress toward achieving the goals). A subset of the se items gathered into a teaching portfolio provides a so und ba s i s for assessing the scholarship of teaching. Glassick, et al. 121 suggest the following standards for evaluating educational innovations : ll Clear goals: I s the basis of the work clearly stated, the que s tion s addressed important and the objectives realistic and ac hiev a bl e? ll Adequate pr e paration : Does the scholar display an un der s tanding of existing sc holarship in the field and the skills n ee ded to asse mble the neces sa ry re so urces and do the work? ll Appropriate methods: Were the methods u se d appropri ate for the goa l s, appl i ed effec ti ve l y, and suitably modified when nece ssa r y? ll Significan t r es ults: Were the goa ls achieved? Did the work contri but e s ignificantly to the field ? ll Eff ec tive pres e ntation: Wa s the work pre se nted effec tively a nd with integrit y in appropriate forums? ll R eflec ti ve c ri tique: Do es the sc holar critically evaluate Assessment of the Scholarship of Teaching hi s or her own work, bringing an appropri ate br ea dth of evi dence to the critique and u s ing the critique to improve the qual ity of future work ? "' -ii' 1 -ii' 6 6 ] g,o u "" -0' i c;s "Statements of leaching philo so phy X Li s t of courses t a u g ht and de ve l oped, r e pre se nt a tiv e in s tructi ona l materi a l s X X R eprese nt at i ve s tud en t product s Learning outcomes assess m e nt data End-of-course s tudent rating s for th e p as t 2-3 years Retrospective se nior rating s X X A lumni ra tin gs X X Peer rating s X X Selfe valuation T eac hin g semina r s and conferences attended bo o k s read jo urn a l s s ub sc rib ed to X Pr ese ntation s in vi ted sem in a r s, a nd workshops o n teach in g given X Published p ape r s a nd monographs X X Publi s hed t ex tb oo k s and co ur s eware X X Awards and other recognition Ex t e rn a l re fe r ences Sprin g 2000 -::: ;,; !;: (.) ::: i'S ,: .; ::: "' "' .::: "Q .!'? 9 <:i "' ;, -::: X X X X X X X X .., "' .,:: u t.J "" s -8 X X X X X X X X X <:i ;, -~ & X X X X X X X X X X X Faculty members do ing educational re search that meets these standards are clearly contributing to the scholarly mission of the univer s ity. They merit advancement up the faculty ladder-tenure promotion, and merit raise s -no less than faculty members who meet institutional stan dards for disciplinary research Table l contains a matrix that may be used to custom-design a pro cess for assessing the components of the scholarship of teaching. Continued on pag e 152. 14 5

    PAGE 52

    .tA ... b-3._c_o_m __: p_u t, n ..: g :...._ ___ __ __.) TEACHING PDE-BASED MODELING TO ChEUNDERGRADUATES Overcoming Conceptual and Computational Barriers KARSTEN E. THOMPSON Louisiana State University Baton Rouge, LA 70803 I ntroducin g partial differential eq uation s ( PDE s) in the undergraduate engineering curriculum can be frustrat ing for both students and instructors. Many st udents gain a dislike of differential equations before the core engi neerin g curriculum even begins As engineering instructors, we then face a number of challenging problems. We mu st help the st udent s overcome conceptual barriers associated with the math and help them envision the phy s ical phenom ena bein g described Additionally we must devise model s for which we can obtain so lution s in limited timefr ames (i.e., class time or homework time). This latt er constraint is impo sed by computational barrier s, which often restrict u s to overly si mpli st ic problem s that have limited engineering relevance. Although the se computational issues have been a s tumblin g blo ck in the past, modern numerical packages prove to be very accessib le even to the undergraduate s tudent a nd have been s hown to improve learning in a number of ways [e.g., IJ Thi s paper describes a n instructional framework that was used for incorporating computational tools into a relatively s hort sectio n on PDE s during an undergraduate mod e lin g course. The rationale i s that by allowing the students to jump headlong into the so lution of real engineering prob lems the e mpha s i s in the classroom can change. Attention can be diverted away from arduous mathematical detail s (fo r th e moment) and focused on broader conceptual issues s uch as the general behavior of classes of equations, or u si n g a model for de s ign considerations. Thi s approach has a number of advantages. The st udent s' concurrent, hand s -on so lution of problems is a powerful method for illu s tratin g fundamentals (t hat otherwise see m abstract). More time can be committed to model development and general behavior (w hich r e main with students longer than the details of so lu tion technique s). And, the a bility to so lve real engineering problem s illustrates to the stu dent s the true power of mathem at ical modelin g. 1 46 A number of computational packages are avai l ab le an d appropriate for und ergra duate education Mathcad, Maple, Mathematica MATLAB and Polymath are a ll co mmon in chemical engineering education. 11 41 Fluent is especia ll y ef fective for CFD applications.'5 1 We foc u s on MATLAB in thi s paper largely because of it s uniqu e PDE Toolbox. At LSU MATLAB i s provided to the s tudent s via the dep a rtment's PC network. ( An academic unit can provide MATLAB on a numb er of PC s without a lar ge inv es tment b y purchasing a classroom kit.) Additionally Mathworks ha s recently relea sed a s tudent version of MATLAB (whic h i s the professional version plus popular toolkit s) that h e lp s s tudent s who wish to work at home In the remainder of thi s paper we will discu ss the context for thi s approach and one po ssi bl e s trateg y for breakin g down conceptual barri ers in the classroom. At the end, three example problems will be so lved using MATLAB, illu strat in g the type of modeling exercises that can be assig n e d to complement conceptual discussions in the classroom. CONTEX T The idea s and examples described her e were developed as part of a senior-level math-modeling course at LSU, which i s de sc ribed below When beginning the PDE sectio n of the course it was helpful to consider th e contexts in which our Kar s ten E. Th o m p son is an assistant profes sor of chemical engineering at Louisiana State Universit y where he has been since 1996 He received his BS degree from the University of Colorado and his PhD from the University of Michigan. He teaches courses in numerical methods math modeling and transport phe nomena His research interests include flow in porous media numerical methods and fluid mechanics Copyright ChE Division of ASEE 2000 Chemi c al Engine e rin g Edu c ation

    PAGE 53

    students typica ll y are introduced to PDE s. Unfortunately, the se encounters are infrequent and generally presented as asides to the main material. Consider three of the most likely places that ChE undergraduate s run into PDE s. First, they may arise at the end of an introductory numerical method s course. But u s ing Chapra and Canaie 1 6 1 as a guide, we see that PD Es are relegated to Chapter 29 s ugge s ting that the topic probably receives a quick overview n ear the end of thi s type of class. PDEs are u s ually presented again in the under gra duat e fluids course in the form of the Na vierStoke s equations, but the emphasis here is placed on the flow phenomena rather than the math ematics ( rightl y so), and the so l ved examples are nearly always one-dimensional problem s that reduce to ODEs. He n ce, for chemical e ngineering s tudent s, the la sti ng impre ss ion of PDE s usually comes from dealing with the un s teady heat equation Whil e the eq uation and it s application are fairly easy to grasp the mathematic s are not ; th e most common introductory so l ution is for a se mi-infinite domain which calls for abstract boundary conditions, a similarity transform a nd an error-function so lution. ( Surely the frequent appearance of erf on comical s tudent-org a ni zat ion Ts hirt s s hould tell u s some thing! ) Whi l e experiencing PDEs in a variety of contexts emphasizes their importance it likewi se see m s to leave the student s so m ew hat un sett led and without a firm foundation for the s ubject. The se considerations h e lped formulate a n approach that was u sed when given the luxury of s pending three or four weeks on the topic of PD Es with se nior-le ve l chemical engineers. At LSU the clas s in which this occurs i s Developm e nt of Mathematical Model s It affords a number of uniqu e opportunities for the instructor. Bein g a math course among other more popular e lecti ves, it attracts so me of the better undergraduate s along with a few Ma s ter s st udent s from our gra duate program a nd from indu stry, a nd the se st udent s have more-or-le ss completed th e 'pri nciple s' classes Also beyond the class's s trong modeling emphasis, the topical coverage is l eft largel y to th e instructor 's di scret ion (see Rice and Do' 7 l for a template of the co ur se offered at LSU) Finally at the point in the se me s ter when PDE s are introduc e d the st udent s hav e them se l ves d erived a number of PDE s governing tran s port and reaction e ngineerin g problems that s ub se quently were reduced to s imp l er form. Hence their foundation for model development i s s trong and they can cl ear ly see the need to mov e b eyo nd algebraic equations and ODEs in order to make full u se of their model s. APPROACH The educational objective i s to a llow the s tud ents to so l ve complex PDE s of real engineering interest, without sac rificin g coverage of fundamental mathematical behavior. To realize thi s goa l an approach is u se d whereby classroom coverage includes topics s uch as the origin of the equations v i s u a lization of the so lution s pace qualitative beh av ior and tie s between the physical phenomena and mathematic s. These topic s are potential concep tual barrier s, and overcoming th em can make the terminology and the detailed mathemat ic s le ss intimidating At the sa m e time to prevent excessive classroom time bein g de vo ted to numeric a l techniques or s peci a li zed softwa re the st udent s are le ft mo st l y on their own to pur s ue numerical so lution s to homew ork or modeling projects Thi s approac h works well (g iven good numerical tool s) because the so lution s that the st ud e nt s h ave themselves worked out are highly effective illu s tration s of conceptual topic s. In the past it would ha ve proved le ss feasible becau se, until the in str uctor covered so lution techniques, the s tudent s lacked the tools to fully exp l ore the mathematical models with which they were working. The intent i s not to de-emphasize the crucial subject of so lvin g PDE s. Rather these issue s are po s tponed until the foundation i s s tronger. If implemented properly, this ap proach s har es many positive attributes of just-in-time l earni n g' emp lo yed by Finlayson: l 8 l modeling project s evolve so th a t just as s tudents id entify the ne ed for n ew mathematical tool s, the rele va nt s ubject s are addressed The benefit of incorpor ati n g sof tware s uch as MATLAB is that topical coverage in the classroom can remain fundamental w ithout s lowing the st udents progress toward qu a ntit a tive so lution s. Spring 2000 The ideas and examples described here were developed as part ofa senior-level math-modeling course at LSU . When beginning the PDE section of the course it was helpful to consider the contexts in which our students ty picall y are introduced to PDEs. Unfortunatel y, these encounters are infrequent and generall y presented as asides to the main material. 14 7

    PAGE 54

    CONCEPTUAL BARRIERS Conceptual barrier s are never clear-cut, and they of course depend on background learning sty le and ability. But we will attempt to generalize a se t of hurdle s that stand between more easily conceptualized phy s ical behavior and the corre sponding mathematic s. An example is the categorization of PDEs; th e mathematical definition s of parabolic versus hy perbolic PDEs probably see m abstract at first but exp l aining the analogy to the synonymous heat and wave equations makes thi s classification more tangible. so lution technique s and the problem s were more often than not restricted to one spatial dimen s ion. From a numerical perspective inno va tive idea s have been pre se nted for pro gramming solutions to PDE s a t an introductory leve] ,l 10 11 1 but the se too are dimensionally restrictive In contrast, mod ern numerical sof tware gives the student more flexibility with re s pect to the type of problem s that can be so lved and the way in which they can be explored. Software such as MATLAB's PDE Toolbox allows two dimensional problems of arbitrary geometry to be set up and solved in a matter of minutes. In MATLAB the system geometry is entered using a graphical user interface (GUI) that strongly resembles familiar drawing programs Equa tions and boundar y conditions are chosen from radio-button menus and clicking with the mou se result s in mesh discretization solution by the finite element method and a wide range of three-dimen s ional color graphical output. Nu merical output is only slightly harder to manipulate, requir ing a review of MATLAB 's built-in pdetool sc ripts The PDE toolbox can be used at an introductory level without extensive knowledge of the ba sic MATLAB software. Hence the time s pent in familiarizing students with the so ftware can be kept relatively s mal I. In 1997 the PDE toolbox was introduced to the s tudent s during a si n g l e This development of strong tie s between analogous physi cal and mathematical behavior provides a basis for breaking down conceptual barriers. Emphasis is placed on the behav ior of general classes of PDE s, with concrete examples u se d to illustrate these tie s. It is hoped that the foundation and comfort level that can be achieved by this approach offsets the risk of se nding the s tudents headlong into numerical solu tions (without much knowledge of the associated techniques). No single list of conceptual barriers i s comprehensive. Table I s how s a li s t of topic s that are addressed se quentially by the author during a three-week sec tion covering PDEs The first couple of points are general, but the latter part of the li s t applies to seco nd-order PDE s, for two reason s. First, many second-order equations can be packaged neatly into elliptic, hyperbolic and parabolic categories, which aids in generalizing behavior. Second the se are the type s of equations amenable to solution in MATLAB 's PDE Toolbox, which was a n essential part of the approach. The in-cla ss overview of seco nd order PDE s was taken largel y from Crandal1 ,l 9 l who doe s an excellent job of tying qualitative behavior of the equations to quantitati ve math ematics ( numerical and analytic). The approach promotes picturing a PDE as a family of s urface s, the correct s urface being pinned down by the appropriate boundary and ini tial conditions. Crandall explains the difference between equilibrium problem s and propagation problem s, which ties in nice l y to a discussion of the characteristic curves for para bolic and hyperbolic equations. COMPUTATIONAL BARRIERS Effective use of a math model re quire s, of course, a so lution In the pa s t a significant time inve s tment was required to introduce analytic 1 48 TABLE 1 Syno p s i s of Co n ceptua l Topics Discussed During Classtime Dur in g the PDE Section of a Modeling Course Topic used to address conceptual barrier Ind epe nd e nt and dependent variab l es Pi c turin g th e so luti on s urfa ce Second-order PDEs Equi librium versus propagation Characterist ic lin es Finite difference methods Stability and num e ric a l diffusion Synopsis o{lhe classroom discu ssio n Use a 'fam iliar eq u a tion to emp ha s i ze independent/dependent and ho w they define a PD E An ODE describes a family of c ur ves; one or more boundary conditions pin down the curve of int e r est. Similarly, many PDEs can be pictured as a s urf ace, pinned a t th e edges. These are categorized as e llipti c parabolic, or h ype rb olic It helps to understand physical beh avior assoc iat ed with eac h in the form of Poisson's eq u at i o n the h ea t eq uation and the wave eq uation Elliptic eq u ations describe equ ilibrium or jury problem s 1 91 -the boundaries (whic h are fixed on a ll s id es) w h olly dictate the s h ape of the int erior. Prop aga ti o n problems ha ve an open b ou ndar y. The s urfa ce evo l ves w ith tim e, pinned on the sides by boundary co nditioin s and at th e front by initial co nditi o n s. Characteristic lin es a re sudden c h anges in the s lope of the s urfa ce ( like creases), brought abo ut by c hange s in boundary cond ition s. Parab o li c equations hav e one c haracteri s ti c line; changes a t the boundary are propagated in s t ant l y but weakly into th e interior. H yper b o li c eq u a ti ons ha ve two c h aracteristic line s. C h anges a t the boundary are propagated int o the interior at full stre n gth but a t a finite spee d. There are analogous c hara cteri s tic lin es for the finite difference method. These p art l y dictate step size in propagation p roblems. When u si n g num erica l so luti ons we must be wa r y of errors inherent to the so lu tion t ec hnique Many of these stab ility issues r e late to th e form of the gove rnin g eq uati on Chemical Engine e ring Education

    PAGE 55

    class session held in the de partment computer lab. In 1999 it was introduced via a step-by-step instruction set handed out to the students, thus requiring no class time. The use of software should not de-emphasize the impor tance of teaching analytic and numerical techniques for solv ing PDEs. Instead it should al low us to embrace the introduc tion of PD Es at the undergradu ate level (if their use enhances the fundamentals being taught) and encourage exploration and critical thinking early on. t 0 5 -0 5 -1 1 5 -1 5 -1 -0 5 0 0 5 1 5 Length domain (boundaries along the centerlines of two neighbor ing fins) or 1/32 of the do main ( one boundary along a fin and one along a fluid line of symmetry) We show the former approach below. For unidirectional flow, the Navier-Stokes equations re duce to 11 31 (1) au 2 p-=G(t)+v' 2 u at where u is the velocity in the direction of flow G(t) is the pressure gradient in this diThree examples are given below These were chosen to illustrate a number of points. First they are indicative of the Figure 1. V e lo c i ty m a p in e ntir e h e atex chan ge r domain tak e n from und e r g raduat e s tud e nt s' s o luti o n. rection, and v'~ is the Laplacian for the two direc tions orthogonal to flow. Hence for a steady flow, G is constant and the equation retypes of problems easily solved by MATLAB (most impor tantly, those with arbitrarily complex boundary geometries). Second, one each of an elliptic parabolic and hyperbolic equation are shown. Third and most important they typify problems where the student clearly understands the engi neering relevance and the manner in which their models can be used in design. This last aspect becomes especially effective when a model can be tied to an experiment as is shown in Example 1 or for instance in reference 12. The MATLAB scripts for these examples can be found on the web at E x ample 1 S tead y Laminar Flo w in a Finned Heat Ex chan g er For parallel laminar flow through a duct the Navier-Stokes equations reduce to Poisson s equation (a scalar equation since there is only one velocity component) This example is for flow in the annulus of a heat exchanger in LSU s ChE measurements laboratory. This example was of particular interest to students in the math modeling class because many of them had used this apparatus for a pressure-drop-versus flowrate experiment in which they calculated the friction factor for the annular region Additionally, the complex ge ometry in the annulus makes the problem an excellent candi date for solution using MATLAB The annular space in the heat exchanger is contained be tween r = 1.05 inches (OD of inner pipe) and r = 2.469 (ID of outer pipe) It contains 16 symmetrically placed fins of width 1/32 inch and length 1/2 inch, as shown in Figure 1. Symme try allows the solution to be performed in either 1/16 of the Spring 2000 duces to Poisson s equation v' ~ u=-G/ (2) Zero-velocity boundary conditions are used along all sur faces of the heat exchanger. Depending on how the FEM domain is chosen line s of symmetry are likely to arise in which case symmetry boundary conditions (n-v'u=0) are also used. In MATLAB the elliptic equation is written as -div[ c grad( u)] +a u = f (3) where div and grad are the divergence and gradient op erators respectively Hence one would specify c=l, a=0, and f=G/ to perform the calculation. The solution using MATLAB s PDE Toolbox involves four steps : I. Map the domain using the GUI ( The geometry can be drawn crudel y u s ing the mou s e and then refined by double-clicking on the various polygon s to type in precise vertex positions ) 2. Select th e governing equation and boundary conditions. The GUI contains a radio-button interface that allows the user to s pecify the type of equations and boundary conditions along with values of parameter s 3 Solve the problem After step two, the solution consists of clicking two buttons: one to generate and refine the FEM mesh and the second to s olve the problem. A wide range of graphical output is available. 4 Quantitative analysi s. Thi s last step requires slightly more user experience since the command-line interface must be used. For instance value s of velocity at each node in the mesh are contained in an array that can be exported to the MATLAB workspace. There are a series of 'pdetool' commands that can then be used to perform interpolation, integration etc. Integration is used to calculate volumetric flowrate a s a function of the pre s sure gradient that was 149

    PAGE 56

    1 5 1 2 5 1 ;;> 0 75 OS ~ C2 0 25 -0 25 1 1 2 5 1 5 1.75 2 2 5 2 5 2 75 32 5 3 5 1.4 1.2 0 8 0 6 0 4 0 2 -0 2 1.5 2 5 3 5 Length(cm ) Figure 2. Geometry of 1116 of flow region in heat exchange r (top); z-direction (perpendicular to page) velocity contours (bottom). 0 4 0 3 0 2 I 0 1 i 0 !:-0 1 0 2 -0 3 240 -0 4 1.2 1 4 1.6 1.8 2 2 2 4 2 6 220 Lenglh(cm) 0 4 200 0 3 0 2 180 I 0 1 160 i 0 .!:--0 1 -0 2 140 -0 3 120 -0 4 1 2 1 4 1 6 1 8 2 2 2 4 2.6 Length (cm) 100 0 4 0 3 BO 0 2 e 0 1 "i 0 !:--0 1 0 2 -0 3 -0 4 1 2 1.4 1 6 1.6 2 2 2 4 2 6 l~(cm) Figure 3. Snapshots of temperature profiles during transient heatin g (t=7 sec, t=2.4 min, and t=l2 min). 150 impo sed during specification of the PDE Figure 2 shows how the annulus geometry (l/16 of the entire domain) is defined u s ing various simple geometric shapes and shows contours of the z-direction velocity (i.e., velocity is perpen dicular to the page) The contours have not bee n assigned numeri cal values in this figure because the velocity depend s on one s specific choice of viscosity and pre ss ure gradient. Their shape remain s fixed however. Integration of the velocity profile to determine flowrate gives a fanning friction factor f=19.5/Re whereas s tudents typically ob tain values between f=20/Re and f=25/Re in the experiment (using an effective area approach). When qua n titative analysis is u se d one can i n troduce the students to issues of numerical accuracy. Grid refinement is trivial in MATLAB, requiring only the click of a button for successive refinements of the FEM mesh. Using various levels of grid refinement in this example (the s tandard mesh followed by two successive refinements ), one obtains the following values for the friction factor: 20.12/Re, 19 .65 /Re and 19 .52/Re. Returning to Figure I one can see the velocity profile for the entire annular region where the lighter s hading indicates higher velocities. While numerical solution over the entire annular region is less efficient than breaking it along line s of symmetry the resulting graphic is more appealing than Figure 2. Figure I was taken directly from two students' homework ;C 14 l they were able to work the problem after only a short tour of MATLAB during one of the class periods. ( The students used inches rather than cm, causing the discrepancy of scale with Figure 2.) Example 2 Trans i ent and Stead y Heat Transfer in a Finned Heat Exchanger While the heat exchanger de sc ribed in Example 1 is not u se d for heat-transfer experiments at LSU the concept of heat transfer from a fin is both important and readily amenable to visua l ization. 0 4 0 3 0 2 240 E 0 1 "t 0 220 .!:--0 1 -0 2 200 -0 3 -0 4 180 1.2 1 4 1.6 1 8 2 2 2 4 2 6 Length(cm) 160 0 4 140 0 3 0 2 120 E 0 1 "i 0 100 ~-0.1 -0 2 80 -0 3 -0 4 1 2 1.4 1 6 1.8 2 2 2 4 2 6 Length (cm) Figure 4. Steady temp e rature profiles for high (top) and low (bottom) fin-side heat transfer coefficien t s Chemical Engineering Education

    PAGE 57

    The example s h own here can illu s trate the unsteady evolu tion of a temperature profile as we ll as the steady-state temperature profile for various heat-transfer coefficients The transient problem underscores the behavior of parabolic PDEs while the s teady problem help s to illustrate engi neering fundamentals : the conditions under which fins are useful for increa s ing heat-transfer efficiency Parameters are taken from Example 27-3 in Bennett and Myers,l1 5 1 and the fin dimen sions are based roughly on the elliptic examp l e above. The ge ometry is made somewhat more complex to illustrate the versa tility ofMATLAB's GUI SURFACE HIGH DENSITY REGION can produce wave-like behavior MATLAB ( to the author's knowledge) i s not equipped to handle convective transport. One can of course s imulate vibration problems or certain problems involving s ound waves Instead, the examp l e shown here i s a highly s implified illustration of seismic exploration It was cho s en because of its intuitive ap peal to an engineer of any discipline. Artificial seismic waves are Figure 3 shows the fin at t=7 sec, t=2.4 min and t=12 Figure 5. G e om e tr y u se d f o r w a ve pr o pagati o n exa mpl e used in oil exploration or geo physical a n alysis to map s ub surface structure The wave so u rce may be eit h er on the surface or lowered into a we ll and the responses to the wave a t various detector l ocatio n s are interpreted to give the mapping On a simplis ti c min The s hading show s the evolution of the temperature profile in time (an effect that is much more dramatic in color) Figure 4 s hows two steady-state profiles for a hi gh externa l heat-trans fer coefficie nt [1500 Btu/(hr ft F ) ] and a low externa l h eat -t ran s fer coefficient [1 Btu/(hr-ftF)]. The graphics shown here help students conceptualize the con ditions und er which fins can significan tl y increase heat transfer. We make two final points. First the evolution of the temperature profile (es pecially in color) is helpful for under scoring the behavior of a parabolic PDE : a cha n ge in the boundary condition h a s immediate but weak influence through out the fin and the temperature evolu tion is smooth Second one can easily envision numerous exercises that cou ld be performed to illustrate important be havior. For instance a spatially varying h eat-transfer coefficient (discus s ed in Bennett and Myers) is impossible to im pose for even simp l e analytic solutions but can easily be incorporated into the MATLAB so luti on Example3 Wave Propagation in a Heterogeneous Material Applicatio n s of the basic wave eq u tions are l ess frequent in c h emical engi neering. While s tron g convection effec t s Sprin g 2000 Figure 6 W ave pr o pa g ation a nd r e fl ec tion in a h e tero ge n eo us d o main Whit e co rr e sponds t o hi g h e r-pr e s s ur e tran s i e nt s level the propagation of pres s ure waves in the ground is described in the wave equation 11 6 1 V 2 p __!_ a 2 p = 0 c 2 at 2 where c is the wave ve lo city depen dent primarily on the material proper ties. A time-dependent pressure must be defined at the wave source. Along reflective boundarie s of the domain, one s pecifies n Vp = O. Figure 5 s hows the geometry used for thi s s imple example. The interest ing features are the slope of the l ower boundary ( which could be interpreted as a geologic bedding plane) and the inclusion of a material heterogeneity at the lower right. These two fea tures make the response more inter e s ting, but preclude solution by ana lytical means To s olve the problem a pressure spike was induced ( via a rapid l y decaying exponential function) a lon g the top boundary at t=0. Figure 6 is a qualita tive illustration of the resulting wave's behavior. The lighter sha din g (w hi c h represents the traveling high-pressure front) propagates downward reflects off of the bottom boundary and then return s to the s urface (where it wo uld be detected). Although length and time s cale s are not included si n ce the ex1 51

    PAGE 58

    ample is qualitative, one can see the influence of the lower density inclusion where the wave gets held up and one can envision how the surface response allows a map of the subsurface to be generated. From a mathematical viewpoint, this example helps illustrate typical hyperbolic behavior: the response of the interior to a change at the boundary is delayed, but then felt at full strength once the wave reaches a given point. Using this example in the cla s sroom an otherwise dry discussion of characteristic lines for a hyperbolic equation can become more captivating CONCLUSIONS The use of PDEs in the undergraduate curriculum often has mixed results: Important topics cannot be modeled with out PDEs. On the other hand, the simplicity of so l ution do mains for analytic problems often makes for abstract relation ships to real engineering problems, and the mathematical de tails of an analytic solution can distance students from the original objectives. This paper presents effective uses of modem numerica l software for solving real engineering problems at the under graduate level, which i s an increa s ingly popular approach among chemical engineering educators. The quick learning c u rve for certain numerical software allows students to be gin exploring a model's behavior almost immediately Class room time can then be used to break down conceptual barriers as s ociated with PDEs. It i s hoped that thi s ap proach lay s a better foundation and better prepares stu dents for later material on solution techniques either analytical or numerical. ACKNOWLEDGEMENTS The author would like to acknowledge Charles Varnado for providing the numerical code used to generate Figure 1 and to acknowledge the anonymous reviewers for very in sightful comments on the original manuscript. REFERENCES 1. Harb J N. A Jones R.L Rowley and W.V. Wilding, "Use of Computational Tools in Engineering Education ," Ch e m. Eng. Ed. 30 ( 3 ), 145 ( 1997 ) 2 Mackenzie, J.G and M. Allen Mathematical Power Tools: Maple Mathematica MATLAB and Excel ," Chem. En g Ed. 32 ( 2 ), 156 ( 1998 ) 3 Taylor, R., and K. Atherley Chemical Engineering with Maple Ch e m Eng Ed ., 29 ( 1 ), 56 ( 1995 ) 4. Cutlip, M B. and M. Shacham Probl e m Solving in Ch e mi c al Engineering with Numerical M e thods Prentice Hall ( 1999 ) 5. Sinclair J.L CFD Ca s e Studie s in Fluid-Particle Flow ," Ch e m Eng Ed ., 32 ( 2 ), 108 ( 1998 ) 6. Chapra, S.C ., and R.P. Canale, Num e ri c al M e thods for En gin ee r s 3rd ed. McGraw-Hill ( 1998 ) 7. Rice R.G. and D.D Do Appli e d Math e matics and Mod e l ing for Ch e mical Engin ee r s Wiley ( 1995 ) 8. Finlayson, B.A. Probl e m-Centered Course in Using Multi media ," paper 3513-02 presented at the ASEE Annual Con/52 ference and Exposition Washington DC ( 1996 ) 9 Crandall S.H. Engin ee ring Anal y sis: A Surv ey of Num e ri c al Pro ce dur es, McGraw-Hill ( 1956 ) 10 Wiggins E.G. "Computational Fluid Dynamics on a Spread sheet ," Comp in Ed J ., 7 ( 2 ) 7 ( 1997 ) 11. Cutlip M B ., and M Shacham, The Numerical Method of Lines for Partial Differential Equations ," CACHE N e w s, 47 18 Fall ( 1998 ) 12 Anklam, M R. R.K. Prud homme B.A. Finlayson, Ion Ex change Chromatography Laborator y : Experimentation and Numerical Modeling ," Ch e m. Eng Ed. ,, 31 ( 1 ), 26 ( 1997 ) 13. Leal L.G., Laminar Flow and Conv e ctiv e Transport Pro c e s se s : Scaling Principles and As y mptoti c Analysis, Butterworth-Heinemann ( 1992 ) 14. Varnado, C and K. Loo Math Modeling Project for ChE 4296 LSU ( 1997 ) 15 Bennett, C O ., and J.E M y ers, Mom e ntum H e at and Mass Transf e r 3rd ed ., McGraw-Hill ( 1982 ) 16 B e rkhout A.J ., Appli e d S e ismi c Wav e Th e ory Elsevier ( 1987 ) 0 RA N DOM THO U GHTS Continu e d from pa ge 145 The more types of assessment data collected for a specific component (column of the matrix) the more reliable valid and fair the evaluation of that component. For explanatory notes and literature citations on the different assessment tools see Reference 3 H ow might the scholarship of teaching be inclu d ed in tenure and pr omotion d ecisions? Many academic institutions have begun to acknowledge the scholarship of teaching as a valid component of tenure and promotion (TIP) applications An approach being taken by s everal of the s e in s titutions is to allow faculty members to allocate variable percentages of their total effort to teach ing research, and service, with minimum percentages being specified for each area If more than a certain percentage is allocated to teaching educational scholarship must be in cluded in the faculty member's activities and a teaching portfolio containing a subset of the items in Table I must be included in the TIP dossier. A review committee a s signs separate numerical performance ratings to each of the three areas and weights the ratings by the specified percentages to calculate a composite rating which provides the basis for the deci s ion on tenure or promotion. For ratings of the scholarship of teaching to be reliable and valid the evaluating department should take the following steps: I] F o rmulat e and ann o un ce a n a ssess m e nt and ev aluation plan Decide which items li s ted in Table 1 will be collected in the teaching portfolio taking into account both in s titutional guidelines and con s ideration s specific to the department. Choo s e a system to rate each of the items in the portfolio (e g., rate e ach item on a s cale from O to 10 ), weighting factors for each item a nd weighted score s that s erve a s criteria for adC h e mi ca l Engin ee rin g Edu c ation

    PAGE 59

    equate and superior scholarship. D esc ribe the rating sys tem to all departmental faculty member s who ma y wish to in c lude educational scholarship in their credentials and di s pla y s ev eral examples of excellent portfolio s as models ll Provide training t o portfolio rat ers. Give detailed explana tions of the evaluation criteria to faculty members who will be serving as raters and provide guided practice on sample port folios. ll Collect at l east two indep e nd e nt ratin gs of each portfolio submitted and ha ve the ev aluator s re co n c ile their ratin gs to arrive at a consensus rating. Incorporate the consensus rating into the overall tenure/promotion do ss ier evaluation pro cess. REFERENCES 1. Boyer E. Scholarship Reconsid e r ed: Prioriti es of the Pro fessoriate, Carnegie Foundation for the Advanc eme nt of Teaching Princeton, NJ ( 1990 ) 2. Glassick C E. M T. Huber, and G.I. Maeroff Scholarship Assess e d: E va luation of th e Prof ess oriate Jossey-Bass San Francisco ( 1997 ) 3. Felder R.M A Rugarcia and J.E. Stice The Futur e of Engineering Education V. Assessing Teaching Effective ness and Educational Scholarship," Ch e m Eng E d., in press. 0 E D U CATOR: REKL A ITI S Continued from pag e IO I review articles and has received many invitations as an in vited lecturer. He supports his research through a number of NSF grants and an industrial/univer s ity consortium, CIP AC (Computer Integrated Process Operation s Center) at Purdue. He was one of the founders of CIP AC and served as its director until this year when Profes so r Gavin Sinclair took over. In 1984 he won the Computing in Chemical Engineer ing Award of the Computing and Sy s tem s Technology Divi s ion of AIChE. He was again recognized by AIChE for hi s accomplishments when he wa s named a fellow in 1994. In that same year he also won the best paper award from Com puters & Chemical Engine e rin g for the paper by Jayakumar and Reklaitis Chemical Plant Layout Via Graph Partition ing : Part l. Single Level (Comp. & Chem. En., 33 441 1994). 1994 was a very good year for Rex since that year he also won the ASEE Chemical Engineering Division lecture ship award His award address on Computer-Aided Design and Operation of Batch Processe s" can be found in CEE, 29 76 (1995) Of course, much of the real work of re searc h i s done b y graduate students. Rex has been advisor or co-advisor for 28 PhD students and 37 MS students. He currently advises or coadvises eight PhD students and one MS student. Seven of his past students are now professors: A. Elkamel (Kuwait University) Carl Knopf (Louisiana State Univer s ity) I. Karimi (National University of Singapore) B S Lee (Pukyong National University) E.S. Lee ( Dongguk Univer sity Korea) I.B Lee ( Pohang University Korea ), and G. Yi Spring 2000 (Kynghee University, Korea ). Rex ha s also proved him se lf to be a good citizen of the chemical e ngineering community. He ha s serve d as secre tary vice-president and pr es ident of CACHE and continues to se rve as a trustee of that organization Equally active in AIChE particularly in the Computing and Systems Technol ogy ( CAST) division (in which he has held all of the of fice s), he was an elected director of AIChE until December of 1999 He has also been an active member of the Council for Chemi ca l Re searc h a nd h as serve d on its governing board. He ha s co-edited severa l volumes of conference pro ceedings dealing with proce ss design simulation, computer graphics and optimization He has been Editor-in-Chief of Computers & Chemi ca l Engineering since 1994 and was Co-Editor-in-Chief for the eight years before that. In hi s research and profe ss ional activities Rex has alway s been a good team pla ye r. He has collaborated with a number of past a nd current faculty at Purdue on papers including Paul Andersen, Gary Blau Frank Doyle Lowell Koppel Martin Oko s, Joe Pekn y, Dan Schneider Bob Squires Venkat Venkatasubramanian and Jack Woods. He ha s a l so collabo rated on papers and edited proceedings with a number of well-known chemical engineering professor s from other schools including Larry Biegler Brice Carnahan, James F. Davi s, Tom Edgar Ignacio Grossmann Dave Himmelblau, Richard Mah Da v id Rippin John Seader Jeff Siirola, Aydin Sunol and Doug Wilde While in vo l ve d in this wo rk at Purdue Rex continues to li ve the good life in We s t Lafayette, helping raise his two so n s and being in Janine 's words "a wonderful father" who take s th e role very seriously-when they were much younger, he would tell the boy s wonderful stories of knights and pirate s, a nd take them fishing or sailing, and for family favorite s ki trip s in Colorado George earned his bachelor's de g ree in history from Purdue his master 's from Wake Forest and is now working on his PhD in hi s tory at North eastern University in Bo s ton Victor i s a junior in electrical engineering at Stanford University. Sailing ha s remained an important part of Rex 's life. For s ixteen ye ar s he participated in the famed Chicago-to Mackinac race and even won his class a few times in a 34foot I s lander-but most of the time he finished in the middle of the pack. His family who did not race with him, would watch the boat leave Chicago and then drive to Mackinac I s land to watch the boat s arrive there. They did however enjoy sai ling with him in hi s 19-foot Lightening In addition to being close e nough to sail on Lake Michigan the Lafayette area is near enough to visit family in Chicago for the holidays Rex remain s close to his mother and sister and has b eco me close to Janine 's large rowdy extended family. We are proud to have Rex Reklaiti s at Purdue. He is an hone s t ge nerous witty colleague with high s tandards Fade out to the tune of The Wabash Far Far Away. 0 1 53

    PAGE 60

    t3ij class and home problems ) The object of this column is to enhance our readers' collections of interesting and novel problems in chemical engineering. Problems of the type that can be u sed to motivate the student by presenting a particular principle in cla ss, or in a new light or that can be ass igned as a novel home problem are requested as well as tho se that are more traditional in nature and that elucidate difficult concepts. Manuscripts s hould not exceed ten double-spaced page s if possible and should be accompanied by the originals of any figures or photographs. Please submit them to Professor James 0. Wilkes (e-mail:, Chemical Engineering Depart ment University of Michigan Ann Arbor MI 48109-2136. AN ''OPEN-ENDED ESTIMATION'' DESIGN PROJECT FOR THERMODYNAMICS STUDENTS STEPHEN J. LOMBARDO University of Missouri Columbia, MO 65211 T he project is "open-ended" when the students ask if only one design i s right and an estimation will s uffice when they cannot find an exact value; the descriptive title for the project intentionally reinforces the expectation of this assignment for the student.The project is a preliminary design to evaluate methods of generating elec tricity for a resort island As see n in the accompanying pres s release (Ta ble 1) written in the form of an advertisement for tourism a tropical island abundant in leisure activities i s described; thinly veiled within the ad are various re so urce s that could be used to produce electrical energy. This design effort is assigned in a first course in thermody namic s offered by the Department of Chemical Engineering at the University of Missouri at Columbia All the material typically contained in a "c lassical thermodynamics course (first law seco nd law and power cycles) is covered in this 15-week semester-long course, and most of the st ud ents are first semester Juniors having recently comp l eted an intro ductory chemical engineering course covering materi a l and energy balance s. Although the students are s till in the begin ning stages of their chemical engineering coursework this project meets the need of providing some early design op portunity in the curricu lum The project also incorporates role playing and decision making two important elements of active learning r 11 and critical thinking .r2 1 PROJECT ORGANIZATION The overall objective of thi s preliminary de sig n is to rec ommend methods of energy generation that meet the elec154 tricity needs of the island, with some assessment of environ mental and safety concerns. The methods proposed by s tu dents to date are li sted in Table 2. Organization of the project is shown sc hematically in Fig ure 1, and the project deliverables are divided into two parts: A Pha se I written report and a Phase II written report fo lowed by an oral pre se ntation. For the Phase I report the st udent s, grouped into teams of three or four, propose meth ods for producing e lectricity for which they mu s t provide brief description s along with environmental and safety con siderations. Each group must also provide block diagrams showing the relation s hip between the resource and the en ergy output. The pedagogical role of the block diagram is to force s tudent s to contemplate the physical layout of each energy-generation method. For hydroelectric energy, for example, Figure 2 illustrates the relationship between the reservoir t h e turbine, and the generator. The groups typi cally propose one or two method s per team member and each method requires one to two p ages to describe Stephen J. Lombardo received his BS de gree from Worcester Polytechnic Institute and his PhD from the University of California, Ber keley both in chemical engineering. He worked for seven years in industry in the areas of ceramic materials and ceramic processing be fore joining the Department of Chemical Engi neering at the University of Missouri-Colum bia in 1997 Copyright ChE Division of ASEE 2000 Chemical Engineering Education

    PAGE 61

    TABLE 1 Pre ss Relea se for the Open-Ended Estimation De s ign Project Th e island of ____ is a J OO-square-mile id y lli c s p o t l oca t e d ____ Clima t e, scenery, flora and fauna all co mbin e to mak e this a pr e mier vacation spo t for families, co upl es, and thos e just l ook ing t o get awa y fr o m it all. Abundant r ec r ea ti o n ac ti v iti es are available as well as s p ec ial s i g ht see in g op p or tuniti es Th e prime attrac ti o ns of this island are the mil es of sce ni c beaches w i t h soft wh it e sand Br e ak e rs p o und th e s h o r e c ontinu o usl y, makin g it a ye ar round hav e n for surfers. A special lo c al e is the c apital port lo c ated on th e so uth west c orner of th e island. Com e spend a lan g uid aft e rnoon v i s iting this spot and observe th e spectac ular tidal drop of 40 feet. At l ow tid e, it is a seashe llco ll ec t o r 's paradis e! Don t forg e t th e m o untains h e r e. Active vo l c an oes d o t th e panoramic vistas and Mount Sim o n e rumbl es at l eas t twice a day sp ra y in g a co ntinuous la zy plum e of h o t gases into th e clear blue sky. F or th e intr e pid g uided tour s are mad e to th e rim durin g qui e t e r p e ri o d s. Come see th e m os t spec t acular la v a formations o n Earth Hik e rs c an tak e trip s d ee p into th e h e art of th e i s land, w h e r e 8 00-f oo t w at e tfalls c a sca d e d ow n into r eflec tin g pools F o r th ose e nj oy in g l ess s tr e nuous exe rtions co m e s p e nd th e afte rnoon in th e h o t s prin gs that co ll ec t into natural r oc k bathin g po o l s. F o r th e co n ve nti o nal indi v idual we ha ve fr e sh-water baths ; fo r tho se lookin g for someth in g m o r e exo ti c, we ha ve 2 % natural salt baths -g uarant ee d to improv e yo ur skin ton e and l eave yo u fee lin g h ea lth y. After bathing co m e and e nj oy a massa ge at th e skilled hand s of o ur staff m e mb e r s A nd don t for ge t th e cleansin g effec t s of a mud bath pro v id e l o n gla s tin g th e rap e uti c val u e to yo ur skin. H ang g lid e r s w ill also fi nd plenty of r ec r ea ti o nal o pp o rtunity Th e 300-Joot l ove l y cliffs over l ooking th e ocea n afford prim e laun c h l oca ti o n s Th e island i s r e n ow n ed fo r it s s t ea d y coo l bree ze s that pr ov id e yea r-round han g g lidin g o pp o rtuniti es. And don t worry abo ut bad weather when yo u co m e to v i s it. The sun s hin es 80 % of th e tim e h e r e, with a ye ar round m e an t e mp e rature of 85 F and l ow humidi ty. Rain w h e n it occurs is a lat e-eve nin g eve nt so day plans are n eve r ruin e d Th e i s land i s also abundant with natural r eso ur ces. F o r es ts of hard woo d tr ees cove r over half of th e i s land and co ntain so m e of th e wo rld 's m os t exo ti c species of animals and birds. C o / 0 1 ful and extrem e l y rare s p ec i es of birds ca n o nl y b e fo und h e r e in th e tropi c al fo r es ts Spelunkers w ill also fi nd pl e n ty to do. Mil e -l o n g caves pro v id e an o pp o rtuni ty for all l eve ls of c av e ex pl o rati o n from n ov i ce to ex p e rt Come d ow n and ex pl o r e o ur known caves, o r di scove r n ew ones of yo ur ow n Ma y b e yo u w ill ge t lu cky and find ve ins of pr ec i o us m e tals ; lar ge co al d e p os it s ha ve already b ee n di scove r e d Don t d e la y-co ntact y our tra ve l a ge nt n ow. Distr i bution of Project Description and Press Release (week 7) Preliminary Brainstorming by Teams Th e s econd and main requirement of the Phase I report i s incl udin g a design e qu ation for each method. The l efth a nd s ide of the equa ti on i s in term s of e n ergy or power, a nd the right-hand s id e is in term s of the fu e l or resource a nd ot h er parameters that will allow for the determination of the a m o unt of energy or power that ca n b e o btain ed. No calc ul at i o n s are made h owever, for the Ph ase I report, since the re so ur ces are not we ll defined a t this point in the project. Rather the calc ul atio n s a nd final reco mm e nd a ti o n are p e rform e d for the Ph ase II s ubm issio n. Literature Review by Teams Meeting with Boss (week 10) Submission of Phase I Report (week 11) Return Phase I Report with Feedback (week 12) Meet with Development Authority (week 13) Submission o f Phase II Report with Revised Phase I Report (week 14) Oral Presentat i ons with Peer Review (week 15) Figure 1. Organization and tim e lin e of the open-ended est imation design project in thermodynamics. Spring 2000 Separating the development of th e de s ign eq u ations in Ph ase I from the exec ution of th e calc ul at i o n s in Phase II i s int e nt ional. Students often fi nd a n eq u a ti o n and u se it without critica l assessment. By delaying the a pplication of the eq u ation from the discovery o r derivation so me time wi ll hopefull y be s pent on critically evaluating the u se fulne ss and correctness of the proposed relationship The benefit of assessing the design e qu at i o n b ecomes a pp are nt to the s tudents when for the Ph ase II report they h ave n o t acquired all the va lu es they n eed to co mpl ete the calc ul ations In the case of wi nd e n ergy for examp l e a typical textbook d esign eq u a ti o n 131 for the power P i s TABLE2 Energy Methods Proposed by Students H ydroe l ec tri c Wind Geothermal Hot Lava Tid a l Ocean Thermal Fuel Ceil Fossil Fuel Solar (Photovoltaic ) Hot Springs Nuclear Wave Bi omass Co n version Solar (Therma l ) ( 1 ) where v i s the wi nd velocity and D is the bl ade diam eter of the wi nd turbine The density of air p a nd efficie nc y Tl are Reserv o ir Flow of water Water Outlet Work Electricity ~--.W Generator Figure 2. Example of block diagram submitted b y a student gro up for h y droelectri c e nergy. 15 5

    PAGE 62

    often lumped into an overall coefficient, k. The co rrectne ss of the dependence of th e power on the diameter D 2 and ve locit y, v 3 can quickl y be estab li s hed b y observing that th e power from wind is related to the rate of kinetic e nergy change by m~v 2 / 2 and the ma ss flow rate, m i s equal to pv1tD 2 / 4 The equation for wind energy further illu s trate s when quantitie s can be cla ss ified as known easily estimated, or unknown a nd thus mu s t be det er mined by other mean s The d e n sity of air, for example, can be eas il y obtained from the ideal gas l aw, and it is not crucial, for an es timate if th e mean temperature i s 65 F or 85 F. The value for the diam e ter of the windmill, on the other hand i s not nec essar ily so methin g that can b e es timated but literature so urce s will typicall y mention a range of diameter s evaluated in actual tests. Finally it i s impo ss ible to estimate the average wind velocity of a location with a n y certainty; s pecific va lue s pertain to s p ec ific loc a tion s Wind energy also provide s a n excellent opportunity to reinforce the u sefu ln ess of thermodynamic analysis for cal culating e ner gy tran sfo rmation via the first l aw. On e s tudent for ex ampl e, s tated that the equation for wind power is not really an outcome of th er mod y nami c analysis W e then jointly proceeded to evaluate the case of how the kinetic e nerg y change of a gas flowin g in a pip e can be related to th e s haft work extracted; b y exte n sion, wind e n e r gy is then linked to the conversion of kin et i c energy in the pipe swe pt out b y the blad es of the wind turbine Since th e wind velocity i s also not zero after passing b etween the blade s of the wind mill, thi s p oi nt offers partial insight into why the efficie nc y of wind turbines i s low .13 1 Other m et hod s o f ge n era tin g e n ergyl 3 JOJ h ave in their d s i g n equations a si milarit y th at is of in str uc tio nal val u e for reinforcing the utility of the seco nd law Burning of fossil fuel s nuclear power, ocean thermal energy, and geothermal e n e r gy can a ll have their design eq uation s represented b y th e ge neral form (2) where fl i s the ma x imum efficie n cy, m i s th e ma ss co s umption rate of th e fuel and ~H is the enthalpy change of the fuel. By casting th e design e quation s in this common format the m ax imum efficiency of eac h m e thod can also be compared on a common ba s i s namely, the a b so lut e tem perature of the hot reservoir, Th and cold reservoir, T c from the Carnot e quation: fl= 1-T c /Th. Whil e so me gro up s choose to u se the C a rnot efficiency other teams opt to u se a n empirical va lu e for fl from refer e n ce works. Prior to s ubmi ss ion of the Pha se I report the students have a n opportunity to sc h e dul e a meeting with their bo ss i.e., the in s tru ctor. Since s tudent s seem to crave real world" experience role playin g with th e ir bo ss provide s a goo d opportunity to ge t so m e B efore th e me e tin g the s tudent s are 156 remind e d th at their bo ss does not know the so lution to this project. If thi s were the case, the project would not have be e n assigned. Furthermore, th eir bo ss d oes not want to so l ve th e probl e m for them; that is the res pon sibility of the em pl oyee. The bo ss will how ever, g uide encourage, dir ec t e tc. After s ubmi ssio n and acce ptanc e of the Pha se I report the team s fly" to the i s land and meet w ith th e director of the local development authority (the in s tructor ), who trie s to answer any que s tions rel a ted s pecifi ca lly to th e resources of the i s land. It i s at this me eti n g that l ocat ionspecific informa tion s uch as wind ve locit y, available tidal ba sin, hydro e l ec tric dam hei g ht a nd water vo lum etr ic flow rate are made available to the st udent sbut only in response to their s p cific qu es tion s. The link b etwee n the design eq u a tion s and known/unknown inform a tion then be co mes clear Values for each and every variable in the de s i g n equations mu s t be obtained or estimated in order to b e a ble to co mpute th e pow e r. Each gro up i s a l so g iven individuali ze d informa tion-no two gro up s have the sa m e a mount of resources or population. In addition, n o t a ll of the resources provided are s uffici e nt to m eet the e n ergy need s. By tailoring th e re so urc es in thi s m a nner gro up s cannot co nv erge on the same answer and mu st be comfortable (or un comforta bl e) with th eir own decisions. During the m ee ting with the local development authority, the val ue s of th e resources requested b y th e s tudent s are r eco rded for l ater u se in gra din g the reports. The projects are evaluated for a number of attri bute s. For th e Pha se I r e port the d es i g n equations mu s t be so und a nd the l ayo ut s of the block diagrams logical Second, the over all organization of th e report a nd writi n g are assesse d Gen era lly the Ph ase I reports ha ve not b ee n graded, but rather feedback for improvement is pro v id ed ( their bo ss g i ves feed back not grades) The most common s hort coming of the Pha se I r e port s is a poorl y for mulated design equation. It is either not in term s of a s ingle equation, not in terms of the prim ary fuel or resource (e g for geo thermal e nerg y the power is formu l ated in term s of the working fluid in s tead of in term s of the h eat available from the hot re servo ir ), or formu l ate d in a mann er for which not all values are easily obtainable Th e format for the Ph ase II report is a oneor two-page executive s ummar y containing a spec ific recommendation of how to m ee t the energy need s of the i s land The revi se d de s i g n equations are a l so presented along wit h the val u es u se d to compute the pow er. The value s of the energy mu st be roughly correct in light of the fuel or re so urc es s pecifi e d. Sinc e eac h gro up h as b ee n g i ve n different va lu es for the resources I calculate the a vai l a bl e energy for the methods of each gro up in a s pr ea d s h eet application. In thi s manner it is r e l a tiv e ly straig htforw ard to determin e if each gro up h as correct l y calc ul ate d the right order of ma g nitud e of energy po ssi ble from eac h propo se d method, a nd it is at this l eve l ( namel y, the order of ma g nitude ) th a t the correctness of the calc ul at ion s is assessed. The revised Ph ase I r eport is r es ub C h e mi ca l En g in ee rin g Edu c at io n

    PAGE 63

    mitted as an appendix, and this revision is evaluated for improvements over the original version TA BL E3 St ud ent Fee d back to engineering economics, this desire to perform eco nomic ana l yses seems am bitious on their part. (N umb e r s in parentheses indicate th e number of tim es s imilar types of comme lll s we r e mad e i11 a class of24 s tud e 11t s, 1 8 ofw /zi c /z r es po11ded.) Positive Aspects Ne~ative Aspects Su~~estio11s Finally, the soundness of each group's recommenda tion is strongly weighed. Since at this stage of the design no method of en ergy generation i s strictly right or wrong as long as sufficient energy to meet the requirements is pro vided, the recommendation Oral pre se nt ations (7) Oral presentations o n the la s t d ay of clas s (2) Give m o re s pecific guide line s for report s ( 5 ) SUMMARY An open-ended estima tion design project has been developed for use in a first thermodynamics course Opene nd e d n a ture (5) Oral presentations (I ) R eq uire economic estimates (3) Computationally easy ( 1 ) P ro bl em i s t oo easy ( I ) Give more tim e for project (I) Opportunity t o u se people ski ll s (I) Rol e playing ( 1 ) R esearc ho ri e nt e d project ( I ) of each group reflects its evaluation as to what is most acceptable for a resort island and balances what it sees as the competing needs These are most often a method that is environmentally friendly and aesthetically pleasing Some groups recommend fossil-fuel power plants as being a tried and true technology whereas other groups select several methods in combination. The mo s t common recommen dation to date has been the use of hydroelectric energy The final part of the project is a ten-minute oral presenta tion by each team for peer review. The three grading criteria are organization of the talk, clarity of presentation, and sound nes s of the recommendation. The grading process for the oral presentation is conducted in a manner to give st udent s exposure to an assessment practice used in industry A five point scale is used with three as the average, and the st dents are told that the average of the grades that they award must come close to this average. This is similar to many of the merit raise systems employed in industry where the available "pool" is a fixed percentage of the total salary bud get. Use of this method gives the students an opportunity to weigh the consequences of positive and negative assessment and combats any tendency for grade inflation by the teams. The comments of the students regarding this project were solicited with an anonymous questionnaire and the results are summarized in Table 3 The most often cited positive aspects of the project were the oral presentations and the open-ended nature of the project. The most negative com ments were also related to the oral presentations, two of which can be remedied by better scheduling The most frequently cited suggestion by the students per tains to having more specific guidelines for the project. In the experience of the author from seven years in industry expectations and formats for project presentations and results were limited or non-existent. In general, employees were ex pected to develop their own presentation formats The second most frequently cited suggestion pertains to their wish to per form economic estimates. Since this course is offered rela tively early in the chemical engineering curriculum, well be fore their formal design courses with accompanying exposure Spring 2000 taught early in the chemi cal engineering curricu l um. The project, which contains a number of components that simulate real-world experience, is designed to empha size simplification over complexity. A key aspect of the project is decision making; after the values of power for t h e various methods are calculated, a decision must be made, namely, what ultimately to propose. It is this aspect of the project that is most in line with the expectation of problem solving in industry. The project is also structured to embody many of the elements of critical thinking in that for each team to make a final recommendation, the "s tudents are active, involved, consulting, and arguing with each other..."l 2 J Furthermore, several aspects of active learning such as role playing writing, and pre se nting are used to enhance under standing 111 and retention. 1111 Finally the project exposes the s tudent s to the importance, breadth, and complexity of the numerous issues surrounding the generation of energy. REFERENCES 1. Bonwell Charles C ., and James E. Eison Active Learning: 2. Creating Excitement in the Classroom, ASHE-ERIC Higher Education Reports Washington, DC ( 1991 ) Kurfi ss, Joanne G. Critical Thinking: Theory, Res earch, Pra c t ice, and Pos sibilities, ASHE-ERIC Higher Education Reports, Washington, DC ( 1988 ) 3. McRae, Alexander Janice L Douglas and Howard Rowland, eds., Th e Energy Sour ce book, Aspen Systems Corporation, Germantown, MD ( 1977 ) 4. Kraushaar, Jack J ., and Robert A. Risten Energy and Prob l e ms ofa Technical Soci e ty, John Wiley & Sons New York, NY ( 1988 ) 5. Hunt V. Daniel Handbook of En ergy Technology: Trends and P ers p ectives, Van Nostrand Reinhold Co New York, NY (1982 ) 6 Bisio Attilio, and Sharon Boots eds., Encyclopedia of En e rgy Technology and the Environment John Wiley & Sons New York, NY ( 1995 ) 7 Parker, Sybil P ., ed. McGraw-Hill Encyclopedia of Energy McGraw-Hill New York NY ( 1981 ) 8 Considine, Douglas M ed., En e rgy T echnology Handbook McGraw-Hill, New York NY ( 1977 ) 9. Howes, Ruth and Anthony Fainberg, eds., The Energy Sourcebook: A Guide to Technology, R eso urces, and Policy, American Institute of Physics ( 1991 ) 10 Kreider, Jan F ., and Frank Kreith eds., Solar Energy Hand book McGraw-Hill, New York, NY ( 1981 ) 11. Souza, D A., How th e Brain Learns, National Association of Secondary School Principal s, Reston, VA ( 1995 ) 0 157

    PAGE 64

    .ta .. 6_._c_l_a_s_s_r,_o_o_m ________ ) LOW-COST MASS TRANSFER EXPERIMENTS Part 6. Determination of Vapor Diffusion Coefficient I. NIRDO SH, L.J. GARRED, AND M.H.I. BAIRD Lakehead University Thunder Bay Ontario Canada P7B 5El M olecular diffu sio n determine s the rate of most mass-transfer operations. Determination of the dif fusion coefficient of the key component i s very important for predicting rates of mass transfer and many correlations are reported in the literature for binary and multicomponent systems 1 1 -4 1 Thi s p a per describes a si mple experimental technique for determining binary diffu s ion coefficients for vapor (A)-gas (B) systems in which the vapor is generated by the evapora tion of a pure volatile liquid and the gas is air. The theory i s described in many textbook s on ma ss transfer and unit op erations,1 2-41 and only a brief treatment i s given below for immediate reference. From Fick 's first law of diffusion for the case of stag n ant B in a binary sys tem the flux of A at steady state (NA ) is given by 121 N DABPt ( ) Az RT PA 1 P A 2 ZPBM (I) where D A B is the diffusion coefficient, R is the ideal gas constant, T is the absolute temperature z is the length of the diffusion path p 1 is the total pres s ure P Ai and P A 2 are the partial pressures of component A at the two extremes of the diffusion path, and PsM is the logarithmic mean of the lnder Nlrdosh received his BSc and MSc in chemical engineering from Panjab Univ ersity (India) and his PhD from Birmingham Uni versity (United Kingdom). He joined Lakehead Univer sity in 1981 and his re search interests are in the fields of mineral processing and electrochemical engi neering. Laurie J Garred is Pro fessor of Chemical Engi neering at Lakehead Uni versity. He received his BASc from the University of Toronto in engineering science and his PhD in chemical engineering from the Uni versity of Minnesota His research interests in biomedical engineering fo cus on mathematical mod liquid level at time t = O liquid level at time t = t TT z t Figure 1. The evaporat ion tube. partial pressure s of component B at the two ends of the diffu s ion path Let u s suppose we have a liquid in a tube (see Figure 1) of cross-section a, and in time dt the liquid level in the tube Malcolm Baird received his PhD in chemical engi neering from Cambridge University in 1960. After some industrial experi ence and a post-doctoral fellowship at the Univer sity of Edinburgh he joined the McMaster Uni versity faculty in 1967. His research interests are l i uid-liquid extraction osi......-...11.. eling applications in kidney failure patients main tained on dialysis cillatory fluid flows, and hydrodynamic modeling of metallurgical processes Addr ess: McMaster University, Hamilton Ontario Canada LBS 4L7 Copyright ChE Division of ASEE 2000 158 C h em i cal Engineering Edu c ation

    PAGE 65

    This paper describes a simple e x perimental technique for determining binar y diffusion coe ff icients for v apor ( A )-ga s (BJ s y stems in w hich the v apor i s generated b y the e v aporation of a pure v olatil e liqu i d and the gas is air falls through distance dz. The volume of liquid evaporated will be given by ( If the density of the liquid is PA and the molecular weight is MA> the molar evaporation of A will be equal to p A ( MA and the rate of evaporation, p A dt, can be related to the diffusional flux (N A z a) by the following equation: N a= PA a dz A z MA dt (2) Assuming the liquid level drops very slowly and therefore pseudo-steady-state conditions apply, N Az in Eq. (2) may be substituted for by Eq. (1), giving PA dz DABPt ( ) MA dt = RTzpBM PA, PA 2 which can be rearranged as z dz= C dt where C = DABP1MA (PA PA 2 ) RTPsMPA Equation (4) can be integrated as which yields r zdz =Cf dt Zo Q z2 -z2 ___ o =Ct 2 (3) (4) (5) (6) (7) Eq u ation (7) suggests that a plot of ( z 2 z5) I 2 vs t will be linear, passing through the origin and having a slope C. The value of D A B can therefore be calculated from the mea sured slope C by rearranging Eq. (5) as D CRTPsMPA AB P1MA(PA1 PA2) (8) It may be noted that P A is the vapor pressure of liquid A at T, and pA 2 may be safely assumed to be zero as fresh air flows over the t u be. EXPERIMENTAL PROCEDURE The apparatus is quite simple. It consists of a small glass tube of 1to 2-cm diameter (see Figure 1), a traveling microscope, a so u rce of light to illuminate the liquid menis cus, a thermometer, and a barometer. All these components can be easily found in any chemical engineering laboratory. Spring 2000 The volati l e liquids selected should have high vapor pres sures to get meaningful results in a reasonable time period acetone, pentane and hexane were tested in this study. Tes t s performed in duplicate indicated that the results were repro ducible within the experimental accuracy. The following procedure is recommended: 1. Fill the tube with the volatile liquid to about 0.5 to 1.0 cm from the top Care should be taken to pipet the liquid in the tube to avoid wetting the top empty section of the tube with the liquid. 2. Place the tube in a stand and place the stand in an illuminated fume-hood. 3. Note the atmospheric pressure and the temperature in the fume hood 4. Keep the fume hood fan off and the door fully open (the front glass panel fully raised) to minimize any air turbulence due to suction in the fume hood. 5. Focus the traveling microscope first at the very top of the tube (z = 0) and then at the liquid menisc u s level (z=z 0 ), and immediate l y start the stopwatch. 6 Record the liquid level (z) in the tube with time to obtain a noticeable drop in the liquid depth. This gives the z-versus-t data. 7. Note the atmospheric pressure and the temperature i n the fume hood again. 8. Measure the liquid density at the experimental temperature by weighing a known vo l ume of t h e liquid 9. Plot ( z 2 z5) / 2 versus t and obtain the experimenta l diffusion coefficient from the slope of the plot as per Eq. (8). 10. Predict the diffusion coefficient from the Hirschfelder Bird-Spotz correlation 121 given below and compare with the experimental value obtained in Step 9 above. (9) 11 Repeat the experiment with a pedestal fan an d /or fume hood fan on, and compare the experimental diffusiv i ty values with those with no air circulation in t h e fume hood 159

    PAGE 66

    35 20 30 15 25 'f E t 20 t 10 1 5 "; "; I I !!. !!. 10 500 1000 1500 2000 2500 3000 3500 4000 500 1000 1500 2000 2500 Time (s) Time (s) Figure 2. Pl o t o f ( z 2 z5 ) / 2 vers u s t im e fo r a ce t o n e vapor di ffus in g t h ro u g h sta ti o na ry a ir Figure 3 Plo t of ( z 2 z5) / 2 ve r s u s tim e for h exa n e vapor diffu si n g thro u g h stat i o n ary a i r. P, Vapor ( Nim ') Ace t o n e 98960 H exa n e 99 1 55 Pe ntan e 1 0 0 6 1 6 Pe nt a n e 1006 1 6 P e ntan e 1 006 1 6 P e ntan e 1 006 1 6 1 60 1400 / Fa n Off / 1 2 00 Fan On S l ow / / Fan On Fa s t / Fan On M e dium+ Fum e H ood F a n / 1000 / 'f / / C = 0.4480 t 800 / / / / 600 :7 "; I !!. 400 2 00 C= O O~,' 'c = 0 1 6 80 0 --==::::;=:=---.=...----~-~------~---l 0 500 1000 1 5 00 2000 2500 3000 3500 4000 4500 Time (s) Figure 4. Plot o f ( z 2 z5 ) / 2 ve r s u s tim e for pe ntan e a i r sys t e m w ith va r yi n g l eve l s of air c ir c ul a ti o n TABLE 1 Experimental and Predicted Values ofD A 8 Co mp o n e nt B = stagna n t a i r; p A 2 ass um e d zero p .,' Temp. Air Flow Diffu s ion Coefficient ( 10 .. m 2 / s) % ( N/m 2 ) ( Q C) Fume Pede s tal Experimental Pr e dicted Deviation Hood Fan Fan 23905 2 0 Off Off JO. I 1 0 .8 -6.5 1 6029 1 9 Off Off 8.6 8. 1 +6.2 76500 27 Off Off 8.5 9.2 -7 6 76500 27 Off S l ow 25 5 76500 27 Off Fas t 63.6 76500 27 On Me dium 67.9 C h e m i c al Eng in eering Ed u cat i on

    PAGE 67

    RESULTS AND DISCUSSION The results for acetone and hexane at atmospheric pres sure, room temperature and with no forced air circulation through the fume hood are plotted in Figures 2 and 3, respec tively, and for pentane without and with forced air flow through the fume hood in Figure 4. As can be noted the ( 2 2) plots of z zo I 2 versus t for acetone and hexane (see Figures 2 and 3, respectively) yield linear lines passing through the origin with a good fit of data point s (R 2 >0.986). The experimental and the predicted values of diffusion coef ficients, calculated from the slopes of these plots and the first line on Figure 4 for pentane (with no forced circulation in the fume hood), and from the Hirschfelder-Bird-Spotz correlationr 21 (Eq. 9), respectively, are given in Table 1. The results suggest that the experimental technique is simple and gives reasonable agreement between experimen tal and predicted values. The last three experiments performed with forced air flow in the fume hood (see Table 1 and Figure 4) indicate an increase in the apparent value of D AB, as expected. These results reflect a very important limitation of this procedure i.e. for ensuring "s tagnant B" conditions (to obtain a good agreement between the experimental and the predicted val ues of D AB) on which the development of Eq. ( 7) is depen dent, undue air turbulence in the fume hood must be ab sent. Any external turbulence can affect the behavior of the gas mixture in th e tube and lead to an increa se in th e mass-transfer rate. It may be noted that for this method to work the den sity of the vapor (A) should be greater than that of air (B) so that there are no natural convection effects in the tube. This i s the case with all common organic liquids CONCLUSIONS 1 Rate of fall in liquid level can be u sed to determine the diffusion coefficient fairly accurately for vapor-gas sys tems where the vapor is generated by the evaporation of a pure volatile liquid and the gas i s stag nant. 2. Experimental diffusion coefficients are within % of the predicted values. 3. Turbulence in the experimental area affects the preci sion of the results. GENERAL REMARKS This laboratory provides students with the opportunity of experiencing how elementary experimental methods can be used to confirm what they read in the classroom The experi ment i s extremely simple and can be completed well within the usual three-hour laboratory period Since linear plots passing through the origin are obtained, only two level read ings, about 20 minute s a part are required for a direct calcu lation of D AB through Eqs. (7) and (8). Spring 2000 We recommend that the class be divided into groups and that different groups s tudy the effects of 1) nature of compo nent A, i.e., study different volatile liquids; 2) degree of turbulence in the work s tation (some effort can be made to quantify the results by measuring air velocity in the fume hood with an anemometer); 3) temperature; a n d 4) natural convection effects in the evaporation tube (this can be stud ied with any liquid with a molecular weight lower than that for air-water being the safest). The students should also be asked to review the analysis of so urce s of error in such a procedure provided by various workers.r 6 8 1 NOMENCLATURE a cross-sectional area of evaporation tube C s lope of ( z 2 25) /2 vs t plot D diffusion coefficient f function k B o ltzmann 's constant M molecular weight N Az s teady-state molar flux of A in the z-direction p pressure R ideal gas constant r molecular se paration at collision T absolute temperature t time z vert ical di sta nce ene r gy of molecular attraction p liquid d ensity Subscripts A co mponent A B compo nent B AB co mponent s A and B BM log-mean average for component B across the diffusion path 0 initial va lue t total z in the z-d ir ec tion beginning of diffusion path 2 e nd of diffusion p at h REFERENCES 1. Perry, R.H ., and C.H. Chilton Chemical Engineers Hand book, 5th ed., McGraw-Hill Book Company, New York, NY pp. 3 -230 ( 1973 ) 2. Tr eyba l R.E. Mass-Transfer Operations, 3rd ed. McGraw Hill Book Company New York NY, pp. 28-31 ( 1987 ) 3. McCabe, W.L. and J C Smith Unit Op eratio n s of Chemical Engin ee ring, 5th e d ., McGraw-Hill Book Company, New York NY Ch 21 ( 1993 ) 4. Geankoplis, C.J., Transport Processes and Unit Operations, 3rd ed., Prentice H a ll Book Company, Englewood Cliffs, NJ p 390 ( 199 3) 5 Perry, R.H. and C H. Chilton, Chemical Engin ee rs Hand book, 5th ed., McGraw-Hill Book Company, New York NY Table 3.8 ( 1973 ) 6. L ee, C.Y., and C.R. Wilke, Ind. Eng. Chem ., 46 2381 ( 1954 ) 7. Rao S S. and C.O. Bennett, Ind Eng. Chem Fund ., 5 573 ( 1966 ) 8. Pommersheim J.M., and B.A. Ranck, Ind. Eng. Chem. Fund. 15, 246 ( 1977 ) 0 161

    PAGE 68

    .t3 11181 ij 111111111 -c u rr_i_c u l u_m _ _____ __ ) INCORPORATING MOLECULAR MODELING INTO THE CHE CURRICULUM R OBERT M. BALDWIN, JAMES F. ELY, J. D OUGLAS WAY, STEPHEN R. DANIEL Colorado School of Mines Golden, CO 80401 C omputers have long been used in the teaching of chemical engineering in order to facilitate complex calculations required for the design and analysis of chemical process equipment (plug-flow reactors multistage distillation columns, etc) The use of computer-based pro cess simulation using commercial software (Aspen Plus ProVision Hysim, etc.) is commonplace in most modern chemical engineering curricula. Today, the availability of powerful molecular modeling software is adding an entirely new vehicle for predicting the behavior of systems and pro cesses based on molecular-scale properties. While the prin ciples are not new, only recently has the computational hardware and software become available that can bring these tools (Gaussian, Spartan Cerius 2 etc) into the chemical engineering classroom. The capability of combinjng compu tation with visualization presents chemical engineering edu cators with important new opportunitie s for enhanced teach ing and learning. PARADIGMS IN CHE EDUCATION Wei 1 1 1 commented on the two paradigms that shaped chemi cal engineering education during the 20th century. The first of these was based on classification of processes and sys tems a s the familiar unit-operations lexicon; this approach dominated the early stages of chemical engineering teach ing. The publication of Transport Phenomena 12 1 marked the beginning of the second paradigm that of the fundamental analytical approach based on rigorous mathematical models of physical systems. Recently a third paradigm for chemical engineering was proposed by Landau,' 31 that of a closer relationship with practice and industry. It is our opinion however, that the third paradigm could and should be cast in the context of better integration of the fundamental molecular processes of chemical physics into chemical engineering. Other educator s have discussed the importance of the microscopic viewpoint in our teaching and 162 research 14 51 but in today's chemical engineering curriculum the basic atomistic concepts learned in organic and physical chemistry are too often left to languish as soon as the spe cific courses dealing with these subject areas have been completed. This is caused in large part by a lack of continu ity between subject matter and by poor integration in terms of teaching of the two disciplines. Important concepts in organic synthesis and molecular structure are rapidly forgot ten by thirdand fourth-year chemical engineering stu dents, just at the time that these concepts should be ap plied (for example in the process design and/or reaction engineering courses) At the Colorado School of Mines (CSM), we have recently completed a top-to-bottom school-wide redesign of our un dergraduate curriculum As part of trus exercise the under graduate chermcal engineering curriculum was significantly updated and revised. A key philosophical component in this revision process was our desire to incorporate molecular modeling and simulation into the chemistry and chermcal engineering course sequence in order to foster a better apR obe rt M. Bal dwi n is Professor and Head of the Chemical Engineering and Petroleum Refining Department. He received his BS and MS degrees from Iowa State University and his PhD from the Colorado School of Mines His research interests include membrane separations computa tional chemistry fuels science and catalysis Ja me s F. Ely is Professor of Chemical Engineering and Petroleum Refin ing at the Colorado School of Mines He received his BS from Butler University and his PhD from Indiana University His research interests include molecular simulation and thermodynamics. J. Doug l a s W ay is Associate Professor of Chemical Engineering and Petroleum Refining at the Colorado School of Mines He received his BS, MS and PhD degrees from the University of Colorado at Boulder His research interests include novel separation processes membrane tech nology molecular simulation and computational chemistry St eph en R Dan iel is Professor and Head of the Department of Chemistry and Geochemistry at Colorado School of Mines He received his BS MS and PhD degrees in interdisciplinary chemistry / chemical engineering pro grams at the Colorado School of Mines. He teaches courses in organic chemistry inorganic chemistry and analytical chemistry at both under graduate and graduate levels Copyr i g ht C hE Divi s io n of ASEE 2000 Ch e mical Engin ee rin g Education

    PAGE 69

    preciation of the relationship between microscopic and macroscopic phenomena and a better understanding of the importance of chemical physics in determining how molecules interact and react: molecular propertie s <---m 0 1 e-c -ul_ar_____, macroscopic proces s es m o d e lin g As described above, we believe that computer-aided molecular modeling can serve as the catalyst that allows students to make and understand these connections. MOLECULAR MODELING AND SIMULATION The two main components of our microscopic approach to understanding macroscopic pro cesses are molecular modeling and molecular simulation. The distinction between these two is somewhat arbitrary, but in our case we define molecular modeling to be the investigation of isolated molecular assemblies (e.g., single molecules, dimers, etc.) and molecular simulation to be the investigation of collections of interacting molecules The primary tools used to perform molecular modeling are ab initio and semi-empirical quantum mechanics and molecular me chanics,l 7 1 while molecular simulation incorporates the use of molecular dynamics and, for example, Monte Carlo methods_[ B J The primary use of molecular mechanics is to make empirical estimations of equilibrium molecular geometry (e.g., the most energetically favorable structure) and energy by using parameterized force fields. For homogeneous systems molecular mechanics describes the total energy of a molecule as the sum of a distortion energy from an ideal geometry of connected atoms (E 1 ) E, = L E s tr e t c hin g + L Eb e ndin g + L El o r s i on (1) bond s b o nd dih e dr a l an g l es an g l es and the contribution due to non-bonded interactions (E 2 ) that arise from van der Waals and electrostatic interactions E 2 = L L E?W + L L E ii l ec t rosta t ic ( 2) j j The total energy is just E I + E 2 Examples of molecular mechanics force fields are SYBYL I9I and MMFF. II 01 Depending on the nature and applicability of the force field being used to carry out the calculation, this procedure may work well or can give rise to structures, geometries and hence equilibrium energies that are significantly in error In so-called computational chemistry programs, force field calculations are often employed to give a refined structure as a starting point for the ab initio quantum mechanics calculations The second type of computation that is often used for molecular modeling directly involves quantum mechanics with the general mathematical relationship given by the Schrodinger equation Toda y, the availability of powerful molecular modeling software is adding an entirely new vehicle for predicting the behavior of systems and processes based on molecular scale properties. In this equation, V is the momentum operator m refers to a mass, Z is a charge, e is the unit measure of charge, R denotes nuclear positions and r denotes electronic positions 'f' i s the quantum mechanical wave function for the molecule, and E is the energy of the molecule. According to quantum mechanics s olution for the wave function enables one to calculate the energy and other structural properties of a molecule. Although this equation may be written down rather simply, it cannot be solved exactly except for the hydrogen atom (one electron, one proton). While exact solution of this equation for polyatomic molecules is still not Spring 2000 163

    PAGE 70

    feasible, application of a series of approxjmations (in par ticular, the Hartree-Fock approximation) and the advent of powerful desktop computers u s ing highspeed rrucroproces sors has made numerical solutions possible l 6 1 A high degree of computational efficiency can be obtained by using semi-empirical quantum mechanjcal methods that consider only the valence electrons in a molecule and reduce the number of electron-electron interactions by neglecting "ove rlap s" between atomic orbitals. In addition, these model s introduce parameters that have been optimized with experimental data In contrast to molecular mechanics and ab initio quantum mechanic s, molecular simulation technjques examine the behavior of systems composed of a small collection (typi cally 100 to 10,000) of interacting molecules. The method s used include molecular dynamics (i n which the time evolu tion of the molecular system is simulated and monjtored) and Monte Carlo simulation (in which s tatistical methods are used to samp le states of the system according to some pre-defined probability di stribution). The basic method u se d in molecular dynamics is to solve numerically the classical equations of motion of the mol ecules (struc tured mass points) and calculate time averages of quantities such as the configurational energy, pressure, se lf-diffu s ion coefficients, local structure, etc. Typically, the molecular system is sim ulated for picosecond time intervals, which may involve so lving the coupled differential equa tions of motion for several hundred thousand time-steps. The accuracy of these sim ulations is governed by the numerical techniques used and the accuracy of the interac tion potential(s ) that govern the motion and time evolution of the mol ec ule s in the system. For s tructured (polyatomic) molecule s, these potentials might include intra-molecular vibration, rotation, and non-bonded interaction potentials as well as non-bonded intermolecular potentials. The non-bonded potential s are typically parameterized intermolecular potential functions s uch as the Lennard-Jones or Exponential-6 models. An advantage of molecular dynamics is that non-equilibrium propertie s may be calculated with relative ease In Monte Carlo si mulation s, the energy of the molecular system is minirruzed by randomly moving molecules ac cording to some desired probability distribution. Again, the u ser must specify the potential functions, and equilibrium properties can be calculated by statis tical (rather than time) averages A great advantage of this method is the relative ease with which it can be u se d to calculate phase equilibria .r 111 Molecular modeling and simulation is finding widespread applicability and increa se d acceptance in chemical engineer ing practice Estimation of thermophysical properties has become routine. L 1 2 1 Simulation of rheological properties of complex fluids has been demon s trated by Cummjngs and co-workers 11 3 141 and 164 molecular simulation of water ha s been s hown to give struc tural information that i s more reliable than even the most precise measurements can yield 1 1 51 In addition, u se of molecular mechanics and quantum chemistry calculations to determine orbital occupancy has been s hown to be important in under sta nding and design of new materials s uch as catalysts 1161 so rbent s,l' 71 and reactive polymer membranes. 11 8 1 MOLECULAR SIMULATION IN TH E UNDERGRADUATE CHEMISTRY CURRICULUM At CSM, st udent s first encounter application s of molecu lar modeling in their sop homore-l eve l organic chemi s try course seq uence. Calculations are facilitated using Spartan, which i s a user-friendly computational quantum chemistry software package. Computational quantum chemistry problems are assigned essentially as se lf -pace d discovery exercises in Organic I and II. Students first use Spartan to carry out quantum calcu lations in order to investigate str ucture/ stability relationshjps for typical hydrocarbon s, functional groups, a nd reactive intermediates (radica ls carbocations, carbenes etc .). Most calculations are carrie d out using geometry optimization at the semi empirical AMI level: some calculations require ab initio methods with hjgher levels of theory (3-21G) ** In either case the computations can be rapidly completed using either Spartan 's PC-based software or the Unix workstation version; a total of approximately sixty licen sed copies of Spartan are avai lable to c hemical e n gineering students in open computer lab s on both platforms. Spartan's ability to calculate and display electron density and molecular orbital s urface s is exploited in the organic course se quence where the focus is on under sta nding the mechani sms of chemical reactions The relationships between electronic str ucture molecular orbital density, and chemical reactivity are also developed using the visualizatio n capabilities of the software. For ex ample, when s tudying nucleoprulic s ubstitution reactions the s tudents u se Spartan to compute HOMO ( highest occu pied molecular orbital) and LUMO (lowest unoccupied mo lecular orbital) surfaces for the reactants, and then relate the electron transfer tiling plac e in the frontier orbitals to the observed regiochemistry and selectivity of the reaction. For both the organic and physical chemistry seq uences Spartan 's capabilities of calculating and graphically render ing e lectron den sity molecular orbital s urface s greatly facili tates the s tudent 's under standi ng of the relationship be tween molecular propertie s and s uch important concepts Austin M et hod 1 : a se mie mp i r ic al mol ec ular orbital m e thod A basis set in whi c h each inn e r-shell atomic orbital is written in t e rms of thr ee Gau ss ian fun c tions and each valen ce -shell atomic orbital is split into two parts writt e n in t er m s of two and on e Gau ssi an s, respectively. C h e mi ca l Engine e ring Edu c ation

    PAGE 71

    as stability and reacti v ity These concepts are further elaborated during the third year phy s ical chemistry sequence Here the s tudent s are exposed in cl ass to the theory and so me of the m at hemati ca l details associated with se tting up a nd arriving at approxi mate solutions to the Schrodinger equation. As an exa mpl e, the Mor se potential e ner gy diagram for a diatomic molecul e is first calculated u si n g measured spectroscop i c ( IR ) data from th e lab and then s imulated u sing Spartan MOLECULAR MODELING IN THE CHE CURRICULUM AT CSM Incorporation of molecular modeling in the chemical engi neering curriculum was first accomplished two years ago in our senior-level reaction engineering course. As an example of the approach bein g u se d an outline for one of the computational chemistry homework probl e m s assigned in thi s cla ss i s shown in Table 1 The problem deal s with sy nthe s i s of chemical-grade etha nol via a S N2 nucleophilic substitution re ac tion in a queou s solution. The objective of the problem is to illu s trate u se of quantum mechanics and computational chemistry in order to generate the thermochemical information required to carry out an analysis of a sim ple indu s trial reaction. Two pos si bl e reactions are proposed differing only in choice of substra te : C2H 5 CI+OH HC 2 H 5 0H+ C l C2 H 5 Br + OH H C 2 H 5 0H + Br As a first s tep s tudents are as ked to inve s tigate the ther mod y namic s of th e reaction s. For thi s p a rt of the problem he ats of formation of all product s and reactants (including solvation energy effects) are es tim ated b y se mie mpirical quantum chemistry methods, and the h eat of reaction co put e d in the normal fashion: products reactants ~HR= L vj~F j L vi~HF i (4) j Calculation of the equilibrium constant require s the free e n ergy change for the re actio n K=exp(, wi/RT) (5) but if e ntropic effec ts are not important (a reasonable as s umption in this case) ~Gi = ~Hi -T~So ~Gi "" ~Hi (6) and the s tandard free energy c hange and hence the equilib rium co n s tant for eac h reaction can be readily estimated TA BL E 1 Exa mpl e Prob l e m in R eaction K i netics Synthesis of c hemi ca lgrade etha n ol ca n be ac hi eved by a nucleophili c s ub s tituti on reaction u si n g h ydroxide ion as the nucleophile a nd a haloethane as the s ub stra t e. For thi s problem, we wi ll investigate th e rates of two synt h esis r eact i ons differing o nl y in the nature of the h a l ogen atom (brom in e vs. chlorine): C 2 H 5 C l + OH H C 2 H 5 OH + C l C2 H 5 Br +OHHC 2 H 5 OH+Br The reaction talces place und er aq u eous conditions Both reactions can be ass um e d to fo ll ow a SN2 (substitution/ nucl eop hili c/b im o l ec lar ) mechanism Hen ce the geo metry a nd co nfi g urati on of th e transition state ca n be ass um e d to be the sa m e fo r both reactions We wis h t o estimate the ratio of the rates of th ese two reactions I. Estimate th e activation ene r gies for both th e forwar d and reverse r eactio n s u s in g Spartan This wi ll require severa l ass umpti o n s regarding th e exact geo m etry of the transition state, namely Assume that the nucleophile ( the attacking gro up ) and l eav in g gro up are both attached to the same carbon atom and are in ax i a l positions (e g., 180 apart) For S N2 reactions trigonal bi-pyramidal geo m et r y at the carbon atom where the nucl eop hil e is attacki n g gives a reasonable a ppro xi m atio n t o the trans ition s t ate. To obtain the energy of the tran s iti on state h ave Spar tan carry out a Semi-Empirical Transition Structure calculation u si n g AM I as the mod e l a nd wa t e r (Water C-T) as the so l vent. Remember to se t up the correct charge and multiplicity for your assu m ed tr ansit i on sta t e. Obtain heats of formation from Spartan for th e i o ni c spec i es ( Semi-Empirical Single P oint Energy, Spring 2000 AM I Water C-T) Obtain h ea t s of formation for th e o ther reactants a nd products using Sem i-Empiri ca l Geometry Optimization as t h e ta sk, AM I as the model, a nd Water C-T as th e so l vent. 2. Using data on heat s of fo rm atio n of the reactant s a nd the produ cts from Spartan, calcu l a t e the h ea t of r eac ti on for both nucleophilic s ub sti tut ion reactions. Which reactio n i s favored if the reactions are under thermod y namic contro l ? Calc ul ate the ratio of th e equilibri u m constants for the se two reaction s at 25 C 3. Calc ul a t e the ratio of the r ate of the s ub s tituti o n reaction for bromoetha n e as the s ub strate to the s ub s tituti o n reac tion w h en ch l oroethane i s the substrate in the temperature ra n ge from 25 C to 1 00 C. What ass umpti o n s are n ecessary to carry o ut thi s calcu l a ti o n ? Are these ass umpti o n s reasonable ? Does th e ratio c h a n ge w i t h temperature ? Why? 4. a) If yo u were go in g to eng in eer a reactor for manufact ur e of chemical-grade ethanol u sing one of these two reactions w hi c h haloethane wo ul d yo u recommend be u sed a nd w h y? Are th e r e a n y important fac t ors th at you are n o t considering in yo ur c hoi ce for a s ub strate ? b ) Would yo u suggest th e proces s be carr i ed o ut at l ow tempera ture or high temperature and why? You may want to cons ult the Chemical Marketing R e port e r ( ref e renc e room, CSM library) for data that will h elp answer this question. Up-to-date information on some c h e micals can also b e found at and . 165

    PAGE 72

    based on the heat of reaction. Using ratios for the two equi librium constants in evaluating the thermodynamic feasibil ity makes this assumption much less restrictive. This analysis shows that both reactions are favorable ther modynamically, with a preference to chloroethane as the substrate (larger equilibrium constant). The reactions are also shown to be under kinetic control, hence the next s tep is to see what difference s may exist in the activation energies for the two reactions. Thi s is accomplished by constructing a hypothetical tran si tion s tate for the nucleophilic s ub sti tution reaction for both reactions, and by using Spartan to estimate the energy of these species. This part of the so lution process draws heavily on the student's background in organic chemistry theory where substitution reaction s are concerned. Once the transition states for both reactions have been constructed, values for the heat s of formation of the reactive intermediates are determined u s ing the Transition Structure Optimization routine (searching for a sa ddle point on the reaction potential energy s urface ) in Spartan. Students can include a calculation of the vibrational s pec trum at this point in order to verify that a rea so nable approxi mation for the tran s ition s tate s pecies has been found by this procedure (at lea s t one imaginary frequency that corresponds to the reaction coordinate of interest). Animation of the largest imaginary frequency in the calculated table of normal mode frequencies provide s convincing evidence that the re action coordinate of int erest corresponds well to th e transi tion state structure. Finally, the activation energies for both reaction s can be readily estimated from the semi-empirical heat s of formation as reactants EA = ~Hr.Ts 2. ~Hr (7) Comparison of the relative rates at any temperature T, then follows directly from (8) From this proce ss, the s tudent s find that the activation TABLE2 Course Outline and Instructional Modules Molecular Perspectives in Chemical Engineering Leaming Obiectives The class introduces st ud en t s to the u se of molecular-scale techniques fo r the prediction of physical properties, transport properties and reaction e n ergetics. Content Summary This class intr o du ces modern methodologies for the est im a ti on of physical transport and reaction propertie s and parameters needed in the design of chemical proce sses. In addition, it serves to enhance stude nt s' molecular-scale intuition through the use of group contribu tion methods molecular simu l at i ons, quantum mechani ca l ca lcul tion s, and molecular vis u a li zation. The class begins with a r eview of th e mi crosco pi c world of ato m s a nd molecules; fundamental l e ngth tim e, and e n e r gy sca l es are discussed Molecular-scale forces and th eir representative potentials are presented. Case st udi es are pursued inv o lving topics s uch as the esti m atio n of diffusion coefficients Module Description v i scos it y, a nd pha se e quilibria as well as tr ans ition -s tat e theory for th e esti m atio n of rate consta nt s in c h emical reactions. R e l evant experi mental techniques that can serve to ver i fy th e molecularsca le calcu l ations are cove r ed. Significant hand s-o n experience in a computer l aboratory and case-st ud y projects i s emp h asized. Topics Covered 1 Comp ut ers and computer simu l ation in chemical engineering 2. Propertie s of fluid s and solids; molecular s tructure prediction methods 3. Co mputati ona l quantum chemistry, int ra mol ec ul ar properties 4. Intermolecular propertie s and forces 5 Intermolecular force s a nd config urati o n a l properties 6. Eq uilibrium molecular dynamics 7 Monte Carlo t ec hniqu es 8. Nonequilibrium molecular dynamics Ideal Gas Properties Vapor-Liquid Equilibria Simulation a nd text to illustrate how molecular motion s g i ve ri se to ideal gas propertie s. Simulation to illu strate h ow inter-molecular int eractions affect the dynamics and VLE of mixtures. Group Contribution s Diffusion in Pol ymers Thermochemical Prop er ti es Structure-Property R e l at i ons hip s Activation Energies lntramolecular Quantum Behavior Quantum m ec hani cal calc ul atio n of Benson groups Molecular dynamics s imul atio n a nd visualization of nitrogen diffu s ion in polysiloxane. Use of computational c h e mi s try to estimate thermochemic a l prop erties Free radi cal polymerization of vinyl c hlorid e to form PVC Use of quantum mechanics to in vestigate thermal cracking of ethane. Quantum m ec hanic s of molecules: potentials vibra ti ons, IR spectra, an d eq uilibrium geo m etr i es. Intermolecular Forces Estimation of intermolecular force u s ing quantum chemistry. Thermodynamics of R are Ga s Mixture s Application of molecular dynamic s simu lati ons to the estimation of mixture properties. 1 66 Chemical Engineer in g Educat i on

    PAGE 73

    energy for the synt he sis using bromoethane as the s ub stra te i s significantly lower thu s suggesting that the reaction will be much faster if this compound is used as the reactant. Finally, costs of the two reactants are compared u si ng d a ta from the commodities literature The approach to so lving thi s problem r e lie s exclusively on the use of molecular modeling to obtain information th a t i s not readily ava ilable from any of th e s tandard data so urc es hence the u se of quantum chemistry to es timate parameter s that are of considerable practical utilit y for both reactor and proce ssde s ign purpo ses is well illustrated. We have recentl y added a new se nior-l eve l course to o ur curriculum, Molecular Per s pecti ves in Chemical Engineer ing." Thi s course present s students with a comprehensive overview of the u se of molecular modeling and s imulation technique s in several different applications, includin g esti m a tion of thermoph ys ical and reaction rate data sorption equilibria and diffu s ion rates, pha se equilibrium s imulation and prediction of transport properties An outline for this course, including descriptions of the computational exercises that are currently in u se, is g i ve n in Table 2. Examples of molecular modelin g exercises u se d in the capstone chemical engineering molecular si mul atio n course can also be found by accessing the CSM website at . CONCLUSIONS Molecular-scale modeling ha s reached a le ve l of so phi s ti cation and acc uracy that make s it an esse ntial a nd hi g hl y u sefu l tool for chemical engineers yet th e m et hod s, ca p bilities and limitations of this tool are n ot yet well known across the chemical engineering profe ssio n. The u se of mo lecular-scale modeling is becomin g increa si ngly important in industry as researchers and product developer s look for ways to cut the costs and time associated with de ve lopm e nt of new products At CSM we h ave a ddre sse d thi s probl em b y inc o rporating atomistic modelin g method s throughout o ur curriculum at the undergraduate level in both th e chemistry and chemical engineering course se quence s. We belie ve that thi s approach represents a new e ducational paradigm in chemical engi neering and we are committed to integration of these con cepts across the curriculum. REFERENCES 1. W e i J .,Adu Chem. Eng., 16 51 ( 1991 ) 2 Bird R.B., W.E Stewart, an d E.N Lightfoot, Transport Ph enomena, Wiley & Sons, New York NY ( 1960 ) 3 Landau R. Chem. Eng. Prag 52 J an ( 1997 ) 4. Drexler KE ., Pro c Natl Acad Sci. US 78 5275 ( 19 81 ) 5. Gibney, K., ASEE Pri sm, 8 ( 1 ) 23 ( 1998 ) 6. Hehre W.J ., L. Radom P .v. R. Schleyer, and J A Popl e Ab Inito Molecular Orbital Th eory, Wil ey & Sons, New York, NY ( 1986 ) Spring2000 7. Burkert, U. and N.L. Allinger Molecular Mechanics ACS Monograph 177 American Chemical Society ( 1982 ) 8. Frenkel D ., and B Smit, Und e rstanding Molecular Simu lation Academic P ress ( 19 96 ) 9. Clark M ., R.D. Cramer III and N va n Opdensch, J. Comp. Chem 10 982 ( 19 89 ) 10. Halgren, T.A., J. Comp. Chem. 17, 490 ( 1996 ) 11. Quirke N. Chem. Eng. Prag., Feb ( 1996 ) 12 Rowley R.L. and J.F Ely, Mol ecu lar Simulation, 7 303 ( 1991 ) 1 3. Moore J.D., S.T. Cui, P T. Cummings, and H D Cochran AIChE J ., 43 3260 ( 1997 ) 14. Cui S.T H.D. Cochran P.T C ummin gs and S.K. Kumar Macromolecul e s 30 3375 ( 1997 ) 1 5. Chia lvo A.A., a nd P .T. C ummin gs, J. Chem. Ph ys 105 8274 ( 1996 ) 16. Hansen, E., and M. Neurock, Computers in C h em. Eng., 22 1045 ( 199 8 ) 17 C h en N. and R.T. Yang Ind. Eng. Chem. R es 35 4020 ( 1996 ) 1 8. Sungpet, A., "Reac tiv e Polymer Membranes for Olefin Sepa rations ," PhD Th es is Co lorado School of Mines ( 1997 ) 0 ,ar.,e-=:tt:-:e=-:r=-=-to=-::t:h:--::e=--::e~d;;i:to=-r=-----) Dear Sir: The article b y Rugarcia, et a l. titled "The Future of Chemi cal Engineering Education" [CEE 34 16 (20 00)] i s inter esting and thought provoking However it begin s with a caricature of a poor lecture a nd returns t o the them e of the inferiority of the le c ture format later in the paper with the asser tion that the s uperiority of a lternati ve methods ... has be en demon stra ted in thousands of e mpiric a l re searc h s tud ie s." This v i ew ha s become widely accepted among the proponent s of n ew" teaching method s At the ri s k of being branded as a Luddite ( probably true ), I am com pelled to offer a modest and purely anecdotal defense of the lecture format. Looking back on m y own experience as an undergraduate the classes that I most enjoyed were all formal lecture s in ph ys ic s, chemistry, and even geology The se lecture s were give n to l arge classes (so m eti me s severa l hundred s tudent s) a nd I am sure that the lecturer s would have been horrified at the thought of following a course textbook or of presenting worked examples during a lectur e. What was presented was an in-depth review stressing the fundamental principles and the logic and coherence of our under s tanding of t h e s ubject. It i s perhap s ironic that the note s from severa l of these courses were later publi s hed as successful textbooks! Well thought-out and well-rehearsed dem onstrat ion experiment s, performed by a teaching assistant, were sometimes included Que stio n s, assignments, and practice examples were handled in p ara llel tutorial s e ss ion s given by either a faculty m e mber or a PhD s tudent each with no more than -------------Continued on pa ge 177 16 7

    PAGE 74

    .,a_5_3._c_l_a_s_s_r_o_o_m _ _____ __,) IS MATTER CONVERTED TO ENERGY IN REACTIONS? P AUL K. ANDERSEN New Mexico State University Las Cruces NM 88003 P erhaps the most famous equation of physic s i s Einstein 's mass-energy relation E=mcz (t) A l though this equation is well known it is often misunder stood to mean that matter is converted to energy (or vice versa) in reactions ; that matter is a form of energy; and that the principle of energy conservation must be modified. Con sider, for example, the following excerpt from a general c h emistry textY 1 Regardle ss of the classification used physical reaction phase change ordinary chemical change chemica l change nuclear reaction-changes in matter involv e the change of matter to energy if the reaction ev olves e nerg y, and the c han ge of energ y to matter if the reaction absorbs ene r gy. Energy and matter are thu s inter c hangeable The scope of the conservation principle is ther efo re e nlarged to includ e energy as a form of matter or matt er as a form of energy .. The convertibility of matter a nd e n ergy i s described by th e equation E = mc 2 predicted by A l bert Einstein in 1 905 ... As we shall see, Eq. (1) does not say that matter and energy are interchangeable, or that matter is a form of energy. Nor doe s it extend the principle of conservation of energy. MA TIER INTO ENERGY? One difficulty with the foregoing quote is an ambiguity over the meaning of t h e word matter. There are at least two common ways to measure the quantity of matter in a body: by its mass or by the numbers of elementary particles it contai n s. The latter is us u ally expressed in moles. It is easy to show that the constituents of matter are not created or destroyed in ordinary chemical reactions For example, hydrogen and oxygen react according to the equation I H z +-Oz H z O 2 On the l eft side of the equation we find two moles of hydrogen and one mole of oxygen ; the same is true of the 168 ot h er side. There is no conversion of matter into energy, or vice versa. What about nuclear reactions ? Consider one of the fission reactions that may occur when a neutron is absorbed by uranium-235 : Z3 5U+ I n l4ZBa+9I Kr+ 3 In 92 0 56 36 0 (2) A count of the protons electrons, and neutrons before and after the reaction shows no change: Before After 92 proton s 92 electrons 144 neutron s 92 protons 92 e l ectrons 144 neutron s Once again we have an example in which the constituents of matter are conserved in an exothermic reaction. There i s no conversion of matter into energy. To be sure, there are processes in which matter can be created or destroyed. In the reaction between an electron and a positron for examp l e, both particles are annihilated and two photons are formed: e-+e+ 2y Nevertheless, we conclude that it is not generally true that matter is converted to energy (or energy into matter) in reactions. If matter i s measured by the moles of atoms or nucleons present, the quantity of matter is unchanged in all chemical reactions and in many nuclear reactions Paul K Andersen is Associate Professor of Chemical Engineering at New Mexico State University. He received his BS from Brigham Young University and his PhD from the Uni versity of California at Berkeley both in chemi cal engineering His research interests include electrochemical engineering and process simu lation. He is author of Just Enough Unix (McGraw-Hill, 2000) and coauthor of Essen tial C (Oxford 1995). Copyrig ht ChE Division of ASEE 2000 C h e mi c al En g in ee rin g Edu c ati o n

    PAGE 75

    INERTIAL MASS AND ENERGY Of course, Einstein's equation refers to mass not moles It is wort hwhil e to co n sider exactly what is meant by mass. In classical mechanics, mass is a measure of two attribu te s of a material body: I The body 's re s istance to acceleration by external forces (" inertial ma ss") 2. The force the body experiences in a gravitational field ( gravitationa l mass ") The title of Einstein 's 1905 paper 121 clearly s hows that he was interested in the first concept: Does the Inertia of a Body Depend on its Energy Content? (1st die Tragheit eines Korpers vo n sei n em Energienhalt abhangig?) The product of in er tial m ass a nd ve lo city is the momen tum of a body Newton's seco nd law of motion can be written as a momentum balance relating the inertial mas s and velocity to the force exerted on the body F = d(mv ) dt (3) It h as been custo m ary in classical mechanics to regard the mas s of a body as a constant, independent of time or velocity (provided the body is not losing or gaining matter). Thus, the mass is usually taken out of the derivation in Eq. (3) dv F=m-=ma dt Einstein cha ll enged the usual assumption that mass is independent of velocity. Using an argument ba sed on the emission of radiant energy (see the Appendix), he derived a relationsh ip between the kinetic energy and the inertial mas s. He concl ud ed, The mass of a body i s a mea s ure of it s energy content; if the energy changes by L, the ma ss changes in the sa me se n se by L/910 20 if the energy is mea s ured in ergs and the ma ss in grams." In other words, (4) INTERCONVERSION OF MASS AND ENERGY? In view of Eq. (4), wou ld it therefore be accurate to say that mass and energy are interconvertible in reactions? The answer i s sti ll no If mass a nd energy were interconvertible, we wou ld expect a negative s ign to appear in the equation illi=-c 2 L'.m (?) But the ma ss a nd energy increase or decrease together so ~E an d ~m must h ave the same s ign Consider once agai n th e fission of uranium as described by Eq. (2). Suppose the reaction i s carried o ut in a closed adiabatic container, whic h a ll ows no work or heat exchange with the su1Toundings. In that case ~E = 0, and Eq. (4) yie ld s ~m = 0. The reaction occurs without any change in the mass of the system What h appens physically is that so me of the energy stored in the uranium nucleus is co n verted to kinetic energy of the Spring 2000 fission products. The temperature of the system rises; but so long as no energy i s exchanged with the surro undin gs, th e overall energy of the system does not cha n ge. Therefore, according to Eq. (4), the ma ss does not c han ge. On the other hand suppose the thermal energy is with drawn from the sys tem as it i s generated by the fissio n reaction. According to Eq. (4), thi s results in a decrease in the ma ss of the sys tem: ~E < 0 implies ~m < 0. Note that it is the withdrawal of energy from the system that causes the mass to decrease; there is n o co n ve r sio n of ma ss to energy in the reaction itself. Indeed if we were to add energy to the system-such as by heating it or accelerat ing it-the mass would increase again. FORMS OF ENERGY I s ma ss then a form of energy? When speaking of forms of energy, we typically mean kinetic, potential and internal en ergy. The total energy of a system may be taken as th e sum E =EK+ EP + U If mass were s imply another form of e n e r gy, we would ha ve to add another term to the equation E = EK + EP + U + mc 2 (?) This is incorrect. According to Einstein, mass is a measure of the ene r gy of the system, not a separate kind of energy Hence it would be proper to write m=~=~(EK+E p+u) (5) C C Note that the mass varies wit h kinetic energy and therefore with velocity. We shall return to thi s point l ater. CONSERVATION OF MASS AND ENERGY An oft-repeated assertion is that Einstein's specia l th eory of relativity modifies the principles of ma ss and energy conservation. This is o nl y half true. Consider th e ge neral balance equation for an extensive quantity in a co ntrol vo lume: (Rate of accumu l atio n )= (net input rate)+ (net genera tion rate) For the energy, E, of the sys tem the balance eq u a tion takes the form dE r = Ei + E ge n dt where a dot over a variable indicates a rate, and the s umma tion is taken over the boundaries of the co ntr o l volume. In thermodynamics we recognize three ways for energy to cross the boundaries : by h eat transfer, by work interactions and by material flows Therefore the e n ergy balance can be written dE L . -= (Q+W+mE)i +E ge n dt (6) where Q rate of heat transfer th.rough boundary i /6 9

    PAGE 76

    w rate of work through boundary i m rate of material flow through boundary i E energy per unit mas s of material Einstein ass u med t h at energy is conserved-it is neither created nor destroyed. In other words, the energy generation rate is zero: E g en = 0 (Conservation of Energy ) This is the assumption usually made in engineering analysis; t h erefore, our energy balance need not be modified to ac count for the effects of relativity. We do, however, have to modify the usual engineering mass balance: dm ~ . dt = L.J mi+ m g en (7) We normally assume conservation of mass m ge n = 0 ( Conservation of Mas s ) R elativity changes this. Recall that the mass of a system is a measure of its energy content. Dividing the energy balance, Eq. (6), by c 2 and assuming that energy is conserved, we obtain 1 dE ( Q w mE 1 2ctt = ~ l2 +2+7 J i Each term in this eq u ation has dimensions of mass divided by time Moreover, mE I c 2 = riun = m Thus dm ~ ( Q W .I dt= ~ l 2+2+m Ji This equation can be rearranged to produce dm ~ ~ ( Q W I dt= ~ mi+ ~ l 2+ 2J I I (8) Comparing Eqs. (7) and (8) term by term, we conclude that ~ ( cj w1 m ge n = ~ l z+zj (Relativity) i C C i In other words, mass i s generated in the system by heat transfer and work. Of course, 1/c 2 = I x w 1 1 s 2 m 2 is so small that the generation rate is usually negligible in practical problems. RELATIVISTIC AND REST MASS According to Eq (5), the mass varies with velocity. To determine the velocity-dependence of mass, consider a closed adiabatic system initially at rest. Suppose that a force F, accelerates the system in such a way as to leave its potential and internal energy unchanged. The energy balance for the system reduces to dE =W dt Substituting E = mc 2 and w = F. v into this equation we obtain 170 :t(mc 2 )=F v (9) The mass is related to the force by the momentum balance Eq (3) F = d(mv) dt Substituting this into Eq (9), we obtain ~(mc 2 )=v d(mv) dt dt Multiplying by 2m and rearranging yields c 2 dm 2 d(mv )2 dt dt (3) Next we integrate, noting that v = 0 at t = 0. The result is c 2 (m 2 -m l )=(mv) 2 Solving form we obtain at last m = 'Yffi o = m o .J1-v 2 /c 2 (10) In this equation, m i s the inertial mass sometimes called the relati v isti c mass ; and mo i s the r e st mas s, which i s the mass of the s y s tem at v =O At velocities much lower than the s peed of light, y "' I and the relativistic mas s coincides with the rest mass. This is usually the case for engineering problems. CONCLUSION We have seen that Eq (1) Ein s tein s mass-energy equa tion, does not predict the interconversion of matter and en ergy in chemical or nuclear reactions. In fact, the constitu ents of matter are conserved in chemical reactions and in many nuclear reactions. Nor ar e mass and energy interconvertible. In s tead what Einstein s howed wa s that the mass of a body is a measure of its energy content; conse quently, the mass increases when the energy does. Because energy is conserved, there is no need to change our usual energy balance. But in some cases it may be desirable to modify the mas s balance to account for the dependence of mass on energy. We may well ask whether any of this matters, s ince chemi cal engineers rarely if ever encounter problems in which relativi s tic effects are s ignificant. There are at least three reasons why it is important. First, if we are going to mention the theory in our classes or textbook s we should try to get it right. Second our students may in the future have to deal with problems in which a sound understanding of E = mc 2 is In r ece nt ye ars, th e pr e f e r e n ce of m a n y ph ys i c i s t s ha s b ee n t o d e fin e th e r es t ma ss as the ma ss of th e sys t e m and t o dr o p th e s ub sc ript 0. Th e ma ss, m th e n b ec om es ind e p e nd e nt o f ve lo ci t y, which ma y b e consid e r e d an advantag e; o n th e oth e r hand th e factor y must b e in cl ud e d ex plicitly in man y e quation s For a li ve l y di sc u s sion of thi s i ss u e s ee r e f e r e n ces 4 and 5 Thi s pap e r ha s adh e r e d to th e mor e traditional d e fin i tion in whi c h th e ma ss, m i s th e in e rtial or r e lati v isti c ma ss. C h e mi ca l Engin ee rin g Edu ca ti o n

    PAGE 77

    required. Finally, the Special Theory of Relativity is one of the major scientific discoveries of the 20 th century. It could be argued that no scientist, engineer, or mathematician can be truly educated w ith out a proper understanding of this theory REFERENCES 1. Brescia, Frank, et al., Fundam entals of Chemistry Academic Press ( 1966 ) 2. Einstein, A., Does the Intertia of a Body Depend on Its Energy Content?" ( Ist die Triigheit eines Kiirper s von seinem Energienhalt abhiingig? ), Annalen der Physik, 17 ( 1905 ) 3. Einstein, A. On the Electrodynamics of Moving Bodies ( Zur Elektrodynamik bewegter Kiirper ), Annalen der Physik 17 ( 1905 ) 4. Okun, L.B ., The Concept of Mass Ph ysics Toda y, p. 31, June ( 1989 ) 5 Rindler W ., et al., 'Letter s ," Ph ysics Today p 13, May ( 1990 ) 0 APPENDIX: Derivation of L1 E = c 2 Mn --------------------------, Consider two astronauts, Jack and Jill riding their s pace scoo ters far out in interstellar s pace (Space scoo t ers had not been invented when Einstein publi s hed his deri va tion in 1905 but his argument was essentially the sa me as what follows.) Jack is moving away from Jill at a constant ve l ocity v. For the purposes of our analysis, we define two coordinate sys tems as shown in the Figure. The (x,y) -coordinate syste m i s at tached to Jack and moves with him ; the (x *, y*)-coord inate sys tem is attached to Jill. The sys tem s are oriented so that the x-axis and x -axi s are parallel to the direction of v. We want to calculate the energy of Jack and his scoo ter. To do so, we must s pecify which coord in ate sys tem we have in mind Relative to Jill' s (x*, y ) sys tem J ac k i s moving at speed v, giving him a kinetic energy mv 2 Relative to his own (x,y) system Jack is not moving so he h as no kinetic energy. ln eit h er system, Jack and his scooter have the same internal energy. Thus, the difference between Jill's v iew and Jack's view is E -E-.Lmv 2 2 (Al) Now suppose Jack activates hi s la se r beacon which fires two pul ses of light. One pulse ha s energy L/2 and i s emitted at an angle 8 relative to the x-axis; the other also ha s energy L/2 but is emitted in the opposite direction .* Jack 's velocity doe s not change, but the internal energy of Jack and hi s scoo ter decreases by the s um of the energies of the li ght pul ses ( L L I E 2 -E =6E=l 2+2 ) =-L (A2) Jill once again sees things differ e ntl y. As Einstein s howed in a previous paper on re l ativity 131 the energies of the light pulse s appear from Jill's standpoint to be ( I _!:: l l + vcos8 j 2 and ( I _!::l l-vcos8 j 2 Therefore Jill computes the change in Jack s energy to be L l + V cos e + L l V cos e = L l 2 2 (A3) y y V Subtracting Eq. (A2) from Eq. (A3), we obtain R egro uping the terms on the left-hand side of the equation yie ld s Referring back to Eq. (A l ), we see that the left-h a nd side of the e quation equals the change in kinetic energy of Jack and his scooter relati ve to Jill 's (x *, y*) system. Moreover Eq (A2) shows that -L = 6E Thus we can rewrite Eq. ( A4) as (AS) Einstein made u se of the approximation Substituting this into Eq (AS), we obtain 6( mv 2 ) = 6E [ ( r] (A6) But if Jack 's ve locity does not change, 6( mv 2 ) = v 2 6m and Eq. (A6) become s 6m = 6E/ c 2 Thi s i s the result Einstein obtained in 1905. (A7) Why two laser pulses? As Einst ein noted light carries momentum If Ja ck fir e d only one pulse, it would tend to acc e l e rat e him in th e direction opposit e th e direction of th e light By using two equal but oppo si t e puls es, there would b e no acceleration. Sprin g 2000 1 7 1

    PAGE 78

    ,a--aa~r:,a:=-b:--:o~r.=a=to=ry~--------..) A LABORATORY FOR GASEOUS DIFFUSION THROUGH PERMEABLE SOLIDS The Time Lag OL IVIER DUFA UD, ERIC FAVRE, LOUIS MARIE VINCENT Ecole Nationale Superieure des Industries Chimiques 54001 Nancy France M ass transfer one of the core areas of classical chemical engineering curricula, is most often pre sented to students through laboratory tutorials dedi cated to unit operations (e g ., distillation, gas absorption, extraction etc.). Laboratory tutorials aimed at a more funda mental approach can also be proposed based for instance on the study of one of the methods leading to diffusion coeffi cient determination in fluid phases (liquid gases, etc)f1 1 To our knowledge however the study of a strict diffusion pro cess in a less conventional phase, such as a permeable solid (po l ymer, adsorbent, microporous material), is seldom at tempted at an undergraduate level in chemical engineering departments. In this paper we will describe an experimental setup that enables study of the transitory mass transfer of a permanent gas through a permeable solid; apart from the simplicity and rapidity argument s, thi s technique, currently referred to as the time-lag method, permits us to stress either the m~ss-transfer or the physical-chemistry aspect in the solution approach proposed to the student. THEORY The time-lag permeation technique was originally con ceived in 1920 by Daynes in order to study the mass transfer through an elastomeric material Y 1 The method was refined and widely used through the years by authors such as Barrerl 3 1 and Crank ,l2 1 and it has been applied successfully to numer ous materials-from catalyst particles to metals and poly mers-and to different sample geometries. Interest in the time-lag technique has been sustained over the pa s t ten years, as shown by VieM 41 in hi s work on permeation through polymer films. A time-lag cell consists of an upper and lower chamber separated by an initially gas-free solid s ample ( s ee Figure 1) A permanent gas is introduced in the upstream part of the cell at time zero; while maintaining upstream pressure con stant the appearance of permeated ga s i s continuously moni tored in the lower compartment using a pressure transducer. A significant pressure rise only occurs after a period called the time lag This time lag, 0 indicates the onset of a quasi steady-state diffusion process which persists until the pres sure in the entire cell equilibrates Two experimental configurations can be used to perform a time-lag experiment. The first one, called the Wicke0. Dufaud received his Master s Degree in Chemical Engineering at ENSIC Nancy (France) and is now a PhD student in the field of stereophotolithography E Favre is a professor of chemical engineer ing at ENSIC Nancy (France). His major inter ests in research are mass transfer and mem brane separations L.M. Vincent is a research engineer at LSGC-CNRS in data acquisition and electronic applications (no photo available) Copyrig ht ChE Division of ASEE 2000 1 7 2 C h e mi c al En g in ee rin g Edu c ati o n

    PAGE 79

    Kallenbach" diffusion cell experiment, consists of two com partments initially containing the same inert gas at the same constant pressure and separated by the solid film to be tested At time zero, another compound is introduced in the upper chamber and the response to this impulse is monitored in the downstream compartment. As part of our study, we use another configuration termed the "traditional" time-lag experiment, which differs from the previou s one in that a vacuum is initially applied in both compartments. Rutherford has discussed the advantages and drawbacks of each method. 1 5 1 A theoretical computation of downstream-compartment pressure increase can be obtained starting from Fick's sec ond law applied between the permeable sample boundaries l 21 (I) It should be stressed that Eq (1) applies only for a constant diffusion coefficient, which is a special case; the didactic system selected for this study (oxygen/nitrogen/silicone rub ber) i s in agreement with this assumption. Frisch extensively discussed the more complex general situation of a concen tration-dependent diffusion coefficient. 1 6 1 In this case, Eq. (1) has to be rewritten as ~[D(c/c]= ac ac at at (2) For a flat sample the following boundary conditions can be postulated: a film initially free from gas, the attainment of equilibrium at the inlet gas-polymer interface according to a Henry-type expression, and a near-zero concentration of gas at the downstream face: c(x,0)=0 c(O t)= Co= SPo c(L t)=cL =0 (3a) (3b) (3c) where L is the sa mple thickness, P 0 the upstream pressure and S ( deriving from Henry s law expression) is usually called sorption coefficient. Again Henry 's law validity cor responds to a simple and special case; it is important to note that numerous systems, especially those involving glassy polymers (e.g polyethyleneterephthalate used in carbon ated beverage packaging) show strong deviations from Henry 's law. Nevertheless the assumption of a constant sorption coefficient is correct for the system selected for this study. Equation (1), subject to the experimental boundary condi tions Eq. (3), can be integrated by Laplace transforrn l 2 J Spring 2000 where n is an integer. The net gas flowrate can be computed from the concentra tion profile based on the integration of Fick's first law with respect to time. The resulting downstream pressure increase is P A RTDP 0 [s SL 2 2SL 2 (-1)"+ 1 ( -Dn 2 rc 2 t I ] L = t6D + rc 2 D L..i -n2 -exp l L2 j n = I (5) When a quasi-steady-state prevails the transient summation terms are negligible and an asymptotic solution is reached P 2 2 Transient s tate Quasi steady Towards state equilibrium ,c ---:>< E----X: ,-----------, ~.__ ___ __,, 'P Time lag 8 Figure 1. Schematics of a time-lag apparatus and experimen t. for the downstream pre ss ure PL P ( ) A RTSD P 0 ( L 2 I Lt= VL l t-6D ) (6) This equation reveals that the pressure-time plot shows a linear rise and allows determination of the following param eters (see Figure 1 ): L 2 8=6D a= A DRTS P 0 VL known from the intercept (7) known from the asymptotic slope (8) In other words the above analy s is shows that an interpre tation of the early events of a time-lag experiment allows simultaneous determination of the three main quantities char acterizing mass transfer: the diffusion coefficient (D), the Henry law sorption coefficient of the gas in the solid (S), and the product of both, usually called permeability, ,!O=SD (9) 1 73

    PAGE 80

    The objective of the laboratory tutorial described in this paper is to show and analyze the possibilities and limitations of a time-lag ex periment performed according to the preceding data treatment. MATERIAL AND METHODS The experimental apparatus we set up consists of a stainless-steel permeation cell (see Figure 2), within which a circular polydimethylsiloxane film purchased from Dow Corning (Silastic sheeting, thickness 125 diameter 8 cm active surface area 30 cm 2 ) is inserted. The film is mechanically supported by a brass or Inox frit (Poral) sufficiently porous to assert that its mass-transfer resistance is far lower than the one of the analyzed material. A Viton T M seal ensures the airtightness of the cell and also reduces the film surface area exposed to the gas. An overview of the complete laboratory setup designed in our workshop is shown in Figure 3. For the sake of convenience, a simple low-cost stainless-steel filter holder can be purchased (for instance, Millipore model 4404700) and works equally well in this application The two compartments of the cell are connected to a vacuum pump (Alcatel Pascal 1015 SD) while the upper one can be fed by a permanent gas (nitrogen or oxygen) from a bottle by opening a needle valve (Nupro ABVT 1). The upstream compartment pressure is con trolled by a Bourdon manometer. An active strain gauge (Edwards ASG NW16 2000 mbar) enables downstream compartment pressure to be monitored. The electric signal ranging from Oto 10 volts is sent to a digital display (Edwards ADD). Afterwards, an analog-to-digital 24-bit converter (Sigma Delta) with an integrated oscillator (Linear Technology L TC2400) is used to allow a computerized acquisition at a maximum frequency of 20 Hz. A personal computer with the Test Point program carries out the analog data processing. It permits screen display of the downstream pressure rise in mil libars versus time in seconds and the information backup in a data extended file. This file is then imported into a spreadsheet pro gram (such as Excel) to de termine the diffusion and sorp tion coefficient. Figure 4 shows an example of the com plete pressure rise for illustrative purposes. LABORATORY TUTORIAL After a short review of the principles and the theoretical de velopment underlying the tech nique (see the section on "Theory"), we ask the students to familiarize themselves with the equipment and then operate 174 Upstream compartment 0-ring Slot Viton seal Solid sample Frit Figure 2. Exploded view of the permeation ce ll. 13 II 1 2 14 Figure 3. Overall set up : 1. Gas bottle (pure oxygen or nitrogen] 2. Bourdon manome ter 3 Thermoregulated bath 4. Solvent res e rvoir 5. H e ating r e sistance 6. Pressure gauge and digital displa y 7 Magnetic stirr e t 8. Pump 9 Solid sampl e 10 Porous support 11 Downstream volume 12 Thermoregulated w ater flow 13. Thermore g ulated water flow 14. Computerized data acquisition. Ch e mi c al En g in ee rin g Edu c ation

    PAGE 81

    it according to the following in st ruction s: G) Start thermostated bath and apply a primary vacuum in both co mpartment s. Attention s hould be paid at this s tag e to the fact that the down s tream s ide only ha s to be deg assed in a first ste p before tr ea ting the up stream part. If not film disruption will occur as soo n as a reverse pr ess ure differen tial take s pla ce ( th e sa mple i s not s upported on the up s tr ea m s ide ). @Ev a luate the cell's leakage rate by clo si n g all co nnec tion s around the module (vacuum pump gas bottle s, etc .). Thi s i s of crucial importance s ince any back gro und in do w s tream pre ss ure increa se will affect d a ta treatment. An aver age value of about 0.1 millibar per hour s hould be achiev able ; if not cell sc rew s s hould be tightened as we ll as the 0ring and sa mple po s itioning checked. The above va lue s hould be kept in mind if error calculation on D or S are needed Neverthele ss, it i s ne g ligible in regard to PDMS permeabil it y and experiment duration in the pre se nt case. (1) A typical time-l ag experiment can then be undertaken for a g iven pre ss ure and temperature. D epe ndin g on th e duration of the tutori a l 2 to 4 diff ere nt pr ess ur e and/or temperature va lues ( typically ranging b e tween 20 and 80 C and 0.5 to 5 bar ) can be explored with pure nitrogen and oxygen The st udent is asked to assess D and S va lue s a nd check the consistency with literature dat a_P91 The key im portance of downstream co mpartment vo lum e V can be di s cussed at thi s s tage (see Eq 8); in our case dir ec t as well as indirect mea s urement s credit thi s chamber with a 585 cm 3 vo lum e, which fits the need s regardin g ga u ge se n s iti vi t y air lo sses, a nd mat er ial permeability. 0 8 10000 20000 t (s) Fig u re 4. E xa mple of an experimental result obtained b y th e se tup d escr ib ed in this work ( downstream pressure vs. time) including the d e terminati on of time-lag int ercep t (e) Oxygen permeation throu g h a 125-m thi ck Silastic film; upstream pressure 4 bar temperature 50 C acquisi tion freq uen cy 5 H z. Spring 2000 An error calculation ca n be optionally performed ba se d on an analysis already di sc ussed by Paul and DiBenedetto 1 1 0 1 and l a ter by Si ege l and Coughlin ci i I showing that the technique can lead to s mall errors in permeability but far greater errors in the diffusion coefficient. For in sta n ce, the relative error on e i s about five time s larger than that o n the s lop e for d ata collected at t = 4 0 : (I 0) Thu s, a 2 % error on the slope mean s a 10 % error on the diffu s ion coefficient-which are commonly assumed values for a si ngle mea s ur e ment. Therefore, the duration of the time-lag experiment s hould be carefully considered. While data taken at time s grea ter than 10 0 will lead to sizeab l e errors in diffu s ivity the attainment of steady s t a te requires durati o n times around 4 0 Nevertheless the erro r in s teadys tate s lope can be greatly minimized by si mpl y r e -evacuating the down s tream cham ber wi th the feed pre ss ure s till a pplied to the top face of the m e mbrane Reclosing the receiver valve ahead of the vacuum pump ( item 8 in Figure 3) allows mea s uring the steady-state s lop e without bend-over problems up to a downstream pres s ure below 1 % of the feed pre ss ure This repeated evacua tion a nd retesting of the s lope gives s tudents an easy way to eva lu ate mea s ur e m e nt reproducibility G) B ase d on the D a nd S va lue s, p ermea bility p can then be calculated from Eq. (9). The student is asked to express the results in Barrer s, the mo st common permeability unit: p( B arrerl010)= 273 v(cm 3 ) L(cm) a(cmHgs 1)_!_ T(K) Pa(cmHg) A(cm 2 ) 76 6.7 .5 ... I[ 6.3 ~6.1 :c ill 5.9 E ; 5 7 II.. 5.5 5.3 0.0028 0.0 0 29 0.003 0 0031 0. 003 2 0. 003 3 0 .0 034 0,00 35 1/Temperature ( K ) ( 11) Fig u re 5. Example of Arr h enius plot s for oxygen and nitro gen permeability varia tion w ith temperature through Silastic film (upstream pressure 2 bar). Comparison be tween experimenta l (so lid lin e) and lit era tur e (dashed line) values. 1 75

    PAGE 82

    Typical oxygen permeability values in PDMS obtained by the setup can be seen in Figure 5 where the good reproduc ibility of measurement is illustrated. @ For a more physicochemical insight, undertaking the st udy of the influence of temperature on sorption, diffu sion, and permeability can be worth the time. A s imple expression i s often shown to describe correctly the results for all three parameters : X=X 0 ex{-!;) X=D,S,or p (12) While the analogy to Arrhenius expression is correct for rate parameters (leading to the often-used vocable activation en ergy of diffusion and activation energy of permeability), it should be explained to s tudents that the same is incorrect for sorption, which is not an activated proces s (Es corresponds to the heat of sorption t.H s, as classically obtained by a van't Hoff plot) The following relationship, linking the two acti vation energies Ep and E 0 with t.H s can be derived by using Eqs (9) and (12) (13) Figure 5 illustrates the good accordance between the experi mental and the literature values for oxygen and nitrogen permeabilities 1 7 12 1 31 Ei 2 =7.4kJ/mol p~ 2 =9180Barrer E~ 2 = 8 2 kJ / mol p~ 2 = 7150Barrer These results are satisfactory in that the oxygen permeability is greater than the one of nitrogen and conversely for the energies of activation. At thi s stage, students can be asked to explore potential separation applications ba se d on thi s pecu liarity, such as gas-separation membrane processes. Typical experimental results are summarized in Table 1. The already-discussed inaccuracies on the intercept lead to significant errors on D values; however D~ 2 remains al ways greater than D~ 2 which agrees with the theory be cause nitrogen's kinetic diameter is larger than oxygen 's (z) The influence of upstream pres s ure can also be option ally investigated, similar to the work already addressed by Koros and Jordan with a silicone-nitrogen system.L 141 CONCLUSION The objective of this work was to point out an easy-to carry-out experiment of didactical importance for the under standing of gas mass transfer in different solid media. The time-lag permeation method is a flexible and powerful tech nique that can give both equilibrium (so rption coefficient S) and transport properties (diffusivity D and permeability go) 176 TA B LE 1 Ex p erimental (exp) and Literature Values ([7]) for Oxygen So r ption, Diff u sion, and Permeability through PDMS for I Bar Upstream Pressure T( C) p p D D s s exp literature exp literature exp literature (Barrer ) ( Barrer) ( m 's ') ( m's ') (Pa') ( Pa') 20 454 479 I.0E-09 l .6E-09 0.34 0.31 30 492 539 7. IE-10 I .8E-09 0.53 0.31 40 574 601 I. I E-09 2.0E-09 0.41 0.30 50 605 667 l.6E-09 2.2E-09 0.29 0.30 70 696 806 2. 1 E-09 2.7E-09 0.25 0.30 in a single experiment; based on the se tup de scri bed in this work, we have shown that reliable and accurate data with regard to the permeabilities of permanent gases can be ob tained, while estimation of D and S is achievable. We fo cused first on simplicity, especially regarding the data-treat ment aspects. We want to stress, however that more so phis ticated approaches (s uch as those propo se d in advanced mass transfer topic s) can be equally well-proposed based on the same setup. A few examples, listed below s how how to open various didactical extensions. A first possibility consists in replacing the silicone rubber membrane with a glassy polymer For instanc e, poly et hyleneterephthalate ( PET) which is readily available since most overhead transparency materials used in proje cto rs are composed of PET ; a more permeable Lexan film produced by General Electric, can also be used. If carbon dioxide is used in plac e of oxygen or nitro gen, exper iment duration would r emain compatible with a laboratory tutorial providing that a thin enoug h film (25m or less) is available. Investigatin g the more-complex case of gas permeation through glassy pol yme rs is of value for the students in order to point out polymer barrier properties. In place of permanent gases, organic-vapors transport cou ld be equally well investigated, based on the vapor ge nerator system con nect ed to the module (s ee Figure 3); in that case, complications arising from the non-constanc y of the diffusion coeffic i ent with conce ntration can lead to comp lex but interesting transport beha vior ( and thus, data treatment). The incidence of a variable D on the experimental time lag has been ex plored by Frisch. 161 Attention should b e paid, however to exp losion hazards or O-ring damage when manipu lating organic vapors Chemica l Engineering Education

    PAGE 83

    In this work, the solid sample is co nsid ered as the si ngle mass-transfer r esis tan ce; nevertheless, boundary-layer resistanc e can arise, particularly when binary mixtures are transported through a very permeable media; in that event, the influen ce of h y drod y namic conditions on overall transfer ( concentration polarization phenomenon) is a good indication that can be best achieved by a magnetic Rushton turbine already existing on the setup The independent temperature ja c kets for the two co mpartments also offer an opportunity to experi ment with the incidence of non-isothermal co ndi tions, alread y shown to strongly affect the observed transfer rate of a pure organic vaporJ 1 51 ACKNOWLEDGMENTS The authors gratefully acknowledge the numerous valu able comments of the manuscript's reviewer s. NOMENCLA T URE A samp l e s urface area a asymptote s l ope c concentration c 0 up stream concentration cL downstream co n centrat i o n D diffusion coefficient E 0 _P energy of activation for diffu s ion ( D) and permeability ( P ) L sample thickne ss P pressure p permeability R perfect gas cons t a nt S sorp ti on coefficient t time T temperature V downstream volume Greek Symbols 0 time la g t.H heat of so rption REFERENCES 1. Cussler, E.L., Diffusion: Mass Transf er in Fluid M edia, Cambridge University Press ( 1984 ) 2. Crank, J ., Th e Math e matics of Diffusion Oxford Science Publications ( 1975 ) 3. Barrer, R.M ., and G Skirrow, Transport a nd Equilibrium Phenomena in Gas -El astomer Systems. I. Kinetic Phenom ena," J. of Poly Sci ., 3 549 ( 1984 ) 4 Vieth, W.R., Diffusion In and Through Pol y m e rs Prin ci pl es and Applications Hanser Oxford University Pres s ( 1991 ) 5 Rutherford, S.W., Review of Time-Lag Permeation Tech nique as a Method for Characterization of Porous Media and Membranes ," Adsorption, 3, 283 ( 1997 ) 6. Frisch, H.L ., The Time-Lag in Diffusion I. J. of Ph ys. Chem ., 62 93 ( 1957 ) 7. Brandrup, J ., and E.H. Immergut, Pol y m er Handbook 3rd ed., Wiley Interscience ( 1980 ) 8. Van Krevelen D.W. Prop e rt ies of Pol ymers: Their Correla tion with Chemical Structure, Th e ir Numerical Estimation Spring 2000 and Pr ediction from Additive Group Contributions, 3rd ed., El sevier Science Edition ( 1990 ) 9. Robb W.L., Thin Silicone Membranes: Their Permeation Properties and Some Applications," Annal s. of New York Acad. Sci. 146 119 ( 1968 ) 10 Paul D ., and A.T. DiBenedetto "Diffusion in Amorphous Polymer s," J of Pol y Sci ., Cl 0, 17 ( 1965 ) 11. Siegel R.D ., and R.W Coughlin, Errors in Diffusivity as Deduced from Permeation Experiments with the Time-Lag Technique ," AIChE J. Symp. Series, 120 68 ( 1986) 12. Barr e r R.M. and H.T Chio, Solution and Diffu sion of Ga ses and Vapors in Silicone Rubber Membranes ," J. of Pol y. Sci ., ClO, 111 ( 1965 ) 13. Yasuda, H ., and K. Rosengren Isobaric Measurement of Gas Perme ab ilit y of Polymers ," J of Appl Poly. Sci. 14 2839 ( 1970 ) 14 Jordan S.M., and W.J. Koros "Permeability of Pur e and Mixed Gases in Silicone Rubber at Elevated Pressures ," J. of Pol y. Sci., B28 795 ( 1990 ) 15 Hillaire A ., and E. Favre, Isothermal and Non-Isothermal Permeation ofan Organic Vapor Through a Dense Polymer Membrane ," Ind. & Engg. Chem. R es., 38 ( 1 ), 211 ( 1999 ) 0 Letter to the Editor Continued from page 16 7. three undergraduates. Used in thjs way, a lecture course provides a hjghly effec tive way not only for the di sse mjnation of information but also for capturing the interest of st udent s. The formal l ec tures does not provide a good format for developing prob lem-solving ski lls for dealing with engineering design or even for presenting and discussing solutions to pre-assigned problem s. Unfortunate l y, in many (if n ot most) unjv ersities the l ec ture format has been widely misused si nc e it h as become the universal workhorse. Thjs may be a more serious issue in engineering education where "design" and problem so lv in g" constitute a major portion of the curriculum Neverthe le ss, withjn the chemjcal engineeri n g curriculum, there are man y s ubject areas that are well-s uited to the lectur e ap proach and, in the hand s of a s killed practitioner and espe cially if s upported by appropriate tutorial sess i o n s, this ap proach ca n be very effective. Essentially this sa me point is made by Wankat and Oreovicz in Teaching Engineering Thi s is one of the references cited in the present article as showing the superiority of alternative approaches! Such a conclusion is hardly s urpri s in g since, in any attempt at a quantitative assessme nt it would be very difficult, if not impossible to establish whether the apparent disadvantages of a lecture course are really intrinsic to the format or ste m from an inappropriate applicatio n of thjs format. There seems to be a c l ear danger that in the curre nt ent hu siasm for new instructional methods, the very real advantages (a nd equally real limitati ons) of the l ecture format will be overlooked. Douglas M. Ruthven University of Maine 1 77

    PAGE 84

    .(.3 111 ij_._1_a_b_o_r._a_t_o_r_:y ________ ) USE OF AN EMISSION ANALYZER TO DEMONSTRATE BASIC PRINCIPLES KEITH B L ODGE Universit y of Minnesota Duluth, MN 55812-2496 I n c hoosing and developing experiments for und ergradu ate laboratorie s, instructors often try to avoid those ex periments that require len g thy or sophisticated chemical analy s i s s uch as experiments related to environmental engi neerin g. But the development of electrochemica l and in frared s ensors ha s made gas ana l ys i s much easier to accomplish Thi s paper de sc ribe s experiments ba se d on the u se of a portable stack gas analyzer, or emission analyzer. In a se n se, the analyzer is modem technology 's replacement for the cla ss ical Or sa t ana l ysis of flue gases;[ll but the analyzer doe s not mea s ure carbon dioxide l evels directly, calculating them instead from the knowledge of fuel type u sing ba sic s toichiometry The modern analyzer pro v ide s a relativel y quick and painle ss way of doing chemical analysis for a gaseo u s system The a nalyzer used here costs $ 6 ,3 50 and was purchased under an NSF ILi grantY 1 With the additional purchase of other more mode s t components for about $250, s imple com bustion experiments with a laboratory burner and a make shift flue were run The result s were used to demon s trate: 1 ) the principles of a rotameter, 2) material balance s ba se d on the measured molar flow of the fuel gas and the measured volume fraction of oxygen in the exiting flue gases, 3) energy balances ba se d on the temperature of the exiting flue gases, and 4) how the kinetics of nitric oxide forma tion can be used to estimate temperature s in the hott est part of the flame The c hemical engi n eering program at the University of Minnesota, Duluth, is relatively small, with five full-time faculty, three temporary instructors, and a grad uatin g cla ss in the range of 20-30 students Without the amenities of lar ger departments (e.g., shop facilities), it is difficult to develop experiments requiring considerable co n struction of apparatus or faculty time. The experiment described here i s relatively straightforward to do-the analyzer does the hard work. It i s intended as an undergraduate experiment, how ever, and s hould not be construed as a research-level experiment or an experiment with which to do thorough studies of combustion The analyzer is se lf-contained in a carryi ng case the size of a large brief case so storage of the experiment s components i s not a problem The department also offer s a minor in environmental engineering and th e intention i s to u se this anal yze r in a n ew environmental-engineering laborator y cour se; other experiments for thi s co urse have been de scribed e l sewhere .r3 1 A g roup of our students u se d the analyzer as part of their project in our capstone de s ign course and the project was successfully entered in a na tional competition. [ 4 l EXPERIMENTAL The apparatus comprises a section of double-walled ga vanized flue pipe (18 in long and 3 in wide) mounted verti cally with a flue cap po s itioned on top ; the se items were purcha se d from a hardware store A Veriflow burner, fitted with an N-2 nozzle (Fisher Scientific ) was po s itioned at the center of the bottom opening of the flue ; it was connected via a rotameter ( Cole Parmer E-32461-58), calibrated in liter s of air per minute ( L air/min), to the laboratory supply of natural gas To obtain temperature s in the exiting flue gas that vary with the flow of natural gas, controlled by the rotameter it was essential to restrict the flow of air into the bottom of the flue A can of suitable dimensions was cut in Keith Lodge is Assistant Professor of Chemi cal Engineering at the University of Minne sota in Duluth. He was educated in the United Kingdom obtaining his BSc from the Univer sity of Warwick and his PhD from the Univer sity of Sheffield He teaches laborator y courses thermodynamics heat transfer com putational methods reactor design and pro cess control Properties of hydrophobic or ganic compounds are his principal research interest Co p yr i g h t C h E Di v i sio n of ASEE 2000 1 78 Ch e m i c al En g in e er i ng Edu c at i o n

    PAGE 85

    half, and a 1-in square hole was punched in the bottom of one half. Thi s half was fitted into the bottom of the flue and the burner was po si tioned at the center of the hole The rotameter was calibrated with a household gas meter (Equipmeter S-275) For the experiments described here, air was not premixed with natural gas before entering the burner so the combustion was controlled by diffusion of air ,15 1 The emission analyzer is a COSA 6000 portable s tack gas analyzer. This measures the volume fractions of 0 2 CO NO x, and SO x using electrochemical cells. (The lifetime of the 0 2 cell is the most limited being in the range of 1-2 years; the other cells will last between 2 and 3 years.) The analyzer pumps the gas to be sampled through the end of a hollow metal probe (12 in long ) at the tip of which is a thermocouple (type K) for measuring the gas temperature. The gas passes from the metal probe via flexible tubing (10 ft long) to a condensate trap clamped to the outside of the main case containing the analytical components of the ana lyzer. This passage of the gas to the analyzer re s ults in the gas entering the train of electrochemical cells at essentially ambient temperature. From the measurements of gas le ve l s and temperature, the in s trument's firmware calculates vario u s quantities The vo ume fraction of CO 2 and the excess air are of mo s t concern here. This require s the user to select the fuel type ; the selec tion includes natural gas, propane butane coal oils light no 2 and no. 6, as well as a programmable option. Other calcu lated quantities include efficiency and effectiveness ; the se are not considered here. For the experiments described here the tip of the probe was positioned just below the center of the flue cap. Two to four sets of data for each of six different flows of natural gas were collected. It was important to let the rotameter setting s tabilize after changing it. The data pre se nted here were comfortably obtained within 3 to 4 hours; the treatment of the data may take most students a lot longer! RESULTS AND INTERPRETATION Calibration of the Rotameter Volumetric flows were measured under ambient conditions (25 C and I atm) through out. The volumetric flow rate of natural gas, V n g read di rectly from the hou se hold gas meter in unit s of ft 3 /min, regre sse d against the reading of the rotameter, V air> in unit s of Umin leads to V n g (Ft 3 I min )=(0.0504 0.00OS)Yair(L/ min) r 2 =0.987 ( I) This is transformed to Yng(L / min)=(l .43.0l)Yair(L/ min) (2) What useful information can we obtain from the value of the s lope? The theory of the rotameter 16 7 1 applied to gases lead s to Spring 2000 (3) Using the density of air at 25 C and 1 atm, the density of our natural gas is 0.579 g/L under the sa me conditions. Methane i s usually the major component of natural gas, and its den s it y181 under the same conditions is 0.657 g/L. Such discrep ancies can cause concern, but it s hould be realized that the composition of natural gas i s dependent on its source and upon the life of the well from which it i s extracted. For example, the analysis of natural gas from 18 locations 1 91 in the U.S. yields the following ranges and median values for the following components: CH 4 98-24 %, 85%; C 2 H 6 700 %, 8 %; N 2 8-0 %, l % ; and CO 2 25-0%, l %. Stoichiometry and Material Balances The emission a nalyzer provide s a measure of the oxygen in the gas sample as a volume fraction ; with the ideal-gas assumption, here and throughout thi s corresponds to a mole fraction, Y o 2 From this mea s urement the instrument calculates the vol ume fraction of carbon dioxide and the fraction of excess air. For the complete combustion of a general hydrocarbon, CmHn, in excess air, represented by the stoichiometric coefficient x the reaction i s (4 m+n J n CmHn +~4 -+x 2 mC0 2 + 2 H20+x0 2 (4) The stoichiometry i s s ummarized in Table 1 for various bases, and in Table 2 the corresponding molar flows and mole fractions in the exiting flue gas are listed. In under sta nding the operation of the gas analyzer, it is recognized that various quantities are calculated from the measured mole fraction of oxygen; these are the mole frac tion of carbon dioxide and the excess air. The ba s ic proce dure is to get an expression for the sto ichiometric coefficient of excess air x in term s of the mole fraction of oxygen (see Table 2, Eqs. Tl and T4 ). Then the mole fraction of carbon dioxide is calculated (see Table 2, Eq. T3). The fraction of excess air is u s ually defined 1 101 by ( moles of) ( mole s of air required ) air fed for comp l ete combustion 4 x Erur = --moles of air required 4 m + n (5) for comp l ete co mbu st ion To apply the appropriate sto ichiometry two cases under which the analyzer operates are identified. For Case 1 it is recognized that the sample strea m passes through a conden sate trap at room temperature When the quantity excess air varies from zero to a value x = X cri,, the water produced will condense to form liquid ; thi s is treated as vapor and liquid in equilibrium Then, quantities applicable to the wet-product stoic hiom etry from Tables 1 a nd 2 are u se d Upon increasing the excess air above x c ri,, there is no longer sufficient water vapor in the product stream to maintain liquid and vapor in equilibrium; this is Case 2, in which all the product s are gaseous, and the gaseous-product stoichiometry is applied 1 79

    PAGE 86

    The va lu e of x c ,i, is calculated by setting the vapor-p h ase mole fraction of water, on the gaseo u s-product basis to its sa turation value of The solution is ( 1 l[ n n (4m+n) l x=xcrit =l-1+--)-2 sa t -m-24 rN/0 (7) rN/0 YH 2 0 If the products from the combustion of methane are passed through the condensate trap at 25C, the value of X c rit is 11.2 mo! (corresponding to about 560% excess air). Thi s value is exceeded in the experiments described here. The dry product stoichiometry is important for und erstand ing the relationships between quantities measured by the gas analyzer and those measured by the traditional Orsat a n a l sis; the mole fractions of gases in the Orsat analys i s are on a dry basi s.P 101 When x < Xcn, then Yo 2 (gas ana ly zer)= ( I YH~O )y o 2 ( Orsat) (8) At 25 C, I y~ 1 0 = 0.97, so the difference in the mole frac tions between the two methods is not great. For the other case, when X > X cr it y 0 2 (gas analyzer)= y 0 2 ( Orsat) /[1 + t 2 )] (9) '/ q+x l+rN / O The quantity q is defined in Table 2. If methane is burned in 600% excess air Uust above the critical va lu e), then TABLE 1 Yo 2 (gas analyzer)= 0.97y O 2 ( Orsat). As the excess air increases further the two measure s of the mole fraction get closer together. In principle the calibration of the rotameter can be u sed to calculate the molar flow of fuel to the burner if the composi tion of the natural gas is known To keep things simple, the n at u ral gas i s treated as pure methane following the usual practice .[1J In this case, the molar flow of methane into the burner at 25 C, as a function of the rotameter se tting is ( ,..JPairPCH 4 -2 Fo mo) CH 4 / mm)=----Y 3 ;, =5.50x 10 Y 8 ;,(LI mm) 16 .043 (IO) This is used in calculations that follow. The mole fractions of oxygen and carbon dioxide and the excess air are plotted against the volumetric flow rate of natural gas in Fig ur e 1 As the flow of n atural gas to the burner increases, the excess air and the fraction of oxygen decrease while the fraction of carbon dioxide increases with th e increasing flow. The largest fraction of carbon dioxide (7%) observed h ere corresponds to the smallest fraction of excess air (60 %) The maximum fraction of carbon dioxide possible is 11 .7% a nd occurs under stoichiometric condi tions (no excess air); the s tudent can deduce this from Eq. (T3) in Table 2 The student can calculate the maximum fractions of carbon dioxide possible for other hydrocarbon s in a si milar way; a standard reference contains the values. 1 81 Energy Balance The first law of thermodynamics for steady-flow syste ms i s ap plicable. There is no shaf t work 800 Stoichiometry for the Co mplete Combustion of a H y drocarbon 20 O N Mo lar flow into Molar flow out of the flue the bu mer a11d gaseous -pr od u ct dry-product wit h liquid water as S p ecies the flue basis basis product, wet basi s CmH,, F o 0 0 0 0 2 ( 4m+n ) -4+x Fo xF 0 xF 0 xF 0 N 2 (4 m+n ) -4-+ x 'Nt0Fo ( 4m+n ) 4 -+x 'N 1 0Fo (4 m+n ) 4 -+x 'N 1 0Fo ( 4m+n ) -4+x 'N!OFO co 2 0 mF O mF O mF 0 n y~~o F-;."' H p 0 2Fo 0 I F O is the molar flow of hydrocarbon into the burner. 2. r N,o is the molar ratio of nitrogen t o oxygen in a ir r N,o = 79.0 1/20 .99 3. F,;."et is the total molar flow o ut of the flue. 4. y~;o i s the mole fraction of water in the va por phase when liquid and vapor are in eq uilibrium. 1 80 () c5 600 O N 15 m 0 )< 0 ., (I) a, 400 ll: "' 10 c ., #., f;J c.. ., 5 !ii'~.v 200 E ::, 0 > ,;;;' 0 0 0 2 4 6 Natural Gas Flow, Umin Figure 1. The measured mole frac tion of oxygen ( ) in the ex itin g flue gas as a function of the volu metric flow of natural gas to the burner measured under ambient con ditions (25 G, 1 atm). The quantities ca l cu lat ed from measured mole frac tion of oxygen are the mole fraction of carbon dioxide ( \J ) and the per centage of excess air ( I::,, ). Chemical En g in ee rin g Education

    PAGE 87

    and it i s assumed that kin e tic energy and potential energy effects are negligible ~H=Q ( I l) The flue pipe is regarded as an open sys tem; ~H i s the enthalpy change from inlet to outlet, and Q is the heat tran s ferred between the sys tem and the s urrounding s. B e cause calculation of the a diabatic flame temperature Q=O i s s tandard for thi s sys tem only the bar e detail s are g iven h ere Many authors[ 10 14 l pre se nt method s for calc ulation of th e adiabatic flame temp era tures. In calculating them here th e molar heat capacity functions of Smith and coauthors 1 14 1 were used and Visual Ba s ic fun c tions for them within an Excel spreadsheet were written "Goal Seek wa s u se d to so lve for the flame temperature s. The re s ulting adiabatic flame t e mperature s and the me as ured flue gas temp era ture s are plotted again st the volumetric flow rate of natu ral gas in Figure 2. The mea s ured flue gas temperatures are u se d to calculate Q directly; thi s, the net heat tran sfe rred turns out to be negative as heat i s lost from the sys tem the flame and the flue to th e s urrounding s. The absolute values of Q as a function of the difference between the temp era ture s of ex it ing flue gases and inlet gas temperature (co n s id e red to be the sa me as the surroundings taken to be 25 C) are plotted in TABLE2 Molar Flows and Mole Fractions in the Gas Stream Leaving the Flue Molar Flows Gaseous-product basi s Dry-product basi s Wet-product ba s i s General Expressions for Mole Fractions Yo 2 =xFo / Fr ( 4m+n J F 0 YN2 =l -4-+x ) Fr rN IO y CO2 = mF o / Fr = ( m / X )Y 02 Mole Fractions of Water Vapor Gaseous-product ba s i s y H 2 o = ( n / 2 )F 0 / F f" 5 Dr yproduct ba s i s Y H 2 o = 0 Wet-product ba s i s Ytt 2 o = YH~o [Tl] [T 2] [T3] Expressio11s for x, the stoichiometric coefficie11t of excess air, i11 terms of y o 2 Gaseous-product b asis x = ( q + n / 2 )y 02 / [ 1( I + r N/O )y 0 2 ] Dry-productbasi s x= ( q xy 0 i)/ [1 -(1+r Nt0 )Yo 2 ] [T4] Wet-product ba s is x = ( q x Y o 2 )/[ ( 1 -y~ ~ o )-( I+ r N,o )y o 2 ] Ill the Equations Above q = m +0.25( 4m + n )rN / O and rN / O = 79.01 /20.99 Spring 2000 Figure 3. Thi s s how s that the rate at which heat is lost from the sys tem increase s with thi s temperature difference in a nonlinear way; thi s provide s a n example of the temperature difference as the dri v in g force" for heat tran sfe r. Kinetics of NO Formation Using other electrochemical cells the analyzer mea s ure s volume fractions of nitric oxide carbon mono xide, and s ulfur dioxide. In Fi g ur e 4, the level ~ of nitri c oxide a nd carbon monoxide are s hown as a function of the vo lumetric flow of natural gas ; no sulfur dioxide was detected ( detection limit~ 1 ppm ). The level of nitric oxide increases s teadil y from 6 to 65 ppm with the flow of natural gas; in contrast the l eve l of carbon monoxide s tarts off at about 17 ppm decrea ses to zero and then increases again at the highe s t flow rate. It is intere s ting that nitric oxide is 3500--t,m-t-,-m-+,-m-+,-m+.-m+.-~-I--u.. 2500 i ::, 8_ 1500 E 500 0 2 4 6 Natural Gas Flow, Umin Figure 2 The measured t emperature of the ex itin g flue gas ( 6 ) as a functio n of the vo lum etric flow of natural gas to the burner measured und er ambient co nditi ons (25 C, 1 atm) Th e ca l c ulat ed adiabatic flame temperature ( J and the es timat ed tempera ture under w hi ch nitri c oxide is formed ( 0 ) are shown !!! -, ai 1/) 1/) .2
    PAGE 88

    produced in this experiment; this is a major concern in many industrial processes[ 5 1 5 1 because of its potential for release into the atmosphere and its subsequent contribution to the formation of acid rain. In contrast to the combustion of a hydrocarbon, the forma tion of nitric oxide from oxygen and nitrogen is endother mic. (I 2) Because of this, the formation of nitric oxide is only signifi cant at elevated temperatures It is instructive to use the known equilibrium and kinetic properties of the reaction to estimate the temperatures that must exist inside the flue. First, the rate of generation of NO within the flue is esti mated; then a simplification of the known kinetics is used to estimate the reaction temperature The mole fractions of the reactants are about 10,000 times larger than the mole fraction of the product; that is, very little of the reactants is converted to product So the flame within the flue may be treated as a differential reactor 11 2 1 61 running at steady state as far as the formation of nitric oxide is concerned Under these assumptions, the mole balance leads to (I 3) where F N o and F N o o are the molar flows of nitric oxide out and into the flue respectively, and r ave is the average rate of generation of nitric oxide within the reactor volume, V. It is reasonable to take F N o ,o = 0 and so (14) The molar flow of nitric oxide out of the reactor is calculated from (I 5) where Y No is the mole fraction of nitric oxide the measure ment provided by the analyzer, and F{l" 5 is the total molar flow of gas out of the flue, given in Table 2. Strictly the molar flow of nitric oxide should be included, but this flow is negligible in comparison to the combined flow of the other species. The effective volume of the reactor is unknown, but the internal volume of the flue available to the flame is about 1 L, and this is used as an approximation for V in the absence of any other information. The estimated rates of generation of nitric oxide are in the range of 0.6-3 0 x 10 6 mol L 1 s 1 Now an expression for the rate of generation of nitric oxide in terms of the concentrations of the chemical species involved is required. Because of its importance in combus tion, the mechanism has received much attention; it is widely believed that the reaction follows the Zeldovich mecha nism. [5 1 7 1 When this mechanism (under fuel-lean conditions) is applied to our data, it turns out that the forward reaction is /82 60 E a. a. 40 n Q) E 20 > 0 --D-NO -6 co 2 4 6 Flow of natural gas, Umin Figure 4. Mole fractions of nitric oxid e and c arbon monoxide measured in the exiting flu e gas as a function of the volumetric flow of natural gas to the burner measured under ambient conditions (25 C 1 atm). dominant; the low levels of nitric oxide seen here are far from equilibrium value s, being approximately an order of magnitude smaller than equilibrium values An abbreviated form of the mechanism that is sufficient to account for the data observed is described; it is also simpler for students who are encountering mechanisms for the first time, to un derstand This abbreviated mechanism is shown in Table 3 From the slow steps the rate of generation of nitric oxide in terms of the concentration of oxygen atom s and concen tration of molecular nitrogen is obtained (16) This expression is obtained from the Zeldovich mechanism in the limit of small concentrations of nitric oxide. To evalu ate this, the concentration of oxygen atoms in terms of measured quantities is needed. This is a useful exercise in thermodynamics I [0]=[0 2 ] ~ (:~ J2 K e q(T) (I 7) where P 0 is the standard pressure of 1 bar. Taking the en thalpy for the dissociation of molecular oxygen to be inde pendent of temperature, the van't Hoff equation gives K e q(T)=K e q(To)e x p[ ~:o ( ) 0 -+ J] (18) where K e q(T 0 ) = exp[-LlG 0 /RT 0 ] and T 0 = 298 K Finally the gas concentrations in term s of the quanitites measured by the analyzer are required. Ch e mi c al En g in ee rin g Edu c ation

    PAGE 89

    [ o ]= Yo 2 P 2 RT [ N ]= YN 2 p 2 RT (19) Here P is the pre ss ure of the reaction, taken to be 1 atm. Table 2, Eq T2 contains the expression for the mole frac tion of nitrogen YN 2 The mole fractions in Eq (19) should strictly be those calculated on the basis of the gaseous-product stoichiom etry, as these are the conditions under which the reaction occurs. Most data were obtained under conditions of excess air for which x < x c ,i, So, in principle, the measured mole fractions should be corrected to give the corresponding val ues on the gaseous-product basis. ( Fwel \ Yi =yi(mea s uredforx 0 2 ,",.HO= +249.17 kJ / mol Entha lp y of formation at 298K L'>.G O = +231. 73 kJ / mo] Gibb s free energy of formation at 298K Step2 k 1 =Aexp(-Ea /RT) A=I x IO Lmol 1 s 1 E = 3 15 kJ/mol Spring 2000 may be inhomogeneous as in a waste incinerator. The three bases upon which the stoichiometry is de scri bed do not lead to mole fractions that differ by very much in the application considered here. To the practicing engineer, the attention to detail may seem unduly fussy, but it i s desirable to train students to formulate a sound theoretical framework from which they can make good practical assumptions and judge ment s This is an example of such a process For whichever basis is appropriate, the mass balance is closed for the primary combustion reaction, assuming it goes to completion. An additional exercise for the st udent is the ex planation of how the molar flow of air 1 83

    PAGE 90

    into flue varies with the flow of natural gas The various temperature data pre se nted here will help the st udent form a s imple picture of a flame a complicated reacting sys tem 151 The first observation i s that the model to explain the formation of NO leads to temperatures larger than the adiabatic flame temperature th e maximum possible flame temperature and the mea s ured exit gas temperature A flame i s far from being a homogeneou s reaction mixture either in it s temperature di str ibution or s pecie s di s tribution. The primary reaction, the combustion of a hydrocarbon her e, occurs in a zone defined by mixing the fuel and air and their subsequent reaction; this i s the so-called comb u stion zone. Becau se the combustion is hi g hly exothermic, the heat gen erated rai ses the temperature of unreacted gases, s uch as excess 0 2 and N 2 to such a level that they s tart to react in a zone, th e po s t-combustion zo ne that i s s patially di s tinct from the combustion zone. Also, it i s interesting to note that the NO-formation and adiabatic-flame temperature s appear to converge upon extrapolation to higher n a tural gas-flow rates where the fuel-air mixture b eco m es s toichiometric. This s u gests th a t NO is formed in zones within which the local fu e air mixture s are sto ichiometric. So qualitatively the fir s t observation c an be explained on the ba s i s of reaction zones with non-uniform t e mperature s and co mpo s ition A seco nd observation supports the idea of non-uniform t e mperature di s tribution. After running th e experiment at th e hi g he s t exit temperature (a bout 1450 F), it was noticed th a t the burner 's nozzle had become partially coated with zinc Thi s mean s that the inside wall of the flu e mu s t have reach e d a tempe rat ure of 787 F the melting point of zinc. Thi s i s much l owe r than th e NO reaction temperature. The use of a s tainle ss s teel flue o r even better a quartz tube so th e str uctur e of the flam e can b e o b serve d would be a worth while improvement. It would also be safe r. The di scre pancy between th e a diabatic fl a me temperatur e a nd th e exiti n g gas t e mperature ma y see m larg e The flu e ca p u se d is de s i g ned to allow for air infiltration throu g h perforation s fabricated in its walls ju s t as th e gas leaves the main part of the flu e. The a s tute s tudent will notice this a nd realize that it will le a d to errors in th e gas compositions and temperature s; they will be lower than they s hou l d be. The cap is de s i g ned to ensure a temperatur e reduction in the gas ju s t before it enters the surroundings. No attempt wa s made to modi fy thi s, or rather to change the po s ition of the prob e, becau se the maximum operating temperature of the probe i s 1550 F for co ntinuou s serv i ce and 2200 F for s hort-term u se Su c h imperfections in laboratory ex periment s, in our opinion, are not a bad thing ; they often help s tud e nts exer cise th eir critical fac ulti es. ACKNOWLEDGMENTS We thank the National Science Foundation for an ILi gra nt award numb er 9451666, a nd the University ofMinne1 84 so t a, Duluth for funding this work Fo r t ec hni ca l assistance we are indebted to D L o n g a nd D Anderson. We ac kn ow edge an anonymous reviewer for tw o helpful observations th at we ha ve included Profe ssor Do ra b Bari a (Apr il 5 194 2 June 1 1999 ) w as the principal inv es tigator of the NSF-ILi grant from which th e analyzer was purchased; thi s pap er i s dedic a t e d to hi s memory. REFERENCES 1. Altieri V.J Gas Anal ysis & T esting of Gaseous Materials, 1st e d American Gas Association Inc. New York NY ( 1945 ) 2 Baria D ., D Dorland R. Davi s, an d K.B Lodge, Incorpo rating Ha zar dous Waste Proc essing in Unit Operations Labo ratories, NSF-ILi grant ( 1994 ) 3. Lodge, K.B., R. Davi s, D Dorl an d and D Baria, "Experi ments in Waste Pr ocess ing for Undergraduates ," Proc ee ing of National Conf ere nc e h e ld in Milwaukee, Wisconsin American Society for Engineering Education ( 1997 ) 4 D eSa ut elle, C., P. Johnson J. Marinoff B Muetzel, B. Pogainis R. Rishavy B. Sc hul er, and J. Shamla Vitrifica tion of High-Level Radioactives Wastes ," Univers i ty of Min nesota Duluth e nt ry in the National Design Competition of the Wa ste Management Education and Research Consor tium, La s Cruces, NM ( 1996 ) 5. Barnard J.A., and J.N. Br a dl ey, Flame and Combustion 2nd e d ., Chapma n and H a ll London England ( 1985 ) 6. Cou l son, J.M., J F Ric h ar d son J R. B ackhurst, and J.H H ar ker Coulson & Richardson 's Chemical Engineering 5th ed Vol. 1 Fluid Flow Heat Transfer and Mass Transfer Butterworth-Heineman L td, Oxford ( 1996 ) 7. de Never s, N., Fluid Mechanics for Chemical En ginee r s, 2nd e d. McGraw-Hill New York, NY ( 1991 ) 8 Perry, R.H., D W Green a nd J.O. Maloney e d s P erry's Chemical Engineer s' Handbook 6th ed., McGraw-Hill Book Com p any, New York NY ( 19 84 ) 9. Johnson A.J. a nd G.H. Auth, eds., Fu e ls and Combustion Ha nd b ook, 1 st e d ., McGraw-Hill New York NY ( 19 5 1 ) 10 Felder R.M. an d R.W. Rousseau, Elem e ntary Principl es of Chemical Pr ocesses, 2nd ed John Wiley & Sons Inc. New York, NY ( 1986 ) 11 San dl er S.I., Chemical and En ginee ring Th e rmodynami cs, 3r d e d ., John Wiley & Sons, Inc ., New York, NY ( 1999 ) 12 Fogler H.S., Elem ents of Chemical R eaction Eng inee r ing, 3r d e d ., Prentice H all PTR, Upper Sad dl e River NY ( 1999 ) 13 Schmidt L.D. Th e Engin ee ring of Chemical R e actions Ox for d University Pr ess New York, NY ( 199 8 ) 14 Smith, J.M. H .C. Van Ness, and M M. Abbott, Intr oduction to Chemical Engine e ring Th e rmod y namics 5th e d ., McGraw Hill N ew York, NY ( 1996 ) 1 5. D av is R.A. "Nitric Oxide Formation in an Iron Oxide Pellet Rotary Kiln Furnace ," J. Air & Waste Manag e. Assoc. 48 44 ( 199 8 ) 16. Levenspiel, 0 ., Chemical Rea ctio n Engin eering, 3rd ed ., John Wil ey & Sons New York, NY ( 1999 ) 17 Hanson R.K. and S. Saliman, in Combustion Chemistry, Gardiner, Jr. W.C. ed. Springer-Verlag Inc New York NY p. 508 ( 1984 ) 1 8. Warnatz, J., in Combustion Chemistry Gardiner Jr., W.C., ed., Springer-Verlag Inc ., New York, NY, p. 508 ( 1984 ) 19 Miller, J. A., and C.T Bowman Mechanism and Modeling of Nitrogen C h emistry in Combustion," Pro g. En ergy Com bust Sci ., 15 287 ( 1989 ) 20. Atkins P W., Physi ca l Chemistry, 5th ed., W.H. Freeman and Company New York, NY ( 1994 ) 0 Ch e mi ca l En g in ee rin g Edu c ation

    PAGE 91

    r \. PO S ITIO NS AVA I LA B LE U s e CEE s reasonab l e rates to advertise. Minimum ra t e, 1 /8 p age, $ 1 0 0 E ac h a dditi o nal co lumn in c h o r porti o n th e r eo f $ 40 UN I VE R SITY O F CINC I NNAT I DEPARTMENT OF CHEM I CAL ENG I NEER I NG The D e p a rtm e nt i s see kin g t o fill t wo t e nur e -tra c k fac ult y p os iti o n s at l eve l s a ppropri a t e t o th e qu a lification s o f th e c andidat es. C a ndidat es s hould h ave a PhD degr ee in c hemi ca l e ngine e rin g o r a cl ose ly relat e d fi e ld.US citi ze n s hip o r p e rm a n e nt r es id e nt s t a tu s i s d es ir e d R es p o n s ibiliti es includ e t e a c hin g a t b o th th e und e r g r a du a t e a nd g radu a t e l eve l s a nd es tabli s hin g a hi g h qual i t y fund e d r esea r c h pro g ram A ppli c ant s mu s t h ave r esea r c h int e r es t s r e le va nt t o th e Ce nt e r fo r M e mbr a n e A ppli e d S c i e n ce a nd T ec hn o l ogy. Ca ndid a t es w ith co m p a ni o n r ese ar c h int e r es t s in biom e dic a l e n g in ee rin g, bi oe n g in ee rin g, o r ph a r m aco l og i ca l e n g in ee rin g a re parti c ul a rl y e n co ura ge d t o a ppl y. Th e D e p a rtm e nt e nroll s a pp rox im a t e l y 3 50 und e r g radu a t e and 50 gra du a t e s tud e nt s a nd pr ese ntl y h as 1 4 fa c ult y. Th e fac ult y includ es an NSF Pr es id e nti a l Y o un g In ves ti ga t o r a nd t wo SF Ca r ee r G ra nt Awar d ees Near l y 22 000 f t 2 o f m o d e rn r ese ar c h s p ace i s a va il a bl e t o th e D e partm e nt mu c h o f w hi c h i s in th e n ew 1 7 5 0 0 0 ft E n g in ee rin g R esea r c h Ce nt e r. Th e C e nt e r for M e mb ra n e A ppli e d S c i e n ce a nd T ec hn o lo gy ha s ex c e ption a l r esea r c h fa c iliti es a nd i s co mpl e m e nted b y th e C e nt e r for C omput e r Aid e d M o l ec ul a r D e si g n a nd th e P o l y mer R esearc h Ce nt e r. Appli ca nt s a r e e n co u rage d t o v i s it th e D e p ar tm e nt 's we b s it e a t < http ://www.c h e mi ca l.u c.e du > Th e Un i ve r s it y o f C in c inn a ti i s o n e of t wo publi c com pr e h e n s i ve r esea r c h uni ve r s iti es in th e S t a t e o f Ohi o. It h as a n e n ro llm e nt o f ove r 3 5 00 0 in a n a ttra c ti ve urb a n e n v ironm e nt w ith a m e trop o lit a n a r ea p o pulati o n o f 1 .5 mil li o n Th e ca mpu s pro v id es uniqu e o pp o rtuniti es for r esea r c h int e r ac ti o n b twee n th e En g in ee rin g Coll ege a nd th e M e di ca l Ce nt e r. Th e U ni ve r s it y h as co mp e titi ve sa l a ri es a nd pro v id es a ttra c ti ve frin ge b e n efi t s for it s e mpl oyees, i ncludin g fr ee tuiti o n fo r s p o u ses a nd c hildr e n Appli ca nt s s h o uld s ubmit a r es um e, publi ca ti o n li s t s t a t e m e nt of t eac hin g a nd r esea r c h pl a n s, a nd n a m es o f thr ee r e f e r e n ces. Appli ca ti o n s r e c eive d b efo r e M ay 1 s t w ill b e g i ve n pref e r e nc e alth o u g h applicant s w ill b e e valuat e d until b o th po s iti o n s a r e fill e d All co rr es p o nd e n ce s h o uld b e a ddr esse d t o: Profe sso r Willi a m 8. Kr a nt z C h a ir o f th e F ac ult y S ea r c h C o mm i tt ee D e p a rtm e nt o f Ch e mi ca l En g in ee rin g U ni ve r s it y of Cin c inn a ti C in c inn a ti Ohi o 45 22 1 01 7 1 Th e U n ive r s i ty oJCi11c i1111 at i i s an Eq u a l Opp o rtu11it y/Affi r111 a ti ve A c tio n E mp loye r

    PAGE 92