ChEmference10 Annual Research Symposium July 13-14, 2010 Outreach 69 & 80 complex I.I.T. Kanpur Supported by Department of Science and Technology, Ministry of Science and Technology, Government of India. Sponsored by Oil and NaturalGas Corporation Ltd. Hindustan Petroleum Corporation Ltd. Tata Research Development and Design Center Bruker Optics Sigma Gases Holmarc Organized by Department of Chemical Engineering Indian Institute of Technology Kanpur Kanpur (U.P.) 208 016 INDIA INDIAN INSTITUTE OF TECHNOLOGY KANPUR KANPUR-208016 Gunjan Kumar Agrahari and Raghwendra Singh Thakur Convener, ChEmference2010 MESSAGE Whenavisionculminatesintoaction,asagaunfolds.Thereflectionoftheideasbehindthe vision paves the way for a mission. ChEmference2010 shines in the light of the concept that was bornthreeyearsagowhenafewvisionaryscholarsofChemicalEngineeringDepartmentat Indian Institute of Technology Kanpur toiled to sow the seeds of a plant that blossoms in 2010.ThusChEmference2010promisestoprovideanopportunitytoyoungresearchersacrossthe country to interchange novel ideas in a multitude of topics ranging from core areas of Chemical Engineering to various multidisciplinary fields.ThefieldofChemicalEngineeringinIndiahasseenatremendousgrowthinthepastfew decadesandiscontinuingtowidenitshorizonsinvolvingasynergyofphysicaland mathematicalscienceswithtraditionalengineeringpracticesandmethods.Thishasopeneda plethoraofavenuestoresearcherstosatisfytheirhungerofinnovationandcomeupwith solutions in sync with the demands of a fast developing society. Howevertopropagatetheemergingideasthescientificcommunityneedstointeractinan environment that helps them unleash their thoughts and speak to peers sharing the common cause and having common goals. ChEmference2010 attempts to create the kind of conglomeration that degenerates into a symposium organized with the objective of providing the research students of our country an opportunity to present their findings to an audience consisting of some of the top scientistsandengineersinthefieldandtootherresearchstudentswhoarekeentoknowthe latest developments. We hope our efforts would be able to win the appreciation from the visitors and prove worthy to our commitments. i Programme vi Invited Speakersvii List of Institutesviii Organizing Committeeix Oral Presentations Session I: Reaction Engineering and Catalysis S101Pre- and post-treatment of gases from a fuel cell processor using ceria-based catalystsParag A. Deshpande and Giridhar Madras [IISc Bangalore] 1 S102Catalytic activity synergism for bimetallic catalystsRucha Paranjpe and Preeti Aghalayam [IIT Bombay] 4 S103The role of flow configuration in microreactors with catalytic insertsVenkat Reddy Regatte and Niket S. Kaisare [IIT Madras] 6 S104Electrodeposited palladium for oxygen reduction reaction in direct methanol fuel cellsKranthi Kumar. M and Raghuram Chetty [IIT Madras] 8 S105Enhancement of esterification reaction using ionic liquids K. Sarvani and P.S.T. Sai [IIT Madras] 10 Session 2: Energy and Environment S201Presentation by Bruker India S202Energy and products yields optimization of Naphtha Stabilization & Fractionation UnitS Kumar, P K Negi, S M Nanoti and M O Garg [IIP Dehradun] 12 S203Application of microchannel reactor for oxidative steam reforming of ethanol and water-gas shift reaction on noble metal 15 ii catalystsA.S.Sandupatla, N.R.Peela, and D.Kunzru [IIT Kanpur] S204Development of diafiltration based membrane separation to purify ligno sulfonates from KBLA. Sarkar, D. Sen and Ch. Bhattacharjee [Jadavpur University] 17 Session 3: Fluid Dynamics and Rheology S301Wetting and adsorption behavior of nonionic surfactants on PTFE surfaceNihar Ranjan Biswal and Santanu Paria [NIT Rourkela] 19 S302Visualization of heat flow in rhombic enclosures: heatline approachR.Anandalakshmi and T.Basak [IIT Madras] 21 S303Computaional modelling of Heat and mass tarnsfer of MHD chemical- Reacting Free convective flow in porous mediaS.Kapoor and S.Rawat [IIT Roorkee] 24 S304Free energy calculations for diblock copolymers in the confinement of two surfaces covered with end-tethered chainsBontapalle Sujitkumar and Upendra Natarajan [IIT Madras] 29 S305Synthesis and characterization of chemically modified sago starch with potential biomedical application Akhilesh V Singh and Lila K Nath [Dibrugarh University] 31 Session 4: Process Control and Operation S401Design of neural controllers for MIMO systemSeshu Kumar Damarla [NIT Rourkela] 33 S402Dynamic simulation and sensitivity analysis of reactive distillation column for isopropyl acetate synthesisNeha Sharma and Kailash Singh [MNIT Jaipur] 35 S403Optimization of the media composition using response surface methodology for the production of cellulase from banana fruit stalkD V R Ravi Kumar, S Chakri, N M Yugandhar and D Sri Rami Reddy [Andhra University, Visakhapatnam] 38 iii S404Preparation of Carbon Micro/Nanofibers for the Adsorptive Removal of Vitamin B12 P. Haldar, Mekala. B, N. Verma and A. Sharma [IIT Kanpur] 42 S405Modelling, simulation and control of boiler power plantK. Sankar, T. K. Radhakrishnan and S. R. Valsalam [NIT Tiruchirappalli and C-DAC,Thiruvananthapuram] 49 Session 5: Bio Process Engineering S501Morphological model to correlate morphology and metabolism in ActinomycetesKirti M. Yenkie, Kamaleshwar P. Singh, Sameer Jadhav and Pramod P. Wangikar [IIT Bombay] 52 S502Synthesis and characterization of interpenetrating network using hydrophilic-hydrophobic acrylicsSudipta Goswami and Kalpana Kiran [BIT Mesra, Ranchi] 54 S503Pb(II) adsorption from aqueous solutions onto clayey soil of Indian origin: equilibrium, kinetic and thermodynamic studiesPapita Saha and Shamik Chowdhury [NIT Durgapur] 57 S504Experiments and modeling on production of biosurfactant through biodesulphurization of hydrotreated dieselS. Bandyopadhyay, R. Chowdhury and C. Bhattacharya [Jadavpur University] 60 S505K- carrageenan biopolymer films reinforced with attapulgite clayS. T. Mhaske, Lokesh Rane, Pravin G. Kadam and Bhushan J. Pawar [ICT Mumbai] 63 iv Poster Presentations P01Fabrication of ultrafine fibers from ElectrospinningM. Ghosh, Karthik Nayani, H. Katepalli and A. Sharma [IIT Kanpur] 66 P02Optimization of process parameters for the production of chitinases under solid state fermentationS. Chakri, D. V. R. Ravi Kumar, N. M. Yugandhar and D. Sri Rami Reddy [Andhra University, Visakhapatnam] 67 P03Isotherm parameters for the sorption of methylene blue onto alkali treated rice husk: comparison of linear and non-linear methods Shamik Chowdhury and Papita Saha [NIT Durgapur] 70 P04Adsorption of aqueous safranine onto rice husk in fluidized bed reactor: kinetic modeling including error analysis Rahul Misra, Praveen Kushwaha and Papita Saha [NIT Durgapur] 72 P05Control of Emulsion Polymerization ReactorJ. Singh, P. Arora, A. Gupta [MNIT Jaipur] 74 P06Equilibrium modelling of binary adsorption of aniline and catechol onto granular activated carbonS. Suresh, V. C. Srivastava and I. M. Mishra [IIT Roorkee] 77 P07Optimization of chemical pretreatment process conditions using response surface methodology (RSM) for ethanol production from banana sheathK. Kumaraguru, M.Virudhagiri and V.Sathiyaraj [Anna University, Tiruchirappalli] 79 P08Study on the dyeing of textile cotton fiber using flame of forest flowerI.Kumar, S. Gupta and P. Saha [NIT Durgapur] 80 P09Assessment on the removal of malachite green dye from waste water using Clayey soilS. Gupta, I. Kumar and P. Saha [NIT Durgapur] 82 P10Software tool for designing small interfering RNAAparna Chaudhary, Sonam Srivastava, Sanjeev Garg [IIT Kanpur] 84 P11Isolation, identication and application of novel bacterial consortium TJ-2 for complete mineralization of aromatic amines resulting from structurally different azo dyes Prashant Barsing, Arti Tiwari, Taruna Joshi and Sanjeev Garg[IIT Kanpur] 86 P12Molecular dynamics study of wetting of water droplet on groove patterned surfacesRavi Chandra, Sandip Khan and J.K.Singh [IIT Kanpur] 88 P13Permeation studies of co2 supported ionic liquid MembraneYogendra Kumar, Yamini Sudha and S. Ashok Khanna [IIT Kanpur] 91 P14Surface interaction and catalytic reactivity of CO2 with H2 over Co/Al2O3 catalystsTaraknath Das and Goutam Deo [IIT Kanpur] 93 v P15Polymer capsule as a micro/nano reactorAuhin K. Maparu, Haider Sami, Sri Sivakumar and Ashok K. Kaul [IIT Kanpur] 95 P16DNA Electrophoresis on Multiple SurfacesTarak K Patra and Jayant K Singh [IIT Kanpur] 97 vi Programme Day 1: July 13, 2010 (Tuesday) 0800 HrsRegistration of participants 0900 Hrs Inauguration by Chief Guest and welcome address by Head, Chemical Engineering 0930 Hrs Keynote address by Prof. K. D. P. Nigam, I.I.T. Delhi 1030 Hrs High Tea 1100 Hrs Commencement of first session with invited talk by Prof. A. N. Bhaskarwar, I.I.T. Delhi 1300 Hrs Lunch 1430 Hrs Commencement of second session with invited Talk by Mr. A. B. Chakraborty, O.N.G.C., New Delhi 1630 Hrs Tea 1700 Hrs Commencement of Poster Session 1800 Hrs Open House Session 1900 Hrs Cultural Program 2000 Hrs Conference Dinner Day 2: July 14, 2010 (Wednesday) 0900 HrsCommencement of first session with invited talk by Prof. Prabhu R. Nott, I.I.Sc. Bangalore 1100 Hrs Tea 1130 Hrs Commencement of second session with invited talk by Dr. V. Runkana, T.R.D.D.C., Pune 1330 Hrs Lunch 1430 Hrs Commencement of third session by Dr. A. Pal, I.I.T. Kanpur 1630 Hrs Informal Meet and Tea 1700 HrsAwards Distribution1800 Hrs Lab Visit vii KEYNOTEADDRESS Prof. K. D. P. NIGAM SPECIALLECTURES Prof.A. N. BHASKARWAR Mr. A. B. CHAKRABORTY Prof. PABHU R.NOTT Dr. V. RUNKANA Dr. ANUPAM PAL viii List of Participating Institutes 1.Indian Institute of Technology Kanpur (Host) 2.Indian Institute of Technology Bombay 3.Indian Institute of Technology Madras 4.Indian Institute of Technology Roorkee 5.Indian Institute of Science, Bangalore 6.National Institute of Technology Durgapur 7.National Institute of Technology Rourkela 8. Jadavpur University, Kolkata 9. Anna University, Chennai 10. Dibrugarh University, Assam 11. M. N. I.T. Jaipur 12. B.I.T. Mesra, Ranchi 13. Andhra University, Vizag 14. I.I.P. Dehradun 15. N.I.T. Trichy 16. U.I.C.T. Mumbai 17. C-DAC, Thiruvananthapuram ix Organizing Committee of ChEmference 2010 Chief Patron:Prof. S. G. Dhande,Director, IIT Kanpur Patron:Prof. Goutam Deo,Head, Dept. of Chemical Engg, IIT Kanpur Advisory Committee Dr. V. Shankar, Associate Professor, Dept. of Chemical Engg, IIT Kanpur Dr. Y. M. Joshi, Associate Professor, Dept. of Chemical Engg, IIT Kanpur Dr. Raj Ganesh S. Pala, Assistant Professor, Dept. of Chemical Engg, IIT Kanpur Dr. Abhijit Chatterjee Assistant Professor, Dept. of Chemical Engg, IIT Kanpur Conveners Raghwendra Singh ThakurGunjan Kumar Agrahari Core Team Jyoti Prasad Chakraborty Yamini Sudha Sistla Sidhharth Sengupta Avinash ChandraSaurabh Singh P. Nageswara Rao Amritha Rammohan Dharmendra PandeySandeep S. PatilSandip Khan Volunteers Aparna Chaudhary Arun P. UpadhyayTarak K. Patra Tanmoy Maitra Neelkanth NirmalPrasad P. L. K. Rajendra Rahul Jagtap K. RakeshManoj GhoshMekala B. Chandan Kumar dasShoaib Hussain Khan Vishal Kusuma Auhin Kumar MaparuAbhai P. Sharma Rupesh Singh Lokesh Tayal Taraknath DasVaibhaw S. ChandelBharat BaldewaDeepthi SantoshiAnurag Pramanik Dharitri RathManish Kaushal Asima Saukat Oral Presentations 1 Pre- and post-treatment of gases from a fuel cell processor using ceria-based catalysts Parag A. Deshpande*, Giridhar Madras Department of Chemical Engineering, Indian Institute of Science, Bangalore 560012, India *Corresponding author: parag@chemeng.iisc.ernet.in Abstract Noblemetalion(PdandPt)substitutedCeO2 wassynthesizedusingsolutioncombustiontechnique. The catalysts were characterized using X-ray diffraction, X-ray photoelectron spectroscopy and trans-mission electron microscopy. The catalysts were used for carrying out the water-gas shift reaction and catalytichydrogencombustion.Thecatalystsshowedhighactivityforbothofthereactions.Equilib-rium conversions were attained using Pt-ion substituted catalyst for the water-gas shift reaction. Room temperatureH2 combustionwasobservedforbothPdasPt-ionsubstitutedcatalystswithlowH2-O2 stoichiometricratio.Themechanismofthewater-gasshiftreactionwasproposedonthebasisofthe spectroscopic observations. Keywords: Water-gas shift reaction; Catalytic Hydrogen Combustion; Spectroscopy; Solution com-bustion; Fuel cells 1. IntroductionSteamreformingofheavierpetroleumfrac-tionsresultsinthegenerationofsynthesisgas whichessentiallyconsistsofalargefractionof H2. Utilization of H2 generated using this method isunsuitableforuseinfuelcellsowingtohigh concentrationsofCOwhichpoisontheelec-trodes.1,2 Therefore, it is required to treat the gas to remove CO and make it rich in H2. The water-gasshiftreaction(WGS)isoneofthemethods to achieve this. Using WGS, the levels of CO in thefeedcanbereducedbelowtheacceptable limits.Fuelcellsoperateathighrecycleratioto makefullutilizationofH2.Inspiteofthis,a small portion of H2 remains unreacted in the ex-haust gas. This may result in build up of H2 con-centrationintheatmospherewithinexplosive limits. The accidents of H2 explosion under such conditionsarereported.3Therefore,treatmentof the exhaust gases from a fuel cell is important to renderitsafeforoutlettotheatmosphere.This study reports two important applications of noble metalsubstitutedCeO2basedcatalysts.WGSis usedasapre-treatmentstepforthepurification oftheinletgastomakeitsuitableforfuelcell applications.Catalytichydrogencombustion (CHC)canbeusedasapost-treatmentstepto make the outlet gas from the processor safe. Both ofthereactionswerecarriedoutusingthecom-bustionsynthesizedcatalystsandthemecha-nisms of the reactions were proposed based upon the spectroscopic observations. 2. Experimental Pd and Pt ion substituted CeO2 catalysts were synthesizedusingsolutioncombustiontech-nique.Asolutionnitrateprecursorsalongwith glycine as fuel was heated in a preheated furnace at350 oC.Thesolutionwasobservedtocatch fireinstantaneouslyandthepowderobtainedaf-tercombustionwasthedesiredcompound.The compoundswerecharacterizedusingpowderX-raydiffraction(PhillipsAnalyticalInstruments, USA),X-rayphotoelectronspectroscopy(Es-calab,VGscientific,England)andtransmission electron microscopy (JEOL). The gas phase reac-tionswerecarriedoutin9mmIDquartztube reactorswithcatalystpackedingranularform betweenceramicwool.Thetemperatureofthe catalystbedwasmaintainedusinganelectric heater equipped with a PID controller. The gases weresentthroughflowcontrollers.Theoutlet gasesfromthereactorweresentthroughanon-line gas chromatograph (Mayura Analyticals Pvt. Ltd.,Bangalore)forcompleteanalysisofthe products.Fig.1showstheexperimentalsetup used for the reactions2 Figure 1.Experimental setup for gas phase reac-tions 3. Results and Discussion TheXRDofthecompoundsshowedthe crystallizationofthecompoundsincubicphase. Fig.2showstheXRDpatternsforCeO2, Ce0.98Pd0.02O2-andCe0.98Pt0.02O2-.TheXRD patterns show the substitution of metal in the lat-ticeasions.Fig.3showstheTEMimagefrom whichnanocrystallinityofthesynthesizedcom-pounds is clear. 20 30 40 50 60 70 80 90Ce0.98Pt0.02O2-Ce0.98Pd0.02O2-CeO2Intensity (a.u.)2 (degree) Figure 2. XRD of the synthesized compounds Figure 3. TEM image of the synthesized com-pound Fig.4showsthevariationofCOconversion withtemperatureduringWGS.ItisclearthatPt substitution was superior as compared to Pd sub-stitution. Equilibrium conversions were achieved within250 oC.4Similarly,thecatalystshave shown high activity for CHC (Fig. 5). Figure 4. Variation of CO conversion with tem-perature during WGS 25 50 75 100 125 150 175 2000.00.20.40.60.81.0H2:O2=2:1 H2:O2=1:1 AirH2 conversionTemperature (oC)40 60 80 100 120 1400.00.20.40.60.81.0H2 conversionTemperature (oC)Pd/CeO2 Pt/CeO225 50 75 100 125 150 175 2000.00.20.40.60.81.0H2:O2=2:1 H2:O2=1:1 AirH2 conversionTemperature (oC)40 60 80 100 120 1400.00.20.40.60.81.0H2 conversionTemperature (oC)Pd/CeO2 Pt/CeO2 Figure 5. Variation of H2 conversion with tem-perature during CHC The spectroscopic evidences support the dual sitemechanismforWGSoverthecompounds 150 200 250 300 350 4000.00.20.40.60.81.0CCO*Temperature (oC) 1% Pd/CeO2 2% Pd/CeO2 1% Pt/CeO2 2% Pt/CeO2150 200 250 300 350 4000.00.20.40.60.81.0CH2*Temperature (oC) 1% Pd/CeO2 2% Pd/CeO2 1% Pt/CeO2 2% Pt/CeO2150 200 250 300 350 4000.00.20.40.60.81.0CCO*Temperature (oC) 1% Pd/CeO2 2% Pd/CeO2 1% Pt/CeO2 2% Pt/CeO2150 200 250 300 350 4000.00.20.40.60.81.0CH2*Temperature (oC) 1% Pd/CeO2 2% Pd/CeO2 1% Pt/CeO2 2% Pt/CeO23 involvingtheadsorptionofCOoverthemetal ionandsplittingofH2Oovertheoxideionva-cancies.5A similar redox mechanism is expected to take place for CHC also.6 4. Conclusions NoblemetalionsubstitutedCeO2 catalysts weresynthesizedusingsolutioncombustion technique.Thecompoundswerenanocrystalline withmetalsubsitutedinionicform.Thecom-pounds showed high activity for WGS and CHC. WGSfollowedadualsitemechanismutilizing noble metal ion and oxide ion vacancies. References [1]V.M.Schmidt,P.Brocheerhoff,B.Hohlein, R.Menzer,U.Stimming,J.PowerSources. 49 (1994) 299-313. [2]F.Vidal,B.Busson,C.Six,O.Pluchery,A. Tadjeddine. Surf Sci. 485 (2002), 502-503. [3]F.Morfin,J.Sabroux,A.Renouprez,Appl. Catal. B: Environ. 47 (2004) 47-58. [4]P.A.Deshpande,M.S.Hegde,G.Madras. AIChE J. 56 (2010) 1315-1324.[5]P.A.Deshpande,G.Madras,AIChEJ.2010 doi. 10.1002/aic.12177. [6]P.A.Deshpande,G.Madras,Appl.Catal.B: Environ. (submitted). 4Catalytic activity synergism for bimetallic catalysts Rucha Paranjpe and Preeti Aghalayam* Dept. of Chemical Engineering, IIT-Bombay *Corresponding author: preeti@iitb.ac.in Abstract Synergism can be defined as the non-additive increase in catalytic activity upon mixing various cata-lyst components. Although well known, an exact representation of such synergism is not yet estab-lished. A commercial catalytic converter used in automobiles has platinum, rhodium and sometimes palladium as the noble metal catalyst and it simultaneously performs the reactions leading to the oxida-tion of hydrocarbons and carbon monoxide and the reduction of nitrogen oxides. Here we demonstrate the synergism in the NO reduction activity for a bimetallic Pt+Rh catalyst. The beneficial effect of the using the two metals is to suppress the production of the harmful byproduct N2O as compared to the monometallic catalyst. A microkinetic model proposed by Mantri and Aghalayam (2007) for the NO-CO reaction on various platinum group metals is able to predict these experimental results very well, when simulations are performed using the commercial reaction kinetics software Chemkin 4.0.2.Keywords: Synergism; Bimetallic Catalysts; Noble metals; Microkinetic modeling 1. Introduction Synergism with respect to catalysts is defined as the non additive increase in the catalyst activ-ityuponmixingvariouscatalyticcomponents and has long been known in the heterogeneous catalysis [1]. But the systematic study on the ef-fectofcombinationofdifferentmetalcatalysts towards reaction mechanism and kinetics is not found in the literature [2]. A good prototype to studythiseffectisthecommercialthreeway catalyticconverter(TWC)usedinautomobile exhaust after treatment. A converter has a num-ber of noble metals (Pt, Rh and Pd) acting simul-taneously as catalysts in order to simultaneously reducethreemajorpollutants(Hydrocarbons, CO and NOx). Since a TWC contains Pt and Rh bothofwhichareactiveindifferentcapacities for the various reactions, the interaction among the two metals with respect to various chemical reactionsisimportantinordertoachieveopti-mumperformancefromboththenoblemetals. Thedifferencesinconversionofthereactants andselectivitiestovariousproductsusingthe monometallic and the bimetallic catalysts need to be analysed. 2. Experiments and Simulations MonometallicPtandRhcatalysts(1wt%) were prepared by the wet impregnation method using Pt and Rh salts on a silica support. A phys-ical mixture of equal amounts of the two mono-metallic catalysts was used as the bimetallic Pt-Rhcatalystinthisstudy.Thecatalystactivity wastestedinafixedbedreactorsetupwitha feed stream containing a fixed CO to NO ratio, using ~200mg of catalysts. The concentrations of CO and NO were measured at the reactor outlet usingaKane9106QuintoxCombustionAna-lyser,asafunctionofthereactortemperature, andotheroperatingparameters.Reactor-scale mathematical simulations were performed using the CHEMKIN 4.0.2 software for equivalent ex-perimental systems (Pure Pt, Pure Rh and Pt+Rh mixture), incorporating the microkinetic reaction mechanism proposed by Mantri and Aghalayam (2007). 3. Results and Discussion The focus of this study is on NO reduction as it is a priority pollutant. Figure 1 shows the NO conversion as a function of the reactor tempera-ture, for the three cases pure Rh, pure Pt, and theRh+Ptcatalyst.Itisevidentfromfigure1 that the catalyst activity for the NO-CO reaction for the bimetallic Pt+Rh physical mixture closely resembles that of the pure Rh catalyst. The ma-thematical simulations are seen to predict the ex-perimental data well in all cases.The bimetallic catalyst in this case has half the amount (weight) of Pt and Rh as compared to 5the pure component catalyst. So if the activity is expressedasthenumberofreactantmolecules converted per noble metal atom, then it can be said that there is an enhancement in the catalyst activity. The main product of NO reduction is N2 and N2O is formed as the byproduct. The predic-tions of the outlet concentrations of N2 and N2O, wereobtainedfrommathematicalsimulations (thesespecieswerenotmeasuredexperimen-tally).Figures 2 shows the beneficial effect of using bimetallic catalysts. The main drawback of Pt catalysts is the tendency to form the undesir-ableside-productN2O.Figure2demonstrates thattheformationisfarlessforthebimetallic catalyst,thanthepurePtone.Thusourwork demonstrates the benefit of metallic catalysts on the basis of both reactant conversion and product selectivity. 4. Conclusions There exists a synergism in the NO reduction activityforabimetallicPt+Rhcatalystasina threewaycatalyticconverter.Intheliterature, thereisnocleardemonstrationofthepotential synergismbetweenPtandRhfortheNO+CO reaction[3,4].Inourwork,wedemonstrate throughbothexperimentsandsimulationsthat the simultaneous presence of Rh and Pt catalysts leadstonon-additiveeffectsonbothreactant conversions and product selectivities. The effect isfavorableastheproductionofharmfulby-productN2Oissuppressedascomparedtothe monometallic Pt catalysts. The ability of the ma-thematicalmodeltopredictexperimentallyob-served features for monometallic and bimetallic catalysts,usingthemicrokineticmodeldevel-oped in our group by Mantri and Aghalayam [5], is also shown. It is proposed to extend this inter-estingworkthroughfurtheranalysisandpara-metricstudies,withtheaimofdevelopingand validating suitable mathematical models for the predictionoftheperformanceofautomobile catalytic converters.

References [1]A.V. Kalinkin, A.V. Pashis and V.I. Bukhtiyarov,Kinetics and Catalysis 48 (2007) 298-304. [2]K. Okumura, T. Motohiro, S. Yoshiyuki and H. Shinjoh, Surface Science 203 (2009) 2544-2550. [3]A.G.v.d Bosch-Driebergen., M.N.H. Kieboom, A.v.Dreumel, R.M.Wolf, F.C.M.J .M.v. Deleft and B.E.Nieuwenhuys,Catalysis letters 2 (1989) 235-242. [4]R.E. Lakis, Y. Cai, H.G. Stenger, J r. and C.E. Layman,J ournal of Catalysis 154 (1995) 276-287. [5]D. Mantri and P.Aghalayam, Catalysis Today 119 ( 2007) 88-93. Figures NO Conversion Vs Temperature02550751000 100 200 300 400 500Temp(0C)%NO conversionPt_Exp Rh_Exp Pt_Rh_ExpPt_Sim Rh_Sim Pt_Rh_Sim Figure 1: Variation of NO conversion (both ex-perimentalandmodelpredicted)withtempera-tureN2O formation01.534.50 100 200 300 400 500Temp(0C)N2O(mole fraction)*E5 Pt Rh Pt+Rh(1: 1) Figure 2: Variation of N2O formation with tem-perature for different Pt-Rh ratios (simulations) 6The role of flow configuration in microreactors with catalytic inserts Venkat Reddy Regatte, Niket S. Kaisare* Department of Chemical Engineering, Indian Institute of Technology Madras Corresponding author: nkaisare@iitm.ac.in Abstract Structural inserts are useful in microreactors to improve the catalyst surface area, enhance mixing and manipulate flow distribution. These structured inserts, such as multi-channel or posted structures, are static structures (like columns or pillars respectively) in the flow channel of a microreactor. In this si-mulation study, the performance of multi-channel and posted catalytic inserts is compared for lean propane/air combustion on Pt catalyst. A comparison with cross-flow configuration with posted inserts is also presented. Significant flow channeling results in poor performance of the cross-flow reactors. A tapered geometry of the inlet manifold is proposed for improving the flow distribution in the cross-post microreactors. The insert geometry does not affect propane conversion in either configuration. Keywords: Flow maldistribution; microreactors; post inserts; channel inserts 1.IntroductionTheadvantagesofmicro-scalesystems[1] includehighratesofheatandmasstransfer, large surface area to volume ratios, and signifi-cant reduction in the experiment time. The sur-face area to volume ratio in micro-scale systems scalesinverselyasthechanneldiameter,and couldbeasmuchastwoordersofmagnitude higherthantheconventionalsystem.However, tomeetthefullpotentialofmicroreactors,fur-therincreaseinsurfaceareaisdesirable,espe-cially in multi-phase systems. Fixed-bed micro-reactorsaredifficulttodesignandoperatein practiceduetotheirhigherpressuredrops[2] and low selectivity. This limits their use only to lower flow rates [3]. Alternatively, one could use preciselystructuredcatalyticinserts,suchas multi-channelorpostedstructures[2,4,5]for the same purpose. The open structure of posts allows for very highsurfaceareawithonlyamodestpressure drop. The narrow residence distribution of gases, and the enhanced mixing [2, 4, 6] increases the conversionforposts[7].However,maintaining uniform flow distribution between adjacent rows of posts is necessary to ensure higher selectivity and reduced pressure drop [8]. Previousstudiesonpostedmicroreactors have focused on understanding the effect of these structuralinsertsonflowdistribution.Inthis study both axial-flow and cross-flow geometries ofpostedmicroreactorsarecompared.Tapered post is proposed for reducing the flow channel-ing observed in cross-post microreactors. 2.Methodology TheComputationalFluidDynamics(CFD) simulationswerecarriedoutusingFLUENT software for various geometries with posted and multi-channel inserts (see contour plots of Figure 2).Amulti-channelreactorconsistsofseven channelswith200micronsgapsizeandwall thickness. The axial-post microreactor consists of a catalytic insert with 0.26 cm in width and 1 cm in length; there are a total of 150 square posts (with d=200 microns) arranged in six rows, with a nominal pitch of 2d.The cross-post configu-ration consists of the same catalytic insert; how-ever, bulk flow is in transverse instead of axial direction.Inordertoimproveflowdistribution in the cross-flow configuration, the outer micro-reactor walls are tapered with an angle of 5.7660. The 2D steady state, laminar and pressure based solver is used to solve the mass, momentum and species equations. The propane combustion reac-tion kinetics is taken from [9]. 3.Results and Discussion A comparison of the four configurations for isothermal conditions (685 K) is given in Figure 1fortheequivalenceratioof0.738.Propane conversions in channel and posted microreactors with axial flow conditions are the same for the entirerangeofflowrates.Theseresultsagree 7qualitatively with the results of [7] that posts and channelsbehaveequallyandtothatofideal PFR.Thepropaneconversioninthecross-post caseislower(byalmost10%athigherflow-rate),aswellaspropanebreak-through(i.e., conversionCeO2>TiO2>ZrO2>Al2O3.The variationofCOconversionwithtemperature on3supportsisshowninFig. 2.Atlow temperatures,theconversionofCOwas limitedbykinetics,whereasattemperatures above 3400C, the conversion was limited by thermodynamicconstraints.WithPt/CeO2-ZrO2/Al2O3catalyst,conversionscloseto equilibriumcouldbeobtainedat3700Cand wt.ofcatalyst/molarflowrateofCO= 11.4 g.h/mol CO. References [1]G.Kolb,V.Hessel,Chem.Eng.J.98 (2004) 1-38. [2] M. Ni, D.Y.C. Leung, M.K.H. Leung, Int. J. Hydrogen Energy 32 (2007) 3238-3247.[3]P.PanagiotopoulouandD.I.Kondarides, Catalysis Today, 112 (2006) 49-52. [4] N.R. Peela, A. Mubayi, D. Kunzru, Catal. Today 147S (2009) 17-23. 00.511.522.533.54600 650 700 750 800 850Temperature, Kmol of product/mol EtOH reactedH2COCH4CO2CH3CHOFig. 1: Variation of selectivities of products with temperature in OSRE (O2/ethanol =0.5; water/ethanol ratio =6) Fig. 2: Variation of CO conversion with temperature for Pt supported on CeO2, TiO2 and CeO2-ZrO2 01020304050607080901000 50 100 150 200 250 300 350 400 450 500Temperature, oCConversion of CO, %equilibriumceriatitaniaceria-zirconia17 Development of diafiltration based membrane separation to purify lignosulfonates from KBL A. Sarkar*, D. Sen, Ch. Bhattacharjee Chemical Engineering Department, Jadavpur University, Kolkata *corresponding author: sarkar.ankur85@gmail.com Abstract In the present study an attempt was made to develop diafiltration assisted membrane separation process to extract lingo sulfonates (LS) from Kraft black liquor (KBL). Here 5 kDa membrane was used to separate LS from the microfiltered raw feed. Six stages of diafiltration were carried out with diluted and non-diluted feed to understand the effect of diafiltration stages on the extent of LS purity. Keywords: Ligno sulfonates, Diafiltration. 1. Introduction Blackliquorisoneofthemosthazardous effluentsthatiscomingoutofanypaper-pulp industry containing mainly lingo sulfonates (LS) alongwithcellulosematerials.Theenviron-mentalimpactofblackliquorresultsnotonly fromitschemicalnature[1],butalsofromits dark coloration that reduces oxygen availability and negatively affects on aquatic fauna and flora. The increasing necessity for water quality pres-ervation too has intensified the need for methods of wastewater treatment within the pulp and pa-per industry. Extensive efforts are being made all over the world to develop treatment schemes for paper mill wastes which could both be economi-cal as well as efficient. The objective of the present work is purifica-tion and concentration of LS present in the black liquor.ThoughUltrafiltration(UF)hasbeen usedworldwide[2]torecoverLSfromblack liquor;butacomplete,neatandeconomical technology has not yet developed. Mainly in the presentworkwehaveattemptedtodevelopa methodology to obtain LS based on the diafiltra-tion process on UF. 2. Experimental Raw feed collected from local paper industry was initially filtered through 20 mesh screen fol-lowedbymicrofiltration(MF)with0.2micron polyethersulfone(PES)membraneresulting 10% (w/v) total dissolved solids (TDS) level in the permeate. Permeate collected from MF (Vis-cosity: 1.05 cP) was introduced to six-stage dis-continuous diafiltration (DD) with volume con-centrationfactor(VCF)2in5kDacrossflow membrane module at 200 kPa in order to obtain purifiedlignin.InonestudyMFpermeatewas directly introduced to the 5 kDa membrane and inanotherobservationMFpermeatewassub-jectedwith50%dilutiontounderstandtheat-tainmentofligninpuritywiththenumberof stages of DD in case of dilution. Retentate ob-tained at sixth-stage of diafiltration was again fed to 5 kDa membrane with VCF 5 to achieve high-er concentration of the lignosulphonates. Finally theconcentratedretentatewasdriedinvacuum oven followed by freeze drying. 3. Results and Discussion Table 1 shows the characterization of the raw feed.Oneoftheremarkableissueswithmem-brane separation process is its continuous fouling whichinherentlyreflectsthereducedlongevity of the membrane. From Figure 1 and 2 it can be observed that 50% dilution yields better perme-atefluxandalsoshowlessruntimetoattain VCF 2. It implicitly shows the less fouling of the membraneandthusincreasedlifetimeofthe membranewhichisadvantageous.Figure3 shows the percentage purity of LS with number of diafiltration stages. From the figure it can be seenthatwith50%dilutionalmost90%purity was attained at third stage, whereas without dilu-tion atleast fifth stage was required to attain 90% purity mark. Therefore dilution of the raw feed indicates that the process becomes more efficient in terms of less energy consumption. 18 Table.1:Characterization of raw feed. ParametersQty Total Suspended Solids (% w/v)0.335429Total Dissolved Solids (% w/v)19.4 Viscosity (Pa.s)0.001603pH9.02 Density (kg.m-3)1117.9 Conductivity (mS.cm-1)1.73 Ash Content (% w/v)8.814 Moisture Content (%w/v)85.256 Chemical Oxygen Demand (ppm)90000 Biological Oxygen Demand (ppm)36000 Lignin (g/L)40.4 Silica (ppm)20,060 Chloride (ppm)2,580 Sodium (ppm)70,620 Carbohydrate (ppm)61,436 0.51.01.52.02.53.05.6 5.1 4.6 3.8 2.8 1.5Permeate Flux (JP) x 106 (m.s-1)Discontinuous Diafiltration Time (h)Feed to 5kDa Membrane without dilution 1st stage 2nd stage 3rd stage 4th stage 5th stage 6th stage Figure.1: Permeate flux versus time plot for six stages of DD without dilution of the MF Perme-ate as feed. 2.02.53.03.54.04.55.05.56.03.3 3.0 2.5 2.0 1.5Permeate Flux (JP) x 106 (m.s-1)Discontinuous Diafiltration Time (h)0.8Feed to 5kDa Membrane with 50% dilution 1st stage 2nd stage 3rd stage 4th stage 5th stage 6th stage Figure. 2: Permeate flux versus time plot for six stages of DD with 50% dilution of the MF Per-meate as feed. 1 2 3 4 5 6020406080100% Lignin purity (dry basis)Diafiltration Stages Without dilution of the feed With 50%dilution of the feed Figure.3: Percentage Lignin purity on dry basis versus diafiltration stages. 4. Conclusions Presentstudyshowsthatfeedchargewith dilution to the membrane accompanied by diafil-tration makes the whole process more productive as well as energy efficient. Also dilution reduces the load on the membrane and thus shows an in-dication of increased lifetime of the membrane. Therefore in short it can be briefly said that 50% dilution of the microfiltered black liquor having 10% (w/v) TDS level gives almost 90% pure LS after third stage of diafiltration. 5. References [1]N. T. K. Oanh, Resour Conserv Recycl 18 (1996) 87-105. [2]E.A. Tsapiuk, M.T. Byrk, M.I. Medvedev and V.M.Kochkodan, JMembr Sci47 (1989) 107-130. 19 Wetting and adsorption behavior of nonionic surfactants on PTFE surface Nihar Ranjan Biswal and Santanu Paria* Department of Chemical Engineering, National Institute of Technology, Rourkela, Orissa, 769008, India *Corresponding author: sparia@nitrkl.ac.in Abstract Wetting and adsorption behavior of two nonionic surfactants (Igepal-630 and Triton X-100) with differenthydrophobicchainlengthhasbeenstudiedonTeflon(PTFE)waterinterface.The kinetics of adsorption was found to be pseudosecondorder for the two surfactants studied here and the equilibrium amount of adsorption ofIgepal-630 is more than that of Triton X-100. The adsorptionisothermsshowthatitfitsbetterwithLangmuirmodelthanFreundlichforboththe cases.ThechangeincontactanglesofIgepal-630andTritonX-100showthatthe reductionof contactangleismoreforIgepal-630.Thereisastraightlinearrelationshipbetweenadhesional tensionandsurfacetensionofaqueoussurfactantsolution.Thisindicatesthattheamountof surfactantadsorbedatsolid-waterinterfaceisequaltotheamountadsorbedatair-water interface.Whereas,nolinearrelationshipwasfoundbetweencosandinverseofsurface tension. Key words: adsorption kinetics, adsorption isotherm, contact angle 1.Introduction Wettingofhydrophobicsurfacesbythe surfactant solutions is very important, owing tothefactthatmanyindustrialprocesses anddailylifeapplicationsareimpossibleto thinkwithoutwetting.Inanywetting process, adsorption of surfactant at the solid-liquidinterfaceandsurfacetensionatthe air-liquidinterfaceplaysanimportantrole. Sincethehydrophobicsurfacesarehaving verylowsurfaceenergy,wettingbypolar solventcanbeenhancedusingsurfactants. Thesurfactantshavinglowsurfacetension valuesatcriticalmicellarconcentration (CMC),andlowsolid-waterinterfacial tension due to adsorption of surfactants may show better wetting property. In view of the widespreadapplicationsofwetting phenomenamanyresearchershavestudied thewettabilityofdifferenttypesof hydrophobicsurfacesbysinglesurfactants [1],mixedsurfactantsystems[2],and different additives [3]. 2. Experimental The PTFE slides of 25.34 mm 1.12 mm and the powder of size 115.7 m were used for wetting and adsorption studies respectively. Triton X100 (TX100) and Igepal-630 were purchased from SigmaAldrich chemicals, Germany. The surface tension and contact angle of aqueous surfactants solutions were measured by the Wilhelmy plate method using a surface tensiometer at 25 ( 0.5) C. For adsorption the concentrations of the surfactants were determined by UV Vis spectrophotometer. 3. Results and Discussion FromtheFigure1(a)itisclearthatthe natureoftheadsorptionisothermforTX100 and Igepal-630 are close to similar, and areofLangmuirtype.Initially,atlow equillibriumconcentration,duetothe presenceofmorefreeaccessiblesitesthe isothermriseslinearlywithahigherslopes, whereas at higher equillibrium concentration formationofplateauregionindicates monolayercoverageof surfactants on PTFE 20 surfaceduetonegligbleintermolecular interactionbetweentheadsorbedsurfactant molecules.Boththeisothermswerefitted withLangmuirandFreundlichmodelsand foundbetterfittingwiththeLangmuir model. Figure1(b)depictsthatthereisa gradualdecreaseincontactanglewiththe increaseinsurfactantconcentrationtill0.1 mM(LogC=-1)forigepal-630,and0.3 mM(LogC=-0.5)mMforTX-100;and abovethoseconcentrationsitremains constant.Thisimplies,thevalueofcontact angle () in the concentration near CMC and higher than CMC remains almost constant.Theshapeofthesecurvesissimilarto thatofadsorptionisothermofIgepal-630 and TX-100 at air-liquid interface. It is seen fromFigure1(c)thatthereisastraight linearrelationshipbetweensurfacetension (LV)andadhesionaltension(LVCos). Astraightlinewithaslope-0.872indicates thatfordifferentsurfactantsatagiven concentrationinthebulkphasethesurface excessatwater-airinterfaceisnotsameas that of Teflon-water interface. 4. Conclusions Theresultsofthisworkcanbe summarized as follows. TheadsorptionisothermforIgepal-630 andTX100areclosetosimilarandareof Langmuir type. Astraightlinearrelationshipexists betweentheadhesionaltensionandsurface tensionofaqueoussurfactantsolution.As theslopeofsurfacetensionandadhesional tensionplot-0.872(1)thisindicatesthat theamountofsurfactantadsorbedatsolid-waterinterfaceisnotequaltotheamount adsorbed at air-water interface. Figure1.(a)AdsorptionisothermsofTX-100 and Igepal-630. (b) Plot of contact angle vsLogC.(c)Plotofsurfacetensionvs adhesional tension. References [1] J. Harkot, B. Janczuk, Applied Surface Sci. 254 (2008) 28252830.[2] Szymczyk, K.; Zdziennicka, A.; Janczuk, B.; Wojcik, W. J Colloid Interface Sci. 293 (2006) 172180. [3] Chaudhuri, R. G.; Paria, S. J. Colloid Interface Sci. 337 (2009) 555562. -4.0 -3.5 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.580859095100105110115 TX-100 Igepal-630 Contact angle ()Log C(b)0.01 0.1 10123456 Amount Adsorbed (m/g)Log C Tx-100 Igepal-630(a)30 40 50 60 70-30-20-10010TX-100 Igepal-630 Adhesional Tension (mN/m)Surface Tension (mN/m)y = -0.872x + 31.901R2 = 0.991(C)21 Visualization of heat flow in rhombic enclosures: heatline approach R.Anandalakshmi, T.Basak* Department of Chemical Engineering,Indian Institute of Technology, Madras. * Corresponding author: tanmay@iitm.ac.in 1. Introduction Natural convection in heat transfer processes arisebecauseofdensityvariationduetothe presence of temperature gradient. Interest in nat-uralconvectioninclosedenclosurecomesfrom itswideapplicationinscienceandengineering. Apartfromregulargeometries,irregulargeome-tries are commonly encountered in various situa-tionslikeatticspaces,solarheating,electronic cooling,foodprocessing,geothermalapplica-tionsetc.Ingeneral,convectionheattransferis studiedbystreamlinesandisotherms.Stream-linesadequatelyexplainfluidflowwhereasiso-therms are quite useful in conduction heat trans-fer. Since fluid and heat flow are coupled in con-vective heat transfer, isotherms are inadequate to study this phenomena.Inordertovisualizeheatflowinconvective heattransferamathematicaltoolcalled`heat-lines'wasproposedbyKimuraandBejan[1]. Heatlines represent path of heat flow, magnitude of heat flow and zones of high heat transfer.Inthepresentstudy,theheatlineconceptis implementedtotwo-dimensionalrhombiccavi-ties with various top angles (=30o, 45o and 75o) under two different boundary conditions: (a) Iso-thermallyheatedbottomwall(b)Non-Isother-mallyheatedbottomwall(Fig.1).Topwallis maintainedadiabaticwhereassidewallsare maintained cold in both the cases. Galerkin finite element method is used to solve the Poisson equ-ationforstreamfunctionsandheatfunctions. Widerangeoffluids(Pr=0.025,0.71,7,and 1000)havebeenstudiedoverarangeofRay-leigh numbers (Ra = 103-105).Effects of Ra and top angles () on fluid flow and heat transfer are analyzedwiththehelpofstreamlines,heatlines and isotherms. 2. Mathematical formulation and simulation Atwo-dimensionallaminarflowmodelwith incompressibleandNewtonianfluidisconsi-dered.Boussinesqapproximationisappliedfor the body force term where density varies linearly withtemperature.Thegoverningequationsfor steadytwo-dimensionalnaturalconvectionflow intherhombiccavityusingconservationof mass,momentumandenergycanbegivenin following dimensionless form (1) The equation for streamfunction is given as (2) The heatfunction () is defined as (3) They satisfy steady energy balance equation to yielda single equation (4) The unique solution of above equation is strong-lydependentonnon-homogeneousDirichlet boundary conditions. A reference value is cho-sen at the top adiabaticwall. The boundary con-ditionsforatthehot-coldjunctionareob-tainedbyintegratingEq.3uptotherequired junction from the reference point. 3. Results and Discussion TheheatlinesforPr=0.015andRa=103 for case1(Fig.2)areperpendiculartothebottom wall as well as left wall. They are smooth circu-lar arcs with low magnitude streamfunctions (||) and heatfunctions (||). Thus, the heat transfer is 22 conduction dominant. As Ra increases to 105 (see Fig.3),thebuoyancyforcesstarttodominate over the viscous forces and circulations in cavity are increased as evident from greater magnitudes ofstreamfunctionandheatfunction.Asymmetric fluidandheatcirculationcellsevolvedueto geometricalasymmetrywithrespecttovertical centerline.Thenon-symmetricflowisfurther inducedbyvariousinclinationanglesofside walls. The left cold wall receives more heat from thelargeportionofthehotbottomwallfor smaller inclination angles. As increases to 75o, thistrendchangescontinuouslyandrightcold wall also starts receiving considerable amount of heatasindicatedbyhighmagnitudeheatfunc-tions (||). It is interesting to observe that closed loop heatline cells in the right portion of the cav-ityoccurinconcentriccirclesandspanningal-most entire cavity. Heatline distribution for case 2 for Pr=0.015 and Ra=105 are shown in Fig. 4. In case 2, both mag-nitudesofstreamfunctions(||)andheatfunc-tions(||)arefoundtobecomparativelylower than the isothermal heating case. Fig. 1:Schematic diagram for various cases.

Fig. 2:Temperature (), streamfunction () and heatfunction()contoursforcase1,Pr=0.025, Ra=103

Fig. 3:Temperature (), streamfunction () and heatfunction()contoursforcase1,Pr=0.025, Ra=105

Fig. 4:Temperature (), streamfunction () and heatfunction()contoursforcase2,Pr=0.025, Ra=105 23 4. Conclusions Thecomprehensiveanalysesofconvective heatdistributioninrhombiccavitiesaredemon-strated using heatline approach. Role of differen-tialheatingforoptimalthermalmixingintwo differentcasesisstudiedviaheatlines.Case1 with=75oisfoundtogiveoptimumthermal mixing. 5. References [1] S. Kimura, A. Bejan, J.Heat Transfer-Trans. of ASME105 (1983) 916-919.[2] T. Basak, S. Roy, Int. J. Heat Mass Transfer, (2008) 3486-3503. 24 Computaional modelling of Heat and mass tarnsfer of MHD chemical- Reacting Free convective flow in porous media1S.Kapoor and 2S.Rawat 1Department of Mathematics, IIT Roorkee, Roorkee-247667. India 2TMPGroup, Materiaux(CEMEF), 1 RueDaunesse, B.P. 207 - 06904 Sophia Antipolis, France. Corresponding author: saurabh09.iitr@gmail.com Abstract Inthepresentstudy,weconsidercomputationalmodelingofthebuoyancy-inducedconvective flow and mass transfer of a micropolar, chemically-reacting fluid over a vertical stretching plane em-bedded in a Darcy- Forchemier porous medium. The finite element method has been used to solve the mathematical model which constitutes a two-point boundary value problem. Such a study finds impor-tant applications in geochemical systems and also chemical reactor process engineering.Keywords:porous media , chemically-reacting, FEM 1.Introduction Flowswithchemicalreactionhasnumerous applicationsinmanybranchesofengineering scienceincludinghypersonicaerodynamics[1, 2],geophysicsandvolcanicsystems,catalytic technologiesandchemicalengineering processes.Manysuchstudieshavebeendone withboundarylayertheory.Acrivos[3]studied the laminar boundary layer flow with fast chemi-calreactions.TakharandSoundalgekar[4]stu-diedthediffusionofachemically-reactingspe-cies in laminar boundary layer flow with suction effects.LaterMerkin[5]consideredisothermal reactiveboundarylayerflows.Morerecently Shateyietalhasstudiedchemically-reactive convective boundary layer flows using asymptot-ic analysis.Littleworkhasbeendoneinanalysingthe chemically-reactiveboundarylayerflowinpor-ousmedia,atopicofgreatimportanceine.g. packed-bedtransportprocesses,geologicalcon-taminationandalsoindustrialmaterials processing.Popetalinvestigatedtheeffectsof bothhomogenousandheterogeneouschemical reactionsondispersioninporousmediausinga Darcian formulation. Aharonov et alstudied the three-dimensionalreactiveflowinporousmedia withdissolutioneffects.LaterFoglerandFreddanalyzedthechemically-reactiveflowinporous media.Theseandmanylaterstudieshoweverdid notconsidertheinfluenceofchemicalreaction or species transfer on the flow regime.2. Mathematical model Considerthe2-D,laminarboundarylayer flow and mass transfer of a micropolar chemical-ly-reacting fluid past a vertical stretching surface embedded in a porous medium. The x-axis is lo-catedparalleltotheverticalsurfaceandthey-axis perpendicular to it. We assume constant mi-cropolarfluidpropertiesthroughoutthemedium i.e. density, mass diffusivity, viscosity and chem-ical reaction rate are fixed. Concentration of spe-ciesinthefreestreami.e.farawayfromthe stretching surface, is assumed to be infinitesimal (zero),anddefinedasC.Temperatureinthe freestreamistakenasT.Thegoverningboun-darylayerequationsfortheflowregime,illu-strated in Fig. 5.1, incorporating a linear Darcian dragandasecond-orderForchheimerdrag, takesthefollowingform,undertheBoussinesq approximation: Continuity eq.0u vx yc c+ =c c(1) Momentum equation ( ) ( )22 11 1 2*a ap pu u u N bu v k g h h g c c u ux y y y k kvv | | c c c c+ = + + + c c c c(2) Angular Momentum 222N N u Nu v Nx y j y j yk | | c c c c+ = + + |c c c c\ . (3) 25 Energy equation 22h h hu vx y yoc c c+ =c c c (4) Spices equation 22c c cu v D cx y yc c c+ = Ic c c (5) Thecorrespondingboundaryconditionsonthe vertical surface and in the free stream can be de-fined now as: 0 : ( ) , 0 , , ,w wuy u U x ax v h h c c N sy| | c= = = = = = = |c\ .(6) : 0, , , 0 y u h h c c N = = = , (7) by introducing the following transformations:1/ 2 1/ 2111( )[ ( )] ( ), [ ] , , ,( )( ) ( ), U(x)=ax, , , ,w wU xxU x f Y Y y u vx y xh h c c U xN U x g Y Cx h h c c vvuv c c= = = = c c = = = (8) Equation (8) reduces the above set of equations Conservation of Momentum:3 22 21 3 21( ) Re Re ( ) 0Rexx x x xx x xFn d f dg d f df df dfB f Gr Gc CdY dY dY dY Da dY Da dYu + + + + = (9) Conservation of Angular Momentum:2 22 2(2 ) 0d g d f df dgg g fdY dY dY dY + + =A(10) Conservation of Energy: 22Pr 0d dfdY dYu u+ = (11) Conservation of Species: 22[ Re ] 0xd C dC dfSc f Sc C CdY dY dY_ + + = (12) The corresponding boundary conditions (6)-(7) are transformed as follows: At220: (0) 0; (0) 1; (0) 1; (0) 1; (0) (0)df d fY f C g sdY dYu = = = = = = (14) As: 0; 0; 0; 0.dfY C gdYu (15) Theshearstressonthesheetsurfaceisdefined as: 1/ 2 1/ 21 1 0( ) ( ) (0) (0)2wydu U UN U f U fdy x xkt k k kv v=| || | | |( | '' '' = + + = + ||( | \ . \ .\ ., at s = 0.5(16) whereastheskinfrictioncoefficientisdefined by: ( )1/ 212Re 1 (0)2wf f xBC C fut| |'' = = + |\ .. The Local Nusselt number can be written as:( )1/ 2Re (0)fx xh xNuku' = = . 3.Numerical solution: Finite element solution to the governing flow equations(9)to(12)withcorrespondingboun-dary conditions (14) and (15) has been obtained . Each element matrix given by equation (41) is of theorder10 10 .Here,wedividethewholedo-maininto80equallineelements.Amatrixof order405 405 is attained on assembly of all the elementequations.Thenonlinearsystemob-tained after assembly is linearized by incorporat-ing the functionsf andU , which are assumed to beknown.Here if and iU arethevalueofthe functionsfandUat the ith node.A system of 346equationleftafterapplyingthegivenboun-daryconditionsissolvedusinganiterative scheme maintaining an accuracy of0005 . 0 . 4. Results and Discussion Thefollowingparametervaluesareadopted in the computations, viz, Grx = 1.0, Gcx = 1.0, _ =1.0,Dax=1.0,Fnx=1.0,Rex=1.0,Pr=0.7, Sc=0.1,B1=0.01,A=1,=1ands=0.5. The results are computed tosee the effect of se-lected important parameters namely Grx, Gcx, _ , Dax, Fnx, Rex, Pr, Sc, B1,A, and s .26 In Fig. 5.2, the variation of velocity versus Y, for various values of the chemical reaction num-ber(_)areshown.Arisein_generatesasub-stantial decrease in velocities. For all values of _ theprofilesdescendfromunityatthewall(Y= 0),andtendasymptoticallytozeroatthefrees-tream (Y ). Therefore clearly chemical reac-tioninducesadecelerationintheflowfield.In equation (12) we observe that the chemical reac-tiontermisnegativeandindeedoppositetothe principaldiffusionterms.Thereforelogically, chemicalreactionwilldelaydiffusivetransport whichinturnwillcorrespondtoretardationin the flow field. Therefore maximum velocity val-ues correspond to the case of zero chemical reac-tion i.e. _ = 0.Conversely,weobservethattemperature functionprofilesi.e.increasewitharisein chemicalreactionparameter,asdepictedinFig. 5.3.Theprofilesarenotaswidelydispersedas forthevelocitydistributions;howeverthereisa clearboostintemperaturesespeciallyatinter-mediateseparationfromthewall.Ourresults agreequitewellforbothvelocityandtempera-ture distributions with those due to Afify [4] who consideredchemicalreactioneffectsonfree convectiveflowandmasstransferofaviscous, incompressibleandelectricallyconductingfluid overastretchingsurfaceinthepresenceofa constanttransversemagneticfield.Temperature profilesgenerallyarelowerincasewithno chemical reaction.Theinfluenceofotherparameterarestudied here5. Conclusions The numerical simulations indicate that:(a)Translationalvelocitydecreases,temper-atureincreases,micro-rotationincreases (inthenear-fieldandintermediaterange from the wall) and mass transfer function decreases with a rise in chemical reaction parameter (_). (b) IncreasingthermalGrashofnumberGrx, increasesthetranslationalvelocity,de-creasestemperaturefunctionvaluesand decreasesmicro-rotation,thelatterinthe regime near the wall. (c)IncreasingspeciesGrashofnumberGcx, increases translational velocity, decreases temperature,decreasesmasstransfer functionandlowersthe micro-rotationat the wall. (d) IncreasinglocalDarcynumberDax,in-creases translationalvelocities butreduc-estemperatureandmicro-rotation,inthe lattercase,againthedepressionismax-imized at the stretching surface (wall). (e)IncreasinglocalForchheimernumber Fnx,reducestranslationalvelocities,but booststhemicro-rotation,inthelatter caseespeciallyatthewallandnearthe wall.(f)Increasing Schmidt number reduces mass transferfunctionbothinthereactiveand non-reactiveflowcases,althoughmass transfer function values are always higher foranyScvalueinthe non-reactivecase (_ = 0). (g) IncreasingPrandtlnumbersubstantially reduces temperature function (u ). (h) Increasing the surface parameter substan-tiallyincreasesmicro-rotationg,particu-larly at and near the wall. (i)Anincreasein,x xGr Gc and xDa leadto anincreaseincoefficientofskinfriction and the rate of heat transfer. (j)Coefficientofskinfrictionandrateof heattransferdecreaseswiththeincrease in chemical reaction parameter. 27 00.510 4 8_ = 0 _ = 1_ = 5 _ = 10_ = 20s = 0.5, Sc = 0.1, Pr = 0.7, Rex = 1, Da = 1, Fnx = 1, Gcx = 1, B1 = 0.01, =1, = 1, Grx = 1 YUFig. 5.2Velocity distribution for different _ 00.510 4 8_ = 0_= 1_= 5_ = 10_ = 20s = 0.5, Sc = 0.1, Pr = 0.7, Rex = 1, Da = 1, Fnx = 1, Gcx = 1, B1 = 0.01, = 1, = 1, Grx = 1 YFig. 5.3Temperature distribution for different _ . 00.81.60 4 8Grx = 10 Grx = 5Grx = 3 Grx = 2Grx = 1s = 0.5, Sc = 0.1, Pr = 0.7, Rex = 1, Dax = 1, Fnx = 1, Gcx = 1, B1 = 0.01, = 1, = 1, _ = 1 YUFig. 5.6Velocity distribution for different Grx 00.510 3 6Grx = 10Grx = 5Grx = 3Grx = 2Grx = 1s = 0.5, Sc = 0.1, Pr = 0.7, Rex = 1, Da = 1, Fnx = 1, Gcx = 1, B1 = 0.01, =1, = 1, _ = 1 YFig. 5.7Temperature distribution for different Grx -1.2-0.40.40 3.5 7 Grx = 10Grx = 5Grx = 3 Grx = 2Grx = 1s = 0.5, Sc = 0.1, Pr = 0.7, Rex = 1, Da = 1, Fnx = 1, Gcx = 1, B1 = 0.01, = 1, = 1, _ = 1 YgFig. 5.8Microrotation distribution for different Grx 00.551.10 4 8Dax = 5Dax = 2Dax = 1Dax = 0.5Dax = 0.1s = 0.5, Sc = 0.1, Pr = 0.7, Rex = 1,Fnx = 1, Grx = 1, Gcx = 1, B1 = 0.01, = 1, = 1, _ = 1 YU Fig. 5.11 Velocity distribution for different Dax 28 00.510 4 8s = 0.5, Pr = 0.7, Rex = 1, Dax = 1, Fnx = 1, Grx = 1, Gcx = 1, B1 = 0.01, = 1, = 1, _ = 1 Sc = 0.1Sc = 0.5Sc = 1Sc = 2Sc = 5Sc = 10 YCFig. 5.16 Concentration distribution for different Sc 00.510 4 8 Sc = 0.1Sc = 0.5Sc = 1Sc = 2Sc = 5Sc = 10s = 0.5, Pr = 0.7, Rex = 1, Dax = 1, Fsx = 1, Grx= 1, Gmx = 1, B1 = 0.01, =1, = 1, _ = 0 YCFig. 5.17 Concentration distribution for different Sc 00.510 4 8Pr = 0.02Pr = 0.05Pr = 0.1Pr = 0.4 Pr = 0.7 Pr = 1s = 0.5, Sc = 0.1, Rex = 1, Dax = 1, Fnx = 1, Grx = 1, Gcx = 1, B1 = 0.01, = 1, = 1, _ = 1 YFig. 5.18Temperature distribution for different Pr 00.260.520 4 8s = 1.0s = 0.75s = 0.5s = 0.25s = 0.0Sc = 0.1, Pr = 0.7, Rex = 1, Dax = 1, Fnx = 1, Grx = 1, Gcx = 1, B1 = 0.01, = 1, = 1, _ = 1 YgFig. 5.19Microrotation distribution for different s

6. References [1]E.Bertolazzi,Int.J.Num.Meth.HeatFluid Flow, 8 (8) (1998) 888-933. [2] J.O. Keller and J.W. Daily, The AIAA J., 23 (1985)1937-1945.[3] A. Acrivos, Chem. Eng. Sci. 13 (1960)[4]H.S.TakharandV.M.Soundalgekar,Aero-technica Missili E. Spazio.58 (1980) 89-92. [5] J.H. Merkinand M.A. Chaudhary, Q. J. Me-chanics Applied Math. 47 (1994) 405-428. [6]M.Q.AliodatandT.A.Alazab,Emirate J.Engg Research. 12(3) (2007) 15-21. 29 Free energy calculations for diblock copolymers in the confinement of two surfaces covered with end-tethered chains Bontapalle Sujitkumar and Upendra Natarajan* Department of Chemical Engineering,Indian Institute of Technology Madras, Chennai, India*Corresponding author: unatarajan@iitm.ac.in AbstractAlatticebasedmodelofdiblockcopolymermeltintercalationinthesurfacescoveredwithend-tetheredchains.Freeenergyandfreeenergychangeiscalculatedbasedonthebasiclatticeandscaling concepts.Ingeneral,theentropicandenergeticfactorsdeterminethefreeenergyofthesystems.Free energy and free energy change curves and their dependence on entropic and energetic factors have been studied. Keywords: Diblock copolymer; free energy; modeling; lattice theory; scaling theory 1. Introduction In this paper we develop a model for a system ofdiblockcopolymermeltintercalatedbetween thesurfacescoveredwithend-tetheredchainsas shown in Fig.1. In section2 the detailed model has beendiscussed,section3containstheresultsand the last section includes the concluding remarks. Figure 1: System under consideration 1.1 Model: Let us consider two surfaces containing end-tethered flexible molecules (End-tethered chains) i.e. Ng/2 tethered chains on each surface, each chain having ng segments, held at a distance greater than the Radius of Gyration[1]Rg of tethered chains, where Rg scales as,(1) withabeingmonomersizeonthelattice.We haveconsideredalatticeofMlayersandL number of lattice sites in each layer [2].Wemakethefollowingassumptionsfor thisthemodel:(1)Tetheredchainsaresparsely distributed;hencetheselfinteractionbetweenthe segmentsoftetheredchainsisnoteffectively considered;(2)Tetheredchainsareathermalwith respect to the surfaces. The thermodynamicWeak Segregationlimitismodeledhere.Inthestudyof molecularstatisticalthermodynamics,thestarting pointisaformofthefreeenergychange (Minimumfreeenergychange)containing physicalcontributionsaswellasinteractive contributions.Thesystemconsideredhereisa grandcanonicalone.Physicalcontributions containentropyofthesystem.Theenthalpy contributions are the interaction energy and elastic energy for the stretching of tethered chains and for freechainsinrespectofsegregationlimits.The freeenergychangeincludes,thefreeenergyfor thesystemoftheblockcopolymermeltinthe confinementofsurfacescoveredwiththetethered chains G1, and the reference states, the free energy forthesystemcontainingthesurfacescovered with tethered chains G2 and the free energy for the diblockcopolymermeltG3.UsingtheScheutjens Fleerslatticetheory [2]andscaling [1]concept G1[4]ismodeled,usingtheScheutjensFleers [2] latticetheoryG2 [4]ismodeledandusingthe modelofohtaandKawasaki [3]G3 [4]is modified.Thefreeenergychangecanbe calculated as (2) For the minimization of the Free energy change, we have (3) 2. Results ThevariableparametersinthissystemareA fraction of A block chain segments, B fraction of B block chain segments, f overall fraction of free chainsegments,gfractionofend-tetheredchain segments, gA Flory-Huggins interaction parameter between tethered and free A block copolymers, gB 30 Flory-Hugginsinteractionparameterbetween tetheredandfreeBblockcopolymers,ABFlory-Huggins interaction parameter between free A and freeBblockcopolymers,AAFlory-Huggins interactionparameterbetweenselfinteractionof freeAblockcopolymers,BBFlory-Huggins interactionparameterbetweenselfinteractionof freeBblockcopolymers,Nnumberofchains,n numberofchainsegmentsinachain,tethering chaindensity,Lnumberoflatticesitesineach layerandMnumberoflatticelayers.Usingthe variousvaluesofandsubstitutingthemin equation3,thecorrespondingvaluesofare obtainedandusingeq.2,Gmin iscalculated.The minimumfreeenergyprofilesforchangesin interactionparameters(Fig.2),forchangesinthe length of the free chains (Fig.3), for changes in the lengthofthetetheredchains(Fig.4),forchanges inthedistancebetweentheplates(Fig.5)are shown. Fig.2 Fig.3 3. Conclusion Fromaboveresults,itcanbeconcludedthatthe minimumfreeenergychangecanbecalculated using the, 1)Maximum values of interaction parameters. 2)Maximum value of free chain length 3)Minimum value of end-tethered chain length. 4)Minimum distance between the plates. 5)Minimum tethered chain density. Fig.4 Fig.5 References [1]P.G.deGennes,Macromolecules,13(1980) 1069-1075. [2]J.M.H.M.ScheutjensandG.J.Fleer,The Journalofphysicalchemistry,83(1979)1619-1635. [3]T.OhtaandK.Kawasaki,Macromolecules, 19 (1986) 2621-2632.[4]Mastersthesis2010inDepartmentof Chemicalengineering,IndianInstituteof Technology Madras, India. 31 Synthesis and characterization of chemically modified sago starch with po-tential biomedical applicationAkhilesh V Singh*, Lila K NathDepartment of Pharmaceutical Sciences, Dibrugarh University, Dibrugarh, Assam, India-786004*Corresponding author: akhileshvikram@gmail.com Abstract This study was undertaken to assess the potential of highly substitute acetylated sago starch (SS) to be used as platform for controlled drug delivery. The acetylated sago starch was synthesized and the two parameters i.e. duration of reaction and temperature were optimized. FT-IR spectra showed the corresponding group attachment inthemodifiedform.XRDpatternexhibitschangeinthephysicalnaturei.e.fromcrystallinetoamorphous. Swelling,hydrationandviscosityvalueexhibitsdecreaseintherespectivevaluesincomparisontoitsnative form which is due to decrease in hydrophilicity value. Keywords:Sago starch, Acetylation, Optimization, FT-IR, XRD, Degree of substitution. 1.IntroductionInthelastfewdecades,anewgenerationof biomaterialhasbeenintroducedaspolymerfor controlledreleaseapplication.Differenttech-niqueshavebeenappliedtoalternativebiopo-lymers.Starchesareaninterestinggroupofna-tivebiomaterialsfortheseobjectives.Acetyla-tion changes the nature of native starch from hy-drophilic to hydrophobic. This modification con-sequently inhibits the characteristic swelling and gellayerformationofnativestarches.Thisin-creasedhydrophobicityforstarchacetatewith higherdegreeofsubstitution,makesthebioma-terialsuitableforcontrolledreleasepolymerin solid dosage forms1,2. Sago starch (SS) commercially isolated from MetroxylansaguRoxb.whichisarichsourceof starch.Thispalmtreeiswelldistributed throughout Southeast Asian region. The objective of this study was to chemically modifytheSSwithhighdegreeofsubstitution andphysicochemicalcharacterizationforitssui-tability in biomedical/pharmaceutical. 2. Experimental Acetylation:Thestarchwastreatedwith acetic anhydride in the presence of pyridine. Firstlystarch(50g)waspregelatinizedandpre-cipitatedwithalcoholandfinallywashedwith distilled water. The preparedgel mass was dried in the oven at 450C overnight. The hard cake of pregelatinizedstarchwasgrindedinamixer grinder and the powder was passed through sieve #80.Thefinepowderwasmixedwith200ml pyridineandafter15min.100mlaceticanhy-dride was added and then it was heated for 1-6 hr andfrom50-1000C..Aftercompletionofthe reactionwetmasswasprecipitatedwithalcohol andwashedtillnosmellofpyridinewasob-served and finally dried at 350C for 48 hr in hot air oven. 2.1Physicochemicalcharacterization:Tocha-racterize the modify starch XRD, FT-IR, viscosi-ty, swelling and hydration study was carried out. 3. Results and Discussion Chemical modification of SS was carried out and it was found that at 1000C and reaction dura-tionof4hrproduceshighestsubstitutedderiva-tive. The degree of substitution was measured by titrimetricanalysis.Forgroupattachmentcon-firmationFT-IRwascarriedoutandthemod-ifiedformshowsabsorptionpeakat1743.6for acetylicgroup.XRDpatternshowstotaldisap-pearanceofcrystallineregionandtransforming intoamorphous,whichmightbeduetogroup attachment and pregelatinization. Hydration capacity of the starch is an indirect techniquetoevaluatewaterholdingcapacity. Fromtheresultsitcanbeconcludedthatdueto hydrophobicityoftheacetylatedderivativeless hydrationwasobservedandsimilarresultswere foundwiththeirviscositydetermination.Dueto hydrophilicnatureofcross-linkedderivativea firmgelwasformedduringhydrationexperi-32 mentandsimilarresultswerefoundinthevis-cosity behavior Figure 1.XRD Pattern of. Modified starches Figure 2.FT-IR Pattern of. Modified starches 4. Conclusions In the foregoing study, acetylated sago starch withdegreeofsubstitution2.6wassynthesized andreactionparameterswereoptimizedforthe same.Modificationwasconfirmedwiththeti-trimetricanalysisandFT-IR.Themodified starchwasmoreamorphousascomparedtothe crystallinenativesagostarch.Theswelling,hy-dration and viscosity values were lower than the native form. Acknowledgements TheauthorsarethankfultoProf.HarkeshB. Singh,Dept.ofchemistry,IIT-Bombayforcar-rying out XRD and FT-IR. 5. References [1]S.Pohja,E. Suihka, M.Vidgren, P.Paronemand J.Ketolainen, J. Controll. Rel 94 (2004) 293-302. [2]O.Korhonen,P.Raatikainen,P.Harjunen, J.Nakari, E. Suihiko,S.Peltonen, M.Vidgren and P.Paronen, Pharm. Res 17(9) (2000) 1138--1143. DESIGN OF NEURALCONTROLLERSFOR MIMO SYSTEM Seshu Kumar DamarlaDepartment of Chemical Engineering, National Institute of Technology, Rourkela, India seshu.chemical@gmail.com Abstract In the present study, the Neural network (NN) based MVSISO (multivariable single input single output) control-lerdesignhasbeenimplementedforanon-linearcontinuousbioreactorprocess.Multilayerfeedforwardnet-works (FFNN) were used as direct inverse neural network (DINN) controllers, which used the inverse dynamics of the decoupled process. For set point tracking; at an operating condition where the cell growth is substrate li-mited,theDINNcontrollersweredesignedforconventionalturbidostatandconcentrationnutristatconfigura-tions. DINN controllers performed effectively well for set-point tracking. Keywords: MVSISO; NN; DINN; turbidostat; Control; nutristat. Introduction Intherecentyears,therehavebeensignificant advancesincontrolsystemdesignfornon-linear processes. One such method is the non-linear inverse modelbasedcontrolstrategy.Neuralnetworks(NN) have the potential to approximate any non-linear sys-tem including their forward & inverse dynamics. The inverse dynamics of the decoupled input-output pairs have been used as controllers. For training the neural networks,theprocessinput-outputdataisgenerated byapplyingapseudorandombinarysignaltothe openloopdecoupled processandthelearningis car-riedoutbyconsideringthefutureprocessoutputsas the reference set point. Application of NN based con-trollersinchemicalprocesseshavegainedhugemo-mentumasaresultoffocusedR&Dactivitiestaken upbyseveralresearchersincludingDonatetal. (1990);Dirionetal.(1995);Hussainetal.(2001); over the decade.Severalresearchershavestudiedthecontinuous bioreactor problem. Cell mass (X g/L, as x1, & y1) & substrate concentrations (S g/L, as x2, & y2) are con-trolled with two manipulated variables, dilution rate (,asu1)andfeedsubstrateconcentration(2],as u2),respectively,thustwodegreesoffreedomis available. Based on the relative gain array (RGA) y1-u2 and y2-u1 pairings are also possible. Modelling Thefollowingmodelequationswereusedforconti-nuous bioreactor process simulation.dx1dt= (p -)1 (1), dx2dt= (2] -2) - x1 (2) p =mcxx2km+x2+k1x22(3), =12.(4),wherepspecific growthandistheyield.Thesteadystatedata which were considered x1s = 0.38 g/L, x2s = 0.05 g/L, s=0.17h-1,2]s = . g/L.Thenominalmodel parametersusedwerepmux = . h-1,m =. gL, = .gg. To avoid control loop interaction ideal decoupling was used & proposed control confi-gurationisshowninFig.1.Theideawastodevelop syntheticmanipulatedinputs-thataffectonlyone process output each. For a (22) system the following relations hold. P (s) is scaled process gain matrix. _1(s)2(s)_ = P (s)(s) _1-(s)2-(s)_(5).Thetargetma-trix;P (s)(s) = _11(s) 22(s)],hence (s) = _11(s) 22(s)] P -1(s)(6).Assuming perfectmodeleachoutputisrelatedtothecorres-pondingsyntheticinputsinthefollowingway, 1(s) = 11(s)1-(s)&2(s) = 22(s)2-(s)(7). The independent SISO tuning parameters can be used foreachcontrolloop,henceindependentNNbased SISOcontrollerscanbedesignedforeachcontrol loopsofthemulti-loopcontrolledprocesstakenup. TheNNsweretrainedwithinput-outputdatagener-atedfromthedecoupledopenloopprocessequation (eq. (7)) representing its inverse dynamics. Hence the resulting inverse NN model can be used as a control-ler typically in a feed forward fashion. In the present study,thetrainingaswellastestingdatabasewas createdbyexcitingthedecoupledopenloopprocess

Design of neural controllers for MIMO system33 withpseudo randombinarysignals(PRBS).Inorder todevelopDINNcontroller,thetrainingofthepro-posed multi layer FF NN (4, 3, and 1) was performed using the Levenberg-Marquardt method. The network predictedtheoutputsofthecontroller(, & 2]) whichactuallyarethemanipulatedvariabletothe process.TheinputsandoutputsofNN(N1,N2& N3) during the training & control:Training Phase: = {(), ( - ), ( - ), ( - )] & =( - ) (8)Control Phase: = {( + ), (), ( - ), ( - )] & = ()(9) Results and Discussion The response of both P control and DINN control for asetpointchangefrom0.38to0.35g/L(inBio-mass) & 0.05 to 0.03 g/L in substrate concentration at t=10 thand20th samplinginstantareshowninFigs.2 &3,respectively,whichreflectperfectsetpoint trackingbyNNcontrollersforaconventionalturbi-dostatandconcentrationnutristat,respectively.For training,thesamplingtimewas0.8timeunitand2 timeunitforsimulation.Thenetworksweresimu-lated up to 80 time units. Conclusions AnNNbasednoninteractingcontrollerdesignhas beenproposedforMIMOsystem.ForaMVSISO system(22),Multiloopcontrolconfigurationwas ascertained using SISO neural network based control-lers,consideringtheinversedynamicsofthede-coupledprocess.DINNcontrollersperformedeffec-tivelyforset-pointtrackingwhilecomparedtocon-ventional Proportional controllers. Acknowledgements It is with a feeling of great pleasure that I express my mostsincereheartfeltgratitudetoProf.Madhushree Kunduforheresteemedguidanceandsupportfor doing this work. References 1.J.S.DonatandT.J.McAvoy,Proceedingsof theAmericanControlConference;IEEE,Pisca-taway, NJ, 1990; pp 2466-2471. 2.J.L.Dirion,M.Cabassud,M.V.LeLann&G. Casamatta,ComputersThem.Engng,19,1995, Suppl., pp. S797-S802.3.M. A. Hussaina, P. Kittisupakornb and W. Daosud, Science Asia, 27, 2001, pp 41-50. Fig.1 Illustration of multivariable control Fig.2 Response of the NN and Proportional controller (in Biomass concentration) for a unit step change in dilution rate. Fig.3ResponseoftheNNandProportionalControl Controller(insubstrateconcentration)foraunit step change in substrate feed concentration 0 10 20 30 40 50 60 70 800.3450.350.3550.360.3650.370.3750.38timeX,BIomass concentrationclosed loop simulations of ANN and P-controller Inverse controllerP-controller0 10 20 30 40 50 60 70 800.0250.030.0350.040.0450.050.0550.060.065timeSubstrate concentration,Sclosed loop simulations of ANN and P-controller Inverse controllerP-controller3435 Dynamic simulation and sensitivity analysis of reactive distillation column for isopropyl acetate synthesis Neha Sharma*, Kailash Singh Department of Chemical Engineering,Malaviya National Institute of Technology Jaipur (Rajasthan), India-302017 *Corresponding author:nehasharma103@gmail.com Abstract In this paper, simulation studies of a reactive distillation column for isopropyl acetate production from isopropnol and acetic acid has been studied. A dynamic model has been developed, which is based on reaction kinetics and incorporates vapor - liquid equilibrium on each tray. A MATLAB program was writtentosolvetheresultingordinarydifferentialequations.ASimulinkmodelwasdevelopedto runinthedynamicmode.Thedynamicsimulationhasbeenusedtounderstandtheeffectofrecycle ratio, reflux ratio, feed tray location, molar ratio of isopropanol to acetic acid, pressure, and total num-ber of trays. The objective of this study is to increase the concentration of isopropyl acetate in the dis-tillate and increase isopropanol conversion. Sensitivity analysis of various operating parameters of re-active distillation column has shown that variation of these parameters implies a larger impact on pu-rity of isopropyl acetate (IPAc). Keywords: Reactive distillation, esterification, isopropyl acetate, sensitivity analysis, simulink 1.Introduction Reactivedistillation(RD)columncombines thekeyoperationsofmostchemicalprocesses intooneunit:chemicalreactionanddistillation. Thecombinationofreactionanddistillation helpsinachievingproductsofhigherpurityand higherconversionofreactantsascomparedto oldconventionalprocesses.Theinformationre-latedtoisopropylacetatesynthesiswithacetic acidinareactivedistillationcolumnisrarely found in the literature except for the vapor-liquid equilibrium(VLE)andkineticsdata[1,2].Sev-eral literature papers are available on RD process design[3-5],butfewpapershavereportedthe sensitivity analysis of RD columns [6-9].In this paper,sensitivityanalysisofreactivedistillation columnforisopropylacetatesynthesiswithace-tic acid and isopropanol is studied. 2.Process Description Adynamicmodelhasbeendevelopedand theresultingordinarydifferentialequationsare solvedbyusingMATLAB.ASimulink modelfilewasdevelopedtoberuninthedy-namicmode.Thecolumnspecificationsaretak-enfromliterature.Totalnumberofstagescon-sideredinthisstudyis14.Feedlocationfor heavyreactantaceticacid(HAc)isstage3and forlightreactantisopropanolisfromreboiler, becausewehavealargestcatalystholdupinthe reboiler. Overhead column containing an isopro-pyl acetate plus water produced by the esterifica-tions reaction and small amount of unreacted al-coholiscondensedanddirectedtodecanter which serves to separate the organic and aqueous layersofthereactionproductmixture.Theor-ganic phase contains isopropyl acetate and small amountofwaterandalcohols.Whereasthe aqueous phase contains about 90-99% water and the remaining is alcohol and acetate. Isopropylacetatesystemexhibitsnonideal phasebehaviorandhasfourazeotropesinthis system.TheNRTL(non-randomtwo-liquid)ac-tivitycoefficientmodelparametersaretaken from the paper by Venimadhavan et al [10]. The esterificationofaceticacid(HAc)withisopro-panol(IPOH)canbeexpressedinthefollowing form: Acetic acid + Isopropanol Isopropyl acetate + water Thechemicalreactionkineticmodelis adoptedfromthepaperbyTangetal[3].They usedasolidacidcatalyzedreactionwithacidic 36 ion-exchangeresin(Amberlyst35wet,Rohm and Hass). 3.Simulation Results 3.1.Effectofrefluxratio:Isopropylacetateyield andcompositionintheproductafterdecanter increasesinitiallyandthendecreasesbeyonda refluxratioof1.5withincreaseintherefluxra-tio as shown in Fig. 1. Maximum composition of isopropyl acetate of around 83.9% was observed at reflux ratio of 1.5 0.1240.1260.1280.130.1320.1340.1360.1380.140 1 2 3 4 5 6Ref luxr at ioYield of IPAc Figure 1. Effect of reflux ratio on isopropyl ace-tate yield in distillate. 3.2.Effectofrecycleratio:Isopropylacetate yieldandcompositionintheproductdecreases continuously with increase in the recycle ratio as showninFig.2.Theresultshowsthatrecycle ratio should be low for high product purity. 0.8050.810.8150.820.8250.830.8350 0.2 0.4 0.6 0.8 1Recycle r at ioIPAc (mole fraction) Figure 2. Effect of recycle ratio on isopropyl ace-tate composition in distillate. 3.3.Effect of molar ratio of isopropanol to acetic acid:Isopropylacetateyieldincreasescontinu-ously with increase in molar ratio of isopropanol to acetic acid in feed as shown in Fig. 3. The re-sultshowsthataslightexcessofisopropanolis favorable because it facilitates liquid-liquid sepa-ration. 0.10.1050.110.1150.120.1250.130.1350.140.1450.2 0.3 0.4 0.5 0.6Yield of IPAc (mol/s)Isopropanol (mol/s) Figure3.Effectofmolarratioofisopropanolto acetic acid 3.4.Effectoffeedtraylocation:Adjustingthe feedtraylocationimpliesagoodeffecton achievingproductyield.Acloseinspectionof Fig. 4revealsthatisopropylacetatecomposition initiallyincreasesandthendecreasescontinu-ously with increase in feed tray location of acetic acid.0.8120.8140.8160.8180.820.8220.8240.8260.8280.832 4 6 8 10 12Feed t r ay location of acet ic acidIPAc (mole fraction) Figure 4. Effect of acetic acid feed plate location onyieldofIsopropylacetatekeepingisopropa-nol feed plate position fixed at 15th plate. 3.5.Effect of Pressure: Isopropyl acetate yield increases continuously with increase in the pres-sure as shown in Fig. 5. The results show the high sensitivity to pressure. 37 0.120.130.140.150.160.170 2 4 6 8 10 12Pr essur eYield of IPAc (mol/s)Figure 5. Effect of pressure on yield of Isopropyl Acetate 3.6.Effectoftotalnumberoftrays:Isopropyl acetateyieldandcompositionincreasewithin-creaseintotalnumberoftraysasindicatedin Fig. 6.Howeverthereisonlymarginalincrease in the yield beyond 19 trays. Therefore, selection of 19 trays is almost optimum. 0.1340.13450.1350.13550.1360.13650.1370.13759 10111213141516171819202122Tot al numberof tr aysYield of IPAc Figure6.Effectoftotalnumberoftrayson product yield 4. Conclusions Sensitivityanalysisofvariousoperatingpa-rametersnamelyfeedtraylocation,refluxratio, recycle ratio, molar ratio of isopropanol to acetic acid, pressure, and total number of trays of reac-tivedistillationcolumnhasshownthatvariation oftheseparametersimpliesalargerimpacton purity of isopropyl acetate (IPAc).According to the results, Acetic acid should be fed at the upper tray and isopropanol should be fed at the bottom tray.Refluxratioof1.5isfoundtobeoptimum forgettingthehighestcomposition.Recyclera-tio should be minimum. Molar ratio of isopropa-noltoaceticacidshouldbekeptmorethan1. The optimum number of trays is found to be 19. References [1]Doherty,M.F.,Malone,M.F."Conceptual design of distillation systems", McGraw-Hill, 2001. [2]L.S.Lee,M.Z.Kuo,FluidPhaseEquilibria 123 (1996) 147-165 [3]Y.Tang,Y.Chen,H.P.Huang,C.C.Yu,AIChE 51 (2005) 1683-1699. [4]I.K.Lia,S.B.Hung,W.J.Hung,C.C.Yu, M.J. Lee, H.P. Huang, Chemical Engineering Science 62 (2007)878 - 898.[5]Y.C.Cheng,andC.C.Yu,Chemicalengi-neering science 60 (2005) 4661-4677. [6]K.J.J.Prakash,A.K.Jana,ChemicalProduct and Process Modeling, 4, A13 (2009) 1-22[7]A.K.Chandrakar,V.K.Agarwal,S.Chand andK.L.Wasewar,InternationalJournalof ChemicalReactorEngineering,5,A81 (2007) 1404.[8]S.Steinigeweg,J.Gmehling,Ind.Eng. Chem. Res. 41 (2002) 5483-5490. [9]A.Singh,R.Hiwale,S.M.Mahajani,R.D. Gudi,J.Gangadwala,A.Kienle,Ind.Eng. Chem. Res. 44 (2005) 3042-3052. [10]G.Venimadhavan,M.F.MaloneandM.F. Doherty, AIChE J. 45 (1999) 546-556. 38 Optimization of the media composition using response surface methodology for the production of cellulase from banana fruit stalk D V R Ravi Kumar*, S Chakri, N M Yugandhar and D Sri Rami ReddyCentre for Biotechnology, Department of Chemical Engineering, College of Engineering, Andhra University, Visakhapatnam 530 003,Andhra Pradesh, INDIA. *Corresponding author: ravidvr@gmail.com Abstract Banana is a major commercial crop of tropical and subtropical countries generating vast agricul-tural waste after harvest. Banana fruit stalk after harvest was used as substrate for the production of cellulase enzyme through solid state fermentation using Cellulomonas uda NCIM 2353. Response sur-face methodology (RSM) involving Box-Behnken design was applied for the optimization of medium constituents for cellulase production. A polynomial model was created to correlate the relation between the three important variables (moisture content, concentrations of glucose and beef extract) and the cel-lulaseproduced.Theoptimalsetofconditionsformaximumcellulaseproductionwasasfollows: moisture content (83.4 % v/w), glucose (1.05 % w/w) and beef extract (1.09 % w/w). The maximum activity of cellulase at these optimum conditions was 10.06 U/ml. This method was efficient because only 15 experimental runs are necessary to assess these conditions and the model accuracy was very satisfactory, as the coefficient of determination, R2 was 0.973.Keywords: Optimization, Response surface methodology, Box-Behnken design, Cellulase, Cellulomo-nas uda NCIM 2353, Banana fruit stalk. 1. Introduction To date, the production of cellulase has been studied in submerged culture process but the rel-atively high cost of enzyme production hindered inindustrialapplicationofcellulosebioconver-sion.Solidstatefermentationisanattractive process to produce cellulase economically due to its lower capital investment and lower operating toproducemicrobialenzymesusinglingo-cellulosic wastes. Banana (Musa sopiestum) fruit stalks abundantly available in banana production fields and markets, appears to be favorable sub-strateforcellulaseproduction,asitischeaply available in the tropical and subtropical countries and has a cellulase content of 23.85% [1]. The traditional one factor at a time technique used for optimizing a variable system is not only time consuming but also often easily mixes the alternativeeffectbetweenthecomponentsand also it requires more number of experiments to determinetheoptimumlevels.Thedrawbacks can be eliminated by optimizing all the effecting parameters collectively by response surface me-thodology(RSM)whichincludesfactorialde-sign and regression analysis.The present work aim at a better understand-ingoftherelationbetweentheimportantinde-pendent variables (nutrients) and dependent vari-able (activity) to determine optimum set of con-ditions for the maximum production of cellulase fromCellulomonasudaNCIM2353usingBox-Behnken design. 2. Experimental 2.1Microorganism: Pure culture of Cellulomo-nasudaNCIM2353wasprocuredfromNCL, Pune, INDIA. 2.2Substrate:Bananafruitstalkwascollected from local market. It was sliced, spread on trays and oven dried at 500C for 48 hrs, grounded and stored at room temperature. 2.3EnzymeproductioninSSF:Erlenmeyer flasks(250ml)containing10gofbananafruit stalk moistened with mineral salt medium [(g/L); Na2HPO4.2H2O, 1.1; NaH2PO4.2H2O, 0.61; KCl, 0.3; MgSO4.7 H2O, 0.01; pH (7.0)] with initial moisture content of 65 % were autoclaved. 10 % (v/w)oftheinoculumwasinoculatedandthe contents of the flasks were mixed thoroughly to ensure uniform distribution of inoculum and in-cubated in a slanting position at 350C for 72 hrs.2.4Enzymeextraction:Enzymeextractionwas 39 carried out by adding 30 ml of 0.1M phosphate buffer (pH 7.0) and then shaking the mixture on an orbital shaker (120 rpm) for 60 min at 280C. The contents were filtered through cheese cotton. The filtrate was pooled together, centrifuged and the supernatant was used for enzyme assay. 2.5Optimizationofselectednutrientsusing RSM: Box-Behnken design and RSM were used tooptimizetheselectedfactors(moisturecon-tent, concentrations of glucose and beef extract) which resulted from the preliminary studies. The ranges of selected variables were given in Table 1. 2.6Enzymeassay: Cellulase activity was deter-mined at 400C by incubating 1 ml of enzyme so-lutionwith1mlof0.1Mpotassiumphosphate with pH (8.0) containing 1 % w/v cellulose. 3. Results and discussion The nutrient composition in the medium for the production of cellulase by Cellulomonas uda wasoptimizedusingBox-Behnkendesignwith middlerangeparameters,asitisapowerful technique for testing multiple process variables. Experiments were carried out as per the design and the average cellulase activity obtained after 5 days fermentation with 15 experiments in tripli-catefromthechosenexperimentaldesignare shown in Table 2. The application of RSM [2] yielded the following regression equation, which is an empirical relationship between the activity and test variables in coded units: Y=-208.8 +2.76 X1 +109.7 X2 +83.46 X3 +0.045 X1X2 0.26 X1X3 3.43 X2X30.015 X1X1 51.9 X2X2 26.53 X3X3 Where Y=cellulase activity, X1, X2 and X3 are thecodedvaluesofthemoisturecontent,con-centrationsofglucoseandbeefextract,respec-tively. Statistical testing of the model was done by the Fischers statistical test for analysis of vari-ance (ANOVA) and the results are shown in Ta-ble3.Thecalculationofregressionanalysis gives the value of the determination coefficient (R2 =0.973) indicates that only 2.7 %of the to-tal variations are not explained by the model and the F-value of 97.88 indicates that the cellulase productionbyCellulomonasudahasagood model fit due to the high values ofR2 and F.The p-values are used as a tool to check the significance of each coefficient, which also indi-cate the interaction strength between each inde-pendentvariable.Thesmallerthep-values,the bigger the significant of the corresponding coef-ficient[3].Table4showedthattheregression coefficients of the linear terms (X1, X2, and X3) andthequadraticcoefficients(X1X1 and X2X2) were significant at 1% level and a quadratic term (X3X3) was significant at 5% level. Response surface plots as a function of two factors at a time, maintaining all other factors at fixed levels (zero for instance) are more helpful in understanding both the main and the interac-tioneffects.Theactivityvaluesfordifferent concentrationofthevariablecanalsobepre-dicted from the respective response surface plots (Fig. 1).Theanalysisrevealedamaximumcellulase activity of 10.06 U/ml which was 5.34 % more than that of the run number (13) achieved at the points where moisture content, concentrations of glucose and beef extract were 83.4 %v/w, 1.05 % w/w and 1.09 % w/w, respectively. The final ex-periment repeated at the optimal settings of the processvariablesproducedcellulaseactivityof 9.89 U/ml which was quite closer to the optimal value predicted by the Box-Behnken design. Fig. 1: The surface plots describing the effect of three independent variables on chitinase activity 40 Table 1: Variables in the experimental plan Coded levelsVariables -10+1 Moisture content (%v/w) 70.080.090.0 Glucose (% w/w)0.81.01.2 Beefextract(% w/w) 0.81.01.2 Table2:TheBox-Behnkendesignmatrixem-ployed for three independent variables in coded units along with observed values Run No.X1X2X3Observed (cellulase ac-tivity) 170.00.81.03.4 290.00.81.05.2 370.01.21.06.24 490.01.21.08.4 570.01.00.83.8 690.01.00.87.9 770.01.01.26.8 890.01.01.28.8 980.00.80.83.8 1080.01.20.85.9 1180.00.81.26.9 1280.01.21.28.45 1380.01.01.09.55 1480.01.01.09.36 1580.01.01.09.3 Table 3: Analysis of variance (ANOVA)SVSSDFMSFp X1+X1X121.11210.5530.960.001 X2+X2X227.65213.8240.560.0008 X3+X3X315.5527.7722.820.003 X1X20.0310.030.090.77 X1X31.1011.103.230.13 X2X30.0710.070.220.65 Error1.7050.34 Total SS 63.8414 SV: Source of variation SS: Sum of squares DF: Degrees of freedom MS: Mean square F: F-value p:p-valueTable 4: Model coefficients estimated by linearregression. FactorRegression Coeff.p-value Intercept-208.80.0008 X12.770.003 X2109.700.003 X383.470.009 X1X1-0.020.004 X2X2-51.900.001 X3X3-26.50.017 X1X20.050.77 X1X3-0.260.13 X2X3-3.440.65 41 4.Conclusions The work has demonstrated the use of a Box-Behnkendesignbydeterminingtheopti-mum conditions leading to the maximum cellu-lase production using banana fruit stalk as sub-strate. This methodology could therefore be suc-cessfullyemployedtoanyprocess(especially withthreelevels),whereananalysisoftheef-fects and interactions of many experimental fac-torsarereferred.Box-Behnkendesignsmaxi-mize the amount of information that can be ob-tained,whilelimitingthenumberofindividual experiments required. Response surface plots are very helpful in visualizing the main effects and interaction of its factors. Thus, smaller and less time consuming experimental designs could gen-erally suffice the optimization of many fermenta-tion processes. References [1] C. Krishna and M. Chandrasekaran, Applied Microbiol. Biotechnol. 46 (1996) 106111. [2]D.V.R.RaviKumar,D.SriRamiReddy, N.M. Yugandhar and G.Hanumantha Rao, J . Bi-ochem. Technol. 3 (2009) 69-74. [3] F. J . Cui, F.Li, Z.H.Xu, H.Y.Xu, K.Sun and W.Y.Tao,Bioresour.Technol.97(2006)1209-1216. 42Preparation of Carbon Micro/Nanofibers for the Adsorptive Removal of Vitamin B12 P. Haldar1*, Mekala. B1, N. Verma1, A. Sharma2 1Department of Chemical Engineering, Indian Institute of Technology Kanpur, 2Department of Chemical Engineering, Indian Institute of Technology Kanpur and DST Unit on Nanosciences, Kanpur 208016 (India) *Corresponding author:parhalda@iitk.ac.in Abstract In the present study we have developed the hierarchal web of carbon micro/nanofibers for the adsorption of vitamin B-12 (VB12). The method of preparation of the substrate, activated carbon fiber(ACF)involvedphysicalactivationofnon-activatedcarbonmicrofibers,usingsteam. Carbon nanofiber (CNF) was prepared by impregnation of ACF with nickel nitrate, followed by calcinations,thenreductionwithhydrogen,andfinallybycatalyticchemicalvapordeposition (CVD) using benzene as the source of carbon. After growing CNF on ACF, Ni particles used as catalystwereremovedbyultra-sonication.Thepreparedmetalimpregnatedhierarchalwebin this study was tested for the adsorption of VB12 over the concentration range of 25-200 ppm in waterunderbothbatchandflowconditions.Theresultsshowsignificantlylargeextentof removal of VB12. In addition, sonicated CNF was found to be superior in comparison to ACF and CNF without sonication.Key words: Activated carbon fiber (ACF), carbon nano fiber (CNF), Vitamin B12, adsorption. 1. Introduction VitaminB12 (cobalamin)isanimportant water-soluble vitamin (MW= 1355 Da) with amolecularsizeof2.09nm.VB12isa cobalt-containingsubstance,redincolor. Themoleculehasaporphyrinunitcalleda corrinring,whichcoordinatesatrivalent cobalt ion. The sixth coordination position is directedtonitrogenofa5,6-dimethylbenzimidazolering.Thisringis positionedperpendiculartothecorrinring [1]. NormalrangeofVitaminB12inhuman bodyshouldbefrom200-900pg/ml.The elevated concentration levels of Vitamin B12 inthebodymaycausecardiovascular disease.Inthisstudy,wehavesynthesized thehierarchalwebofcarbon micro/nanofibers for the removal of VB12 by adsorption.2. Experimental 2.1 Adsorbent 2.1.1 ACF: The non-activated phenolic resin basedfiberswerefirstcarbonizedandthen activatedusingsteaminatubularreactor. Theas-receivedACFimportedfromKynol, Japanweredriedundervacuumandata temperatureof473Ktodriveoffmoisture andotherimpuritiesfromthefiberbefore carbonization and activation. 2.1.2CNF:Continuousincipiencemethod was used for impregnating ACF with nickel nitratesaltdissolvedinacetone.ACFwas wrappedaroundaperforatedtubewhichin turnwasfittedintoanothertubularshell (ReferFigure1).Impregnationwascarried outfor6hundercontinuousflowusinga peristalticpump.Afterimpregnation,the sample was dried in air for 6 hours and then insidetheovenfor6hoursat323Kto removesurfacemoisture.Thereafter,itwas calcined at 573 K for 2 h. During calcination Ni(NO3)2 wasconvertedintoNiO.The 43resultingmetaloxidewasthenreducedto thecorrespondingme