<%BANNER%>

Flame Blowout and Pollutant Emissions in Vitiated Combustion of Conventional and Bio-Derived Fuels

Permanent Link: http://ufdc.ufl.edu/UFE0024935/00001

Material Information

Title: Flame Blowout and Pollutant Emissions in Vitiated Combustion of Conventional and Bio-Derived Fuels
Physical Description: 1 online resource (267 p.)
Language: english
Creator: Singh, Bhupinder
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: biofuels, blowout, cycles, flameless, formation, semiclosed, soot
Mechanical and Aerospace Engineering -- Dissertations, Academic -- UF
Genre: Mechanical Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The widening gap between the demand and supply of fossil fuels has catalyzed the exploration of alternative sources of energy. Interest in the power, water extraction and refrigeration (PoWER) cycle, proposed by the University of Florida, as well as the desirability of using biofuels in distributed generation systems, has motivated the exploration of biofuel vitiated combustion. The PoWER cycle is a novel engine cycle concept that utilizes vitiation of the air stream with externally-cooled recirculated exhaust gases at an intermediate pressure in a semi-closed cycle (SCC) loop, lowering the overall temperature of combustion. It has several advantages including fuel flexibility, reduced air flow, lower flame temperature, compactness, high efficiency at full and part load, and low emissions. Since the core engine air stream is vitiated with the externally cooled exhaust gas recirculation (EGR) stream, there is an inherent reduction in the combustion stability for a PoWER engine. The effect of EGR flow and temperature on combustion blowout stability and emissions during vitiated biofuel combustion has been characterized. The vitiated combustion performance of biofuels methyl butanoate, dimethyl ether, and ethanol have been compared with n-heptane, and varying compositions of syngas with methane fuel. In addition, at high levels of EGR a sharp reduction in the flame luminosity has been observed in our experimental tests, indicating the onset of flameless combustion. This drop in luminosity may be a result of inhibition of processes leading to the formation of radiative soot particles. One of the objectives of this study is finding the effect of EGR on soot formation, with the ultimate objective of being able to predict the boundaries of flameless combustion. Detailed chemical kinetic simulations were performed using a constant-pressure continuously stirred tank reactor (CSTR) network model developed using the Cantera combustion code, implemented in C++. Results have been presented showing comparative trends in pollutant emissions generation, flame blowout stability, and combustion efficiency.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Bhupinder Singh.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Lear, William E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024935:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024935/00001

Material Information

Title: Flame Blowout and Pollutant Emissions in Vitiated Combustion of Conventional and Bio-Derived Fuels
Physical Description: 1 online resource (267 p.)
Language: english
Creator: Singh, Bhupinder
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: biofuels, blowout, cycles, flameless, formation, semiclosed, soot
Mechanical and Aerospace Engineering -- Dissertations, Academic -- UF
Genre: Mechanical Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The widening gap between the demand and supply of fossil fuels has catalyzed the exploration of alternative sources of energy. Interest in the power, water extraction and refrigeration (PoWER) cycle, proposed by the University of Florida, as well as the desirability of using biofuels in distributed generation systems, has motivated the exploration of biofuel vitiated combustion. The PoWER cycle is a novel engine cycle concept that utilizes vitiation of the air stream with externally-cooled recirculated exhaust gases at an intermediate pressure in a semi-closed cycle (SCC) loop, lowering the overall temperature of combustion. It has several advantages including fuel flexibility, reduced air flow, lower flame temperature, compactness, high efficiency at full and part load, and low emissions. Since the core engine air stream is vitiated with the externally cooled exhaust gas recirculation (EGR) stream, there is an inherent reduction in the combustion stability for a PoWER engine. The effect of EGR flow and temperature on combustion blowout stability and emissions during vitiated biofuel combustion has been characterized. The vitiated combustion performance of biofuels methyl butanoate, dimethyl ether, and ethanol have been compared with n-heptane, and varying compositions of syngas with methane fuel. In addition, at high levels of EGR a sharp reduction in the flame luminosity has been observed in our experimental tests, indicating the onset of flameless combustion. This drop in luminosity may be a result of inhibition of processes leading to the formation of radiative soot particles. One of the objectives of this study is finding the effect of EGR on soot formation, with the ultimate objective of being able to predict the boundaries of flameless combustion. Detailed chemical kinetic simulations were performed using a constant-pressure continuously stirred tank reactor (CSTR) network model developed using the Cantera combustion code, implemented in C++. Results have been presented showing comparative trends in pollutant emissions generation, flame blowout stability, and combustion efficiency.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Bhupinder Singh.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Lear, William E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024935:00001


This item has the following downloads:


Full Text

PAGE 1

FLAMEBLOWOUTANDPOLLUTANTEMISSIONSINVITIATEDCOMBUSTI ON OFCONVENTIONALANDBIO-DERIVEDFUELS By BHUPINDERSINGH ADISSERTATIONPRESENTEDTOTHEGRADUATESCHOOL OFTHEUNIVERSITYOFFLORIDAINPARTIALFULFILLMENT OFTHEREQUIREMENTSFORTHEDEGREEOF DOCTOROFPHILOSOPHY UNIVERSITYOFFLORIDA 2009 1

PAGE 2

c r 2009BhupinderSingh 2

PAGE 3

Idedicatethisworktomyparents 3

PAGE 4

ACKNOWLEDGMENTS IwouldliketothankDr.WilliamE.Lear,myadvisor,whohash elpedmethrough allthechallengesduringthecourseofmystudyatUniversit yofFlorida.Iwouldalso liketothankmycommitteemembers,Dr.JamesF.Klausner,Dr .DavidW.Hahn, Dr.DavidW.MikolaitisandDr.JohnEyler,andmyLabManager ,JohnCrittenden, whoI'vealwayslookeduptofortechnicaladvice.Mycolleag uesattheEnergyandGas DynamicsLaboratoryhavebeeninstrumentalinthesuccesso fmywork,throughtheir cooperationandtheirvaluabletechnicalinputs.Myapprec iationalsogoesouttoMark HarrisofFloridaTurbineTechnologies(FTT)forhistechni calsuggestions,andFlorida DepartmentofEnvironmentalProtection(FDEP)fortheirsu pportviaEnergyGrant numberS0323.Lastly,I'mindebtedtomyfriendsandfamilyf ortheirconstantsupport andencouragement. 4

PAGE 5

TABLEOFCONTENTS page ACKNOWLEDGMENTS .................................4 LISTOFTABLES .....................................10 LISTOFFIGURES ....................................11 ABSTRACT ........................................23 CHAPTER NOMENCLATURE ....................................25 1INTRODUCTION ..................................29 1.1Background ...................................29 1.2Motivation ....................................32 1.3ResearchObjectives ...............................37 2LITERATUREREVIEW ..............................40 2.1SemiclosedCycleEnginesandthePower,WaterExtractio nandRefrigeration (PoWER)Engine ................................40 2.2FlamelessCombustion .............................44 2.2.1EectofDiluentGas ..........................54 2.2.2RoleofAutoignitionTemperature ...................55 2.2.3EectofVitiationonSootFormation .................55 2.2.4StabilityinFlamelessCombustion ...................57 2.2.5BiofuelVitiatedCombustion ......................60 2.3CombustionofAlternateFuels .........................61 2.3.1Syngas ..................................61 2.3.2Biodiesel .................................62 2.3.3PalmMethylEster(PME) .......................63 2.3.4RapeseedMethylEther(RME) ....................63 2.3.5DimethylEther(DME) .........................64 2.3.6Ethanol .................................65 2.4SootFormation .................................65 2.4.1SootStructure,PropertiesandHealthconcerns ............65 2.4.2FactorsAectingSootFormation ...................67 2.4.3SootFormationStudies .........................70 2.4.4SootFormationModels .........................72 2.4.5SootFormationChemistry .......................76 2.4.6SootNucleation .............................78 2.4.7SootGrowthandOxidation ......................79 5

PAGE 6

3DESCRIPTIONOFMODEL ............................81 3.1SootFormationModel .............................81 3.1.1MethodofMoments ...........................82 3.1.2GasPhaseKinetics ...........................85 3.1.3PAHPlanarGrowth ..........................86 3.1.4PaticleCoagulation ...........................93 3.1.4.1Continuumregime ......................94 3.1.4.2Free-molecularregime ....................96 3.1.4.3Transitionregime .......................99 3.1.4.4Interpolativeclosure .....................99 3.1.5PaticleAggregation ...........................100 3.1.6SurfaceChemistry-GrowthandOxidation ..............100 3.1.6.1PAHcondensation ......................107 3.1.6.2Particlenucleation ......................109 3.1.7PredictionofSmokeNumber(SN) ...................111 3.2CombustorCSTRNetworkModel .......................111 3.2.1Software .................................111 3.2.2Model ..................................112 3.2.3GoverningEquations ..........................112 3.2.4ModelInputs ..............................113 3.2.5PerformanceParameters ........................115 3.3ModelingApproach ...............................116 3.3.1CaseStudy1:SyngasandMethaneFuelSimulations ........118 3.3.2CaseStudy2:Biofuelsandn-HeptaneFuelSimulations .......118 3.3.3CaseStudy3:StudyofEGRTemperature/EGRlevelonEth anol BlowoutLimits .............................119 3.3.4CaseStudy4:AStudyoftheEectofEGRTemperature/EG R level/ResidenceTime/EquivalenceRatioonSootFormation(AcetyleneFuelCombustion) ......................119 3.3.5CaseStudy5:SootParticleGrowthandOxidationinVit iated CombustionofAcetyleneFuel .....................120 3.3.6CaseStudy6:ModelingofSemi-ClosedCycleEngineCha racteristics usingaKineticallyEquivalentUnvitiatedOpen-CycleCSTR System 121 3.3.7CaseStudy7:ASensitivityAnalysisStudyonCombusti on PerformanceofEthanolFuelClosetoBlowout ............123 4RESULTSANDDISCUSSION ...........................125 4.1CaseStudy1:SyngasandMethaneFuelSimulations ............125 4.1.1EectofFuelCompositionandEquivalenceRatioatEGR =0 ....125 4.1.2EectofAdiabaticExhaustGasRecirculation(EGR) .......130 4.1.3EectofNon-AdiabaticExhaustGasRecirculation .........138 4.1.4EectofFuelCompositionandEGRonExtinctionLimits .....142 4.1.5EectofFuelCompositionandEGRonSootPrecursorFor mation .146 6

PAGE 7

4.2CaseStudy2:Biofuelsandn-HeptaneFuelSimulations ...........147 4.3CaseStudy3:StudyofEGRTemperature/EGRlevelonEthan olBlowout Limits ......................................157 4.4CaseStudy4:AStudyoftheEectofEGRTemperature/EGRl evel/ ResidenceTime/EquivalenceRatioonSootFormation(Acety leneFuel Combustion) ...................................171 4.4.1EectofEGR ..............................171 4.4.2EectofEGRTemperature ......................175 4.4.3EectofEquivalenceRatio .......................180 4.4.4EectofResidenceTime ........................186 4.5CaseStudy5:SootParticleGrowthandOxidationinVitia tedCombustion 191 4.6ComparisonofFrenklachandRichterModelsofSootForma tion ......202 4.7CaseStudy6:ModelingofSemi-ClosedCycleEngineChara cteristicsusing aKineticallyEquivalentUnvitiatedOpen-CycleCSTRSyste m .......210 4.7.1SemiclosedCase .............................210 4.7.2OpenCycleCase ............................210 4.7.3MatchingConditions ..........................211 4.7.3.1CaseA:AnadiabaticOCsystemequivalenttoanSCC system,withlowerreactantinlettemperaturesandlargerresidencetimewithmatchedfreshreactantrows .....211 4.7.3.2CaseB:AnadiabaticOCsystemequivalenttoanSCC system,withlowerreactantinlettemperaturesandmatchedresidencetime .........................212 4.7.3.3CaseC:Anon-adiabaticOCsystemequivalenttoanSC C system,withapositiveheatlossandlargerresidencetimewithmatchedfreshreactantrows. .............213 4.7.3.4CaseD:Anon-adiabaticOCsystemequivalenttoanSC C system,withapositiveheatlossandmatchedresidencetime ..............................214 4.8CaseStudy7:ChemicalKineticSensitivityAnalysisatt heBlowoutLimit forEthanolFuel .................................219 5SUMMARYANDCONCLUSIONS .........................223 5.1CaseStudy1:SyngasandMethaneFuelSimulations ............224 5.2CaseStudy2:Biofuelsandn-HeptaneFuelSimulations ...........225 5.3CaseStudy3:StudyofEGRTemperature/EGRlevelonEthan olBlowout Limits ......................................226 5.4CaseStudy4:AStudyoftheEectofEGRTemperature/EGRl evel/ ResidenceTime/EquivalenceRatioonSootFormation(Acety leneFuel Combustion) ...................................226 5.5CaseStudy5:SootParticleGrowthandOxidationinVitia tedCombustion ofAcetyleneFuel ................................228 5.6CaseStudy6:ModelingofSemi-ClosedCycleEngineChara cteristicsUsing aKineticallyEquivalentUnvitiatedOpen-CycleCSTRSyste m .......228 7

PAGE 8

5.7CaseStudy7:ChemicalKineticSensitivityAnalysisatt heBlowoutLimit forEthanolFuel .................................228 5.8ProposedFutureStudies ............................229 APPENDIX AGOVERNINGEQUATIONSOFCSTRMODEL .................230 A.1TransientCSTR-SingleStreamModel ....................230 A.1.1ReactionKineticsoftheGasMixture .................230 A.1.1.1Equilibriumconstants ....................231 A.1.1.2Rateofreaction ........................231 A.1.1.3Speciesproductionrate ...................232 A.1.2SpeciesConservation ..........................232 A.1.2.1Massbasis ...........................232 A.1.2.2Molarbasis ..........................233 A.1.2.3Concentrationbasis ......................233 A.1.3ConservationofTotalMass .......................233 A.1.3.1Massbasis ...........................233 A.1.3.2Molarbasis ..........................233 A.1.4ConservationofEnergy .........................234 A.1.4.1Massbasis ...........................234 A.1.4.2Molarbasis ..........................237 A.2TransientCSTR-Multi-streamModel ....................238 A.2.1SpeciesConservation ..........................238 A.2.1.1Massbasis ...........................238 A.2.1.2Molarbasis ..........................238 A.2.2ConservationofTotalMass .......................239 A.2.2.1Massbasis ...........................239 A.2.2.2Molarbasis ..........................239 A.2.3ConservationofEnergy .........................240 A.2.3.1Massbasis ...........................240 A.2.3.2Molarbasis ..........................240 BTHERMODYNAMICPROPERTIESOFGASMIXTURES ...........241 B.1SpecicHeats ..................................241 B.1.1CaloricallyIdealGases .........................241 B.1.2CaloricallyNon-IdealGases ......................241 B.2Enthalpy .....................................241 B.2.1ConstantC P Parametrization .....................241 B.2.2NASAParametrization .........................241 B.2.3ShomateParametrization ........................242 B.3Entropy .....................................242 B.3.1ConstantC P Parametrization .....................242 B.3.2NASAParametrization .........................242 8

PAGE 9

B.3.3ShomateParametrization ........................242 CMETHODOFMOMENTS .............................243 C.1BasicStatistics .................................243 C.2LagrangianInterpolation ............................244 REFERENCES .......................................246 BIOGRAPHICALSKETCH ................................267 9

PAGE 10

LISTOFTABLES Table page 3-1Combustionmechanisms ...............................124 4-1SensitivityAnalysisforblowoutresidencetime ...................220 10

PAGE 11

LISTOFFIGURES Figure page 1-1SchematicofaPower,WaterExtractionandRefrigeratio nCycleEngine ....39 3-1SootFormation-ParticulateModel ........................81 3-2SootFormation-GasPhaseKinetics .......................82 3-3LinearlyLumpedSequentialReactionsofPAHMolecule .............87 3-4ProcessofSootCoagulation ............................94 3-5ProcessofSootAggregation ............................100 3-6ProcessofSootSurfaceReactions .........................101 3-7EquivalentSCCsystems.(A)OCandSCC(withnon-adiabat icEGR)CSTR systems,(B)EquivalentSCCsystemwithadiabaticEGRandre ducedairinlet temperature,(C)Equivalentnon-adiabaticOCsystem. .............122 3-8Semi-ClosedCycleCSTRNetworkModel .....................123 3-9ModelInputs .....................................124 4-1ComparisonoframetemperaturesforfuelsMethane,Syn7 5,Syn50andSyn25 asafunctionofequivalenceratioatEGR=0.0, res =5.0ms,P=202.65kPa (2atm),T A =400K. ................................126 4-2ComparisonofcombustioneciencyforfuelsMethane,Sy n75,Syn50andSyn25 asafunctionofequivalenceratioatEGR=0.0, res =5.0ms,P=202.65kPa (2atm),T A =400K. ................................127 4-3ComparisonofCOemissionindexforfuelsMethane,Syn75 ,Syn50andSyn25 asafunctionofequivalenceratioatEGR=0.0, res =5.0ms,P=202.65kPa (2atm),T A =400K. ................................127 4-4ComparisonofUHCemissionindexforfuelsMethane,Syn7 5,Syn50andSyn25 asafunctionofequivalenceratioatEGR=0.0, res =5.0ms,P=202.65kPa (2atm),T A =400K. ................................128 4-5VariationofUHCemissionindexforfuelsMethane,Syn75 ,Syn50andSyn25 asafunctionofcombustioneciencyatEGR=0.0, res =5.0ms,P=202.65 kPa(2atm), =0.4,0.5,0.6,0.7,0.8,0.9,1.0(fuel-leanside)and =1.2,1.6, 2.0,3.0(fuel-richside),T A =400K. ........................128 4-6VariationofcombustioneciencyforfuelsMethane,Syn 75,Syn50andSyn25 asafunctionoframetemperatureatEGR=0.0, res =5.0ms,P=202.65 kPa(2atm), =0.4,0.6,0.8,1.0(fuel-leanside)and =1.2,1.6,2.0,3.0 (fuel-richside),T A =400K. ............................129 11

PAGE 12

4-7VariationofcombustioneciencyforfuelsSyn75,Syn50 andSyn25asafunction oframetemperatureatEGR=0.0and2.0, res =5.0ms,P=202.65kPa(2 atm),T A =400K. .................................130 4-8VariationoframetemperatureforfuelsSyn75,Syn50and Syn25asafunction ofadiabaticEGRat =0.4, res =5.0ms,P=202.65kPa(2atm),T A = 400K. ........................................131 4-9VariationofcombustioneciencyforfuelsSyn75,Syn50 andSyn25asafunction ofadiabaticEGRat =0.4, res =5.0ms,P=202.65kPa(2atm),T A = 400K. ........................................132 4-10VariationofUHCemissionindexforfuelsSyn75,Syn50a ndSyn25asafunction ofadiabaticEGRat =0.4, res =5.0ms,P=202.65kPa(2atm),T A = 400K. ........................................132 4-11VariationofCOemissionindexforfuelsSyn75,Syn50an dSyn25asafunction ofadiabaticEGRat =0.4, res =5.0ms,P=202.65kPa(2atm),T A = 400K. ........................................133 4-12VariationoframetemperatureforfuelsSyn75,Syn50an dSyn25asafunction ofadiabaticEGRat =1.0, res =5.0ms,P=202.65kPa(2atm),T A = 400K. ........................................133 4-13VariationofcombustioneciencyforfuelsSyn75,Syn5 0andSyn25asafunction ofadiabaticEGRat =1.0, res =5.0ms,P=202.65kPa(2atm),T A = 400K. ........................................134 4-14VariationofUHCemissionindexforfuelsSyn75,Syn50a ndSyn25asafunction ofadiabaticEGRat =1.0, res =5.0ms,P=202.65kPa(2atm),T A = 400K. ........................................134 4-15VariationofCOemissionindexforfuelsSyn75,Syn50an dSyn25asafunction ofadiabaticEGRat =1.0, res =5.0ms,P=202.65kPa(2atm),T A = 400K. ........................................135 4-16VariationoframetemperatureforfuelsSyn75,Syn50an dSyn25asafunction ofadiabaticEGRat =1.6, res =5.0ms,P=202.65kPa(2atm),T A = 400K. ........................................135 4-17VariationoframetemperatureforfuelsSyn75,Syn50an dSyn25asafunction ofadiabaticEGRat =1.6, res =5.0ms,P=202.65kPa(2atm),T A = 400K. ........................................136 4-18VariationofUHCemissionindexforfuelsSyn75,Syn50a ndSyn25asafunction ofadiabaticEGRat =1.6, res =5.0ms,P=202.65kPa(2atm),T A = 400K. ........................................136 12

PAGE 13

4-19VariationofCOemissionindexforfuelsSyn75,Syn50an dSyn25asafunction ofadiabaticEGRat =1.6, res =5.0ms,P=202.65kPa(2atm),T A = 400K. ........................................137 4-20ComparisonoframetemperatureforSyn25fuelwithEGRo f0.0,0.2and0.5 asafunctionofequivalenceratioatEGRT=1200K, res =5.0ms,P=202.65 kPa(2atm),T A =400K. .............................138 4-21ComparisonofcombustioneciencyforSyn25fuelwithE GRof0.0,0.2and 0.5asafunctionofequivalenceratioatEGRT=1200K, res =5.0ms,P= 202.65kPa(2atm),T A =400K. .........................139 4-22ComparisonofCOemissionindexforSyn25fuelwithEGRo f0.0,0.2and0.5 asafunctionofequivalenceratioatEGRT=1200K, res =5.0ms,P=202.65 kPa(2atm),T A =400K. .............................139 4-23ComparisonofUHCemissionindexforSyn25fuelwithEGR of0.0,0.2and 0.5asafunctionofequivalenceratioatEGRT=1200K, res =5.0ms,P= 202.65kPa(2atm),T A =400K. .........................140 4-24ComparisonofUHCemissionindexforSyn25fuelwithEGR of0.0,0.2and 0.5asafunctionofcombustioneciencyatEGRT=1200K, res =5.0ms,P =202.65kPa(2atm),T A =400K, =0.4,0.6,0.8,1.0(fuel-leanside)and =1.2,1.6,2.0,3.0(fuel-richside). .........................140 4-25ComparisonofcombustioneciencyforSyn25fuelwithE GRof0.0,0.2and 0.5asafunctionoframetemperatureatEGRT=1200K, res =5.0ms,P =202.65kPa(2atm),T A =400K, =0.4,0.6,0.8,1.0(fuel-leanside)and =1.2,1.6,2.0,3.0(fuel-richside). .........................141 4-26VariationofCOemissionindexforSyn25fuelwithEGRof 0.0,0.2and0.5 asafunctionofcombustioneciencyatEGRT=1200K, res =5.0ms,P =202.65kPa(2atm),T A =400K, =0.4,0.6,0.8,1.0(fuel-leanside)and =1.2,1.6,2.0,3.0(fuel-richside). .........................141 4-27VariationofblowoutequivalenceratioforMethanefue lwithEGR=0.0, MethanefuelwithEGR=2.0(adiabatic),Syn50fuelwithEGR= 0.0,Syn50 fuelwithEGR=2.0(adiabatic)andSyn25fuelwithEGR=0.0as afunction ofblowoutloadingparameter(LP)atP=202.65kPa(2atm),T A =400K. .142 4-28VariationofblowoutequivalenceratioforMethanefue lwithEGR=0.0, MethanefuelwithEGR=2.0(adiabatic),Syn50fuelwithEGR= 0.0,Syn50 fuelwithEGR=2.0(adiabatic)andSyn25fuelwithEGR=0.0as afunction ofblowoutresidencetimeatP=202.65kPa(2atm),T A =400K. .......143 4-29VariationofblowoutequivalenceratioforMethanefue lforEGR=0.0,EGR =0.2(EGRT=1200K),EGR=0.5(EGRT=1200K)andEGR=2.0(adia batic) asafunctionofblowoutLPatP=202.65kPa(2atm),T A =400K. .....144 13

PAGE 14

4-30VariationofblowoutrametemperatureforMethanefuel forEGR=0.0,EGR =0.2(EGRT=1200K),EGR=0.5(EGRT=1200K)andEGR=2.0(adia batic) asafunctionofblowoutequivalenceratioatP=202.65kPa(2 atm),T A = 400K. ........................................144 4-31VariationofreactantmixtureinlettemperatureforMe thanefuelforEGR= 0.0,EGR=0.2(EGRT=1200K),EGR=0.5(EGRT=1200K)andEGR=2.0(adiabatic),T A =400K,P=202.65kPa(2atm)asafunctionofblowout equivalenceratio. ..................................145 4-32VariationofpyreneemissionindexforMethanefuelwit hEGR=0.0,EGR= 1.0(adiabatic)andEGR=2.0(adiabatic)asafunctionofblo woutequivalence ratioatP=202.65kPa(2atm). ..........................146 4-33Comparisonofblowoutrametemperatureasafunctionof blowoutequivalence ratioforEthanol,DME,MB,n-HeptanefuelsatEGR=0.0and2. 0, T A = EGRT=700K,P=202.65kPa(2atm). .....................148 4-34ComparisonofblowoutCOemissionindexasafunctionof blowoutequivalence ratioforEthanol,DME,MB,n-HeptanefuelsatEGR=0.0and2. 0, T A = EGRT=700K,P=202.65kPa(2atm). .....................149 4-35ComparisonofblowoutUHCemissionindexasafunctiono fblowoutequivalence ratioforEthanol,DME,MB,n-HeptanefuelsatEGR=0.0and2. 0, T A = EGRT=700K,P=202.65kPa(2atm). .....................150 4-36Comparisonofblowoutequivalenceratioasafunctiono fblowoutloading parameterforEthanol,DME,MB,n-HeptanefuelsatEGR=0.0a nd2.0, T A =EGRT=700K,P=202.65kPa(2atm). ...................151 4-37Comparisonofblowoutresidencetimeasafunctionofbl owoutequivalenceratio forEthanol,DME,MB,n-HeptanefuelsatEGR=0.0and2.0, T A =EGRT =700K,P=202.65kPa(2atm). .........................152 4-38Comparisonofblowoutequivalenceratioasafunctiono fblowoutmassrow rateforEthanol,DME,MB,n-HeptanefuelsatEGR=0.0and2.0 ,EGRT= 700K,P=202.65kPa(2atm). ..........................153 4-39Comparisonoframetemperatureasafunctionofequival enceratioforEthanol, DME,MB,n-HeptanefuelsatEGR=0.0and2.0, T A =EGRT=700K, res =10.0ms,P=202.65kPa(2atm). ........................153 4-40Comparisonofcombustioneciencyasafunctionofequi valenceratiofor Ethanol,DME,MB,n-HeptanefuelsatEGR=0.0and2.0, T A =EGRT= 700K, res =10.0ms,P=202.65kPa(2atm). .................154 14

PAGE 15

4-41ComparisonofCOemissionindexasafunctionofequival enceratiofor Ethanol,DME,MB,n-HeptanefuelsatEGR=0.0and2.0, T A =EGRT= 700K, res =10.0ms,P=202.65kPa(2atm). .................154 4-42ComparisonofUHCemissionindexasafunctionofequiva lenceratiofor Ethanol,DME,MB,n-HeptanefuelsatEGR=0.0and2.0, T A =EGRT= 700K, res =10.0ms,P=202.65kPa(2atm). .................155 4-43ComparisonofUHCemissionindexasafunctionofcombus tioneciencyfor Ethanol,DME,MB,n-HeptanefuelsatEGR=0.0and2.0, T A =EGRT= 700K, res =10.0ms,P=202.65kPa(2atm), =0.4-1.0. ...........155 4-44Comparisonofcombustioneciencyasafunctionoframe temperaturefor Ethanol,DME,MB,n-HeptanefuelsatEGR=0.0and2.0, T A =EGRT= 700K, res =10.0ms,P=202.65kPa(2atm), =0.4-1.0. ...........156 4-45Comparisonofblowoutrametemperatureforethanolfue lasafunctionofEGR temperatureatEGR=0.0,0.5,1.0,1.5,2.0, =1.0,P=202.65kPa(2atm). 158 4-46Comparisonofblowoutloadingparameterforethanolfu elasafunctionofEGR temperatureatEGR=0.0,0.5,1.0,1.5,2.0, =1.0,P=202.65kPa(2atm). 159 4-47Comparisonofblowoutresidencetimeforethanolfuela safunctionofEGR temperatureatEGR=0.0,0.5,1.0,1.5,2.0, =1.0,P=202.65kPa(2atm). 161 4-48Comparisonofblowoutmassrowrateforethanolfuelasa functionofEGR temperatureatEGR=0.0,0.5,1.0,1.5,2.0, =1.0,P=202.65kPa(2atm). 161 4-49Comparisonofblowoutresidencetimeforethanolfuela safunctionofOH speciesmassfractionatEGR=0.0,0.5,1.0,1.5,2.0,EGRT=4 00,600,800, 1000,1200K, =1.0,P=202.65kPa(2atm). .................162 4-50Comparisonofblowoutresidencetimeforethanolfuela safunctionofH speciesmassfractionatEGR=0.0,0.5,1.0,1.5,2.0,EGRT=4 00,600,800, 1000,1200K, =1.0,P=202.65kPa(2atm). .................162 4-51Comparisonofblowoutresidencetimeforethanolfuela safunctionofCO speciesmassfractionatEGR=0.0,0.5,1.0,1.5,2.0,EGRT=4 00,600,800, 1000,1200K, =1.0,P=202.65kPa(2atm). .................163 4-52Comparisonofnormalizedresidencetimeforethanolfu elasafunctionof normalizedrametemperatureatEGR=0.0,0.5,1.0,1.5,2.0, EGRT=400, 600,800,1000,1200K, =1.0,P=202.65kPa(2atm). ............163 4-53Comparisonofnormalizedblowoutloadingparameterfo rethanolfuelasa functionofnormalizedrametemperatureatEGR=0.0,0.5,1. 0,1.5,2.0,EGRT =400,600,800,1000,1200K, =1.0,P=202.65kPa(2atm). ........164 15

PAGE 16

4-54Comparisonofnormalizedblowoutloadingparameterfo rethanolfuelasa functionofnormalizedOHspeciesmassfractionatEGR=0.0, 0.5,1.0,1.5, 2.0,EGRT=400,600,800,1000,1200K, =1.0,P=202.65kPa(2atm). .164 4-55Comparisonofnormalizedblowoutloadingparameterfo rethanolfuelasa functionofnormalizedCOspeciesmassfractionatEGR=0.0, 0.5,1.0,1.5, 2.0,EGRT=400,600,800,1000,1200K, =1.0,P=202.65kPa(2atm). .165 4-56Comparisonofignitiondelayforethanolfuelasafunct ionofequilibriumrame temperatureatEGR=0.0,0.5,1.0,1.5,2.0,reactantmixtur einlettemperature 1200to2500K, =1.0,P=202.65kPa(2atm). ................165 4-57Comparisonofignitiondelayforethanolfuelasafunct ionofreactantmixture inlettemperatureatEGR=0.0,0.5,1.0,1.5,2.0, =1.0,P=202.65kPa(2 atm). .........................................166 4-58Variationofignitiondelayforethanolfuelasafuncti onofEGRatreactant mixtureinlettemperature=2600K, =1.0,P=202.65kPa(2atm). ....166 4-59Comparisonofblowoutresidencetimeforethanolfuela safunctionofignition delayatEGR=0.0,0.5,1.0,1.5,2.0,EGRT=400,600,800,100 0,1200K, =1.0,P=202.65kPa(2atm). ..........................167 4-60Comparisonofblowoutloadingparameterasafunctiono fignitionnumberat EGR=0.0,0.5,1.0,1.5,2.0,EGRT=400,600,800,1000,1200K =1.0,P =202.65kPa(2atm). ................................167 4-61Comparisonofnormalizedblowoutloadingparameteras afunctionof normalizedignitionnumberatEGR=0.0,0.5,1.0,1.5,2.0,E GRT=400,600, 800,1000,1200K, =1.0,P=202.65kPa(2atm). ...............168 4-62Comparisonofblowoutrametemperatureasafunctionof airstream/EGR temperatureatEGR=0.0and2.0,P=2.0,1.5,1.0,0.5atmand =1.0. ...169 4-63Comparisonofblowoutloadingparameterasafunctiono fairstream/EGR temperatureatEGR=0.0and2.0,P=2.0,1.5,1.0,0.5atmand =1.0. ...169 4-64Comparisonofblowoutresidencetimeasafunctionofai rstream/EGR temperatureatEGR=0.0and2.0,P=2.0,1.5,1.0,0.5atmand =1.0. ...170 4-65Comparisonofblowoutmassrowrateasafunctionofairs tream/EGR temperatureatEGR=0.0and2.0,P=2.0,1.5,1.0,0.5atmand =1.0. ...170 4-66Variationoframetemperatureforacetylenefuelasafu nctionofEGRatEGRT =1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). ...........172 4-67VariationofUHCemissionindexforacetylenefuelasaf unctionofEGRat EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). .......172 16

PAGE 17

4-68VariationofCOemissionindexforacetylenefuelasafu nctionofEGRatEGRT =1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). ...........173 4-69VariationofC 2 H 2 emissionindexforacetylenefuelasafunctionofEGRat EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). .......173 4-70Variationofcombustioneciencyforacetylenefuelas afunctionofEGRat EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). .......174 4-71ComparisonofPAHemissionindicesforacetylenefuela safunctionofEGR atEGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). ......174 4-72Comparisonoframetemperatureforacetylenefuelasaf unctionof airstream/EGRtemperatureatEGR=0.0andEGR=2.0, =2.0, res = 1.0ms,P=202.65kPa(2atm). ..........................175 4-73ComparisonofUHCemissionindexforacetylenefuelasa functionof airstream/EGRtemperatureatEGR=0.0andEGR=2.0, =2.0, res = 1.0ms,P=202.65kPa(2atm). ..........................176 4-74ComparisonofCOemissionindexforacetylenefuelasaf unctionof airstream/EGRtemperatureatEGR=0.0andEGR=2.0, =2.0, res = 1.0ms,P=202.65kPa(2atm). ..........................177 4-75ComparisonofC 2 H 2 emissionindexforacetylenefuelasafunctionof airstream/EGRtemperatureatEGR=0.0andEGR=2.0, =2.0, res = 1.0ms,P=202.65kPa(2atm). ..........................177 4-76Comparisonofcombustioneciencyforacetylenefuela safunctionof airstream/EGRtemperatureatEGR=0.0andEGR=2.0, =2.0, res = 1.0ms,P=202.65kPa(2atm). ..........................178 4-77ComparisonofPAHemissionindicesforacetylenefuela safunctionof airstream/EGRtemperatureatEGR=0.0, =2.0, res =1.0ms,P=202.65 kPa(2atm). .....................................178 4-78ComparisonofPAHemissionindicesforacetylenefuela safunctionofEGR temperatureatEGR=2.0, =2.0, res =5.0ms. ................179 4-79ComparisonofPAHemissionindicesforacetylenefuela safunctionofEGR temperatureatEGR=0.0, =2.0, res =5.0ms. ................179 4-80Comparisonoframetemperatureforacetylenefuelasaf unctionof equivalenceratioatEGR=0.0andEGR=2.0,EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). .........................181 4-81ComparisonofUHCemissionindexforacetylenefuelasa functionof equivalenceratioatEGR=0.0andEGR=2.0,EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). .........................182 17

PAGE 18

4-82ComparisonofCOemissionindexforacetylenefuelasaf unctionof equivalenceratioatEGR=0.0andEGR=2.0,EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). .........................182 4-83ComparisonofC 2 H 2 emissionindexforacetylenefuelasafunctionof equivalenceratioatEGR=0.0andEGR=2.0,EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). .........................183 4-84Comparisonofcombustioneciencyforacetylenefuela safunctionof equivalenceratioatEGR=0.0andEGR=2.0,EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). .........................183 4-85ComparisonofPAHemissionindicesforacetylenefuela safunctionof equivalenceratioatEGR=0.0andEGR=2.0,EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). .........................184 4-86ComparisonofPAHemissionindicesforacetylenefuela safunctionof equivalenceratioatEGR=0.0andEGR=2.0,EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). .........................185 4-87Comparisonoframetemperatureforacetylenefuelasaf unctionofresidence timeatEGR=0.0andEGR=2.0, =2.0,EGRT=1000K,P=202.65 kPa(2atm). .....................................186 4-88ComparisonofUHCemissionindexforacetylenefuelasa functionofresidence timeatEGR=0.0andEGR=2.0, =2.0,EGRT=1000K,P=202.65 kPa(2atm). .....................................187 4-89ComparisonofCOemissionindexforacetylenefuelasaf unctionofresidence timeatEGR=0.0andEGR=2.0, =2.0,EGRT=1000K,P=202.65 kPa(2atm). .....................................187 4-90ComparisonofC 2 H 2 emissionindexforacetylenefuelasafunctionofresidence timeatEGR=0.0andEGR=2.0, =2.0,EGRT=1000K,P=202.65 kPa(2atm). .....................................188 4-91Comparisonofcombustioneciencyforacetylenefuela safunctionofresidence timeatEGR=0.0andEGR=2.0, =2.0,EGRT=1000K,P=202.65 kPa(2atm). .....................................188 4-92ComparisonofPAHemissionindicesforacetylenefuela safunctionofresidence timeatEGR=0.0, =2.0,EGRT=1000K,P=202.65kPa(2atm). ....189 4-93ComparisonofPAHemissionindicesforacetylenefuela safunctionofresidence timeatEGR=2.0, =2.0,EGRT=1000K,P=202.65kPa(2atm). ....189 4-94ComparisonofPAHemissionindicesforacetylenefuela safunctionofresidence timeatEGR=0.0andEGR=2.0, =2.0,EGRT=1000K,P=202.65 kPa(2atm). .....................................190 18

PAGE 19

4-95Comparisonofsootvolumefractionforacetylenefuela safunctionoframe temperatureatEGR=0.0andEGR=2.0, =2.0, res =1.0ms,P=202.65 kPa(2atm). .....................................192 4-96Comparisonofaveragesootparticlediameterforacety lenefuelasafunction oframetemperatureatEGR=0.0andEGR=2.0, =2.0, res =1.0ms,P =202.65kPa(2atm). ................................192 4-97Comparisonofaveragesootparticlesurfaceareaforac etylenefuelasafunction oframetemperatureatEGR=0.0andEGR=2.0, =2.0, res =1.0ms,P =202.65kPa(2atm). ................................193 4-98Comparisonofzero th sootmomentproductionratecontributions(nucleation, coagulation,surfacegrowth)foracetylenefuelasafuncti onoframe temperatureat =2.0,EGR=0.0, res =1.0ms,P=202.65kPa(2atm). .194 4-99Comparisonofzero th sootmomentproductionratecontributions(nucleation, coagulation,surfacegrowth)foracetylenefuelasafuncti onoframe temperatureat =2.0,EGR=2.0, res =1.0ms,P=202.65kPa(2atm). .195 4-100Comparisonofzero th sootmomentproductionratecontributions(nucleation, coagulation,surfacegrowth)foracetylenefuelasafuncti onoframe temperatureatEGR=0.0andEGR=2.0, =2.0, res =1.0ms,P=202.65 kPa(2atm). .....................................195 4-101Comparisonofrstsootmomentproductionratecontri butions(nucleation, coagulation,surfacegrowth)foracetylenefuelasafuncti onoframe temperatureat =2.0,EGR=0.0, res =1.0ms,P=202.65kPa(2atm). .196 4-102Comparisonofrstsootmomentproductionratecontri butions(nucleation, coagulation,surfacegrowth)foracetylenefuelasafuncti onoframe temperatureat =2.0,EGR=2.0, res =1.0ms,P=202.65kPa(2atm). .196 4-103Comparisonofrstsootmomentproductionratecontri butions(nucleation, coagulation,surfacegrowth)foracetylenefuelasafuncti onoframe temperatureatEGR=0.0andEGR=2.0, =2.0, res =1.0ms,P=202.65 kPa(2atm). .....................................197 4-104ComparisonofA i (PAHs)emissionindicesforacetylenefuelasafunctionof rametemperatureat =2.0,EGR=0.0, res =1.0ms,P=202.65kPa(2 atm). .........................................198 4-105ComparisonofA i (PAHs)emissionindicesforacetylenefuelasafunctionof rametemperatureat =2.0,EGR=2.0, res =1.0ms,P=202.65kPa(2 atm). .........................................198 19

PAGE 20

4-106ComparisonofA i (PAHs)emissionindicesforacetylenefuelasafunctionof rametemperatureat =2.0,EGR=0.0andEGR=2.0, res =1.0ms,P =202.65kPa(2atm). ................................199 4-107Comparisonofsurfacegrowthrateforacetylenefuela safunctionoframe temperatureat =2.0,EGR=0.0, res =1.0ms,P=202.65kPa(2atm). .200 4-108Comparisonofsurfacegrowthratecontributions(PAH condensation,C 2 H 2 addition,O 2 oxidation,OHoxidation)foracetylenefuelasafunctionof rame temperatureat =2.0,EGR=2.0, res =1.0ms,P=202.65kPa(2atm). .200 4-109Comparisonofsurfacegrowthratecontributions(PAH condensation,C 2 H 2 addition,O 2 oxidation,OHoxidation)andOHoxidationreactionsforace tylene fuelasafunctionoframetemperatureat =2.0,EGR=0.0andEGR= 2.0, res =1.0ms,P=202.65kPa(2atm). ....................201 4-110Comparisonofsootmassdensityforacetylenefuelasa functionoframe temperatureat =2.0,EGR=2.0, res =1.0ms,P=202.65kPa(2atm). .201 4-111ComparisonofUHCemissionindexforacetylenefuelas afunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandF renklach mechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). ..........202 4-112ComparisonofCOemissionindexforacetylenefuelasa functionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandF renklach mechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). ..........203 4-113ComparisonofC 2 H 2 emissionindexforacetylenefuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandF renklach mechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). ..........204 4-114Comparisonofcombustioneciencyforacetylenefuel asafunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandF renklach mechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). ..........204 4-115ComparisonofA 1 emissionindexforacetylenefuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandF renklach mechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). ..........205 4-116ComparisonofA 2 emissionindexforacetylenefuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandF renklach mechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). ..........205 4-117ComparisonofA 3 emissionindexforacetylenefuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandF renklach mechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). ..........206 20

PAGE 21

4-118ComparisonofA 4 emissionindexforacetylenefuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandF renklach mechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). ..........206 4-119Comparisonofbinmassfractionsforacetylenefuelas afunctionoframe temperatureatEGR=0.0andEGR=2.0,EGRT=1500K, =2.0, res = 1.0ms,P=202.65kPa(2atm). ..........................207 4-120Comparisonofbinmassfractionsforacetylenefuelas afunctionoframe temperatureatEGR=0.0andEGR=2.0,EGRT=1600K, =2.0, res = 1.0ms,P=202.65kPa(2atm). ..........................207 4-121Comparisonofbinmassfractionsforacetylenefuelas afunctionoframe temperatureatEGR=0.0andEGR=2.0,EGRT=1700K, =2.0, res = 1.0ms,P=202.65kPa(2atm). ..........................208 4-122Comparisonofbinmassfractionsforacetylenefuelas afunctionoframe temperatureatEGR=0.0andEGR=2.0,EGRT=1800K, =2.0, res = 1.0ms,P=202.65kPa(2atm). ..........................208 4-123Comparisonofbinmassfractionsforacetylenefuelas afunctionoframe temperatureatEGR=0.0andEGR=2.0,EGRT=1900K, =2.0, res = 1.0ms,P=202.65kPa(2atm). ..........................209 4-124Comparisonofbinmassfractionsforacetylenefuelas afunctionoframe temperatureatEGR=0.0andEGR=2.0,EGRT=2000K, =2.0, res = 1.0ms,P=202.65kPa(2atm). ..........................209 4-125Comparisonoframetemperatureforethanolfuelasafu nctionof airstream/EGRtemperatureinSCCsystemandanequivalentu nvitiatedOC systematEGR=0.0andEGR=2.0, =1.0, res =10.0ms. ........216 4-126ComparisonofUHCemissionindexforethanolfuelasaf unctionof airstream/EGRtemperatureinSCCsystemandanequivalentu nvitiatedOC systematEGR=0.0andEGR=2.0, =1.0, res =10.0ms. ........217 4-127ComparisonofCOemissionindexforethanolfuelasafu nctionof airstream/EGRtemperatureinSCCsystemandanequivalentu nvitiatedOC systematEGR=0.0andEGR=2.0, =1.0, res =10.0ms. ........217 4-128ComparisonofOHemissionindexforethanolfuelasafu nctionof airstream/EGRtemperatureinSCCsystemandanequivalentu nvitiatedOC systematEGR=0.0andEGR=2.0, =1.0, res =10.0ms. ........218 4-129Comparisonofblowoutresidencetimeforethanolfuel asafunctionofOH speciesmassfractionatEGR=0.0and2.0,EGRT=400,600,800 ,1000,1200 K, =1.0,P=202.65kPa(2atm). ........................220 21

PAGE 22

4-130NormalizedOHSensitivityIndicesbeforerameblowou tatEGR=0.0. ....221 4-131NormalizedOHSensitivityIndicesbeforerameblowou tatEGR=2.0. ....222 22

PAGE 23

AbstractofDissertationPresentedtotheGraduateSchool oftheUniversityofFloridainPartialFulllmentofthe RequirementsfortheDegreeofDoctorofPhilosophy FLAMEBLOWOUTANDPOLLUTANTEMISSIONSINVITIATEDCOMBUSTI ON OFCONVENTIONALANDBIO-DERIVEDFUELS By BhupinderSingh December2009 Chair:WilliamE.LearMajor:MechanicalEngineering Thewideninggapbetweenthedemandandsupplyoffossilfuel shascatalyzed theexplorationofalternativesourcesofenergy.Interest inthepower,waterextraction andrefrigeration(PoWER)cycle,proposedbytheUniversit yofFlorida,aswellas thedesirabilityofusingbiofuelsindistributedgenerati onsystems,hasmotivatedthe explorationofbiofuelvitiatedcombustion.ThePoWERcycl eisanovelenginecycle conceptthatutilizesvitiationoftheairstreamwithexter nally-cooledrecirculatedexhaust gasesatanintermediatepressureinasemi-closedcycle(SC C)loop,loweringtheoverall temperatureofcombustion.Ithasseveraladvantagesinclu dingfuelrexibility,reduced airrow,lowerrametemperature,compactness,highecienc yatfullandpartload,and lowemissions.Sincethecoreengineairstreamisvitiatedw iththeexternallycooled exhaustgasrecirculation(EGR)stream,thereisaninheren treductioninthecombustion stabilityforaPoWERengine.TheeectofEGRrowandtempera tureoncombustion blowoutstabilityandemissionsduringvitiatedbiofuelco mbustionhasbeencharacterized. Thevitiatedcombustionperformanceofbiofuelsmethylbut anoate,dimethylether,and ethanolhavebeencomparedwithn-heptane,andvaryingcomp ositionsofsyngaswith methanefuel.Inaddition,athighlevelsofEGRasharpreduc tionintherameluminosity hasbeenobservedinourexperimentaltests,indicatingthe onsetoframelesscombustion. Thisdropinluminositymaybearesultofinhibitionofproce ssesleadingtotheformation ofradiativesootparticles.Oneoftheobjectivesofthisst udyisndingtheeectofEGR 23

PAGE 24

onsootformation,withtheultimateobjectiveofbeingable topredicttheboundaries oframelesscombustion.Detailedchemicalkineticsimulat ionswereperformedusinga constant-pressurecontinuouslystirredtankreactor(CST R)networkmodeldeveloped usingtheCanteracombustioncode,implementedinC++.Resu ltshavebeenpresented showingcomparativetrendsinpollutantemissionsgenerat ion,rameblowoutstability,and combustioneciency. 24

PAGE 25

NOMENCLATURE SymbolDescriptionA I Ignitergaussian-pulseamplitude,kg/s A R Reactorsurfacearea,m 2 C a ;C d ;C h ;C r ;C s Constants C j Cunninghamslipcorrectionfactorforsizeclassj P valve Valvepressuredrop,Pa EI j EmissionindexofspeciesS j ,g/kg-fuel f l ;f x;y l Gridfunctionsforcoagulation f v Sootvolumefraction G r Momentproductionrateduetocoagulation,mol/m 3 s h Specicenthalpy,J/kg Kn j Knudsennumberforparticlesizeclassj Kc j Constantforsizeclassj Kc 0j Constantforsizeclassj K v Valvecoecient,kg/s Pa k ratio Constant K Rateconstantmatric,1/mol s K f Constant k j Rateconstant,1/mol s LP Loadingparameter,mole/s m 3 Pa n L C C AromaticC-Cbondlength,nm M j MolecularweightofspeciesS j ,kg/mole M r r th concentrationmoment,mol/m 3 m B Block-massadditioninaHACAmechanism m l IndividualmassadditionreactionstepinaHACAmechanism m i;s Blockispeciess m s MassrowrateofStreams,kg/s m T Totalmassrowrate,kg/s m valve Massrowrateofvalve,kg/s m i Numberofcarbonatomsinasootparticle m s Sootmassdensity,mg/m 3 N j Numberofparticlesofsizeclassj n Constant n s MolarrowrateofStreams,mole/s P CombustorCSTRpressure,m 3 Q int Interfaceheattransferrate,J/s m 2 R Gasconstant,kJ/kg K R m Recirculationratio(massbasis) R n Recirculationratio(molebasis) R r Momentproductionrateduetocoagulation,mol/m 3 s R u UniversalGasConstant,kJ/kmole K 25

PAGE 26

SymbolDescriptionS Speciesconcentrationvector,mol/m 3 s Sootsurfacearea,cm 2 /m 3 SN Smokenumber s j Speciessurfaceproductionrate,mole/s m 2 t 0 Ignitergaussian-pulsepeakingtime,s U Internalenergy,J V R ReactorVolume,m 3 W r Momentproductionrateduetosurfacechemistry,mol/m 3 s Y j;s MassfractionofspeciesS j instreams Greek Air-fuelRatio st Stoichiometricair-fuelRatio c i;j Collisionfrequencyforcoagulation Van-derwaal'scollisionenhancementfactor c Combustioneciency(combinedconversioneciency) Equivalenceratio r r th sizemoment,mol/m 3 I Ignitergaussian-pulsewidth(givenasfull-width-half-maximum),s res Residencetime,s j Speciesproductionrate,mole/s m 3 SubscriptsA Air F Fuel R EGR j Speciesindex Superscriptsc Continuumcoagulationregimeforsootformation f Free-molecularcoagulationregimeforsootformation t Transitioncoagulationregimeforsootformation sc SurfacecondensationofPAHonsootparticules AbbreviationsASUAirseparationunitBOBlowoutCICompressionignitionCFDComputationalruiddynamicsCGCCoolgascooler 26

PAGE 27

AbbreviationsCNGCompressedNaturalGasCCGTCombinedcyclegasturbineCSTRContinuouslystirredtankreactorCRGTChemicallyreformedgasturbineCOSTAIRContinuousstagedaircombustionDMADierntialMobilityAnalyzerDMEDimethylEtherEIEmissionIndexEGRExhaustgasrecirculationEGRTExhaustgasrecirculationtemperatureExVExhaustvalveEPFMEulerianparticlerameletmodelFLOXFlamelessoxidationFWHMFull-width-halfmaximumHGCHotgascoolerHiRCHighRecirculationCombustionHPCHigh-pressurecompressorHPTHigh-pressureturbineHTCHightemperaturecombustionHACAHydrogen-abstraction-acetylene-additionHCDIHomogenousChargeDiusionIgnitionHiTAC/HTACHightemperatureaircombustionHPACHighlypreheatedaircombustionHPRTEHighpressureregenerativeturbineengineHRSGHeatrecoverysteamgeneratorICInternalcombustionJSRJet-stirredreactorIGCCIntegratedgasicationcombinedcycleLPLoadingparameterLDVLaserdoplervelocimetryLNILowNOxInjectionLPCLow-pressurecompressorLPTLow-pressureturbineLTCLowtemperaturecombustionLLNLLawrenceLivermoreNationalLaboratoryLTHRLowtemperatureheatreleaseMKmodulatedkineticsMAIMainairinletMBMethylButanoateMFRMassrowregulatorsMILDModerateandintensivelowoxygendilutedNAAQSUSnationalambientairqualitystandardNOCNano-organiccarbon 27

PAGE 28

AbbreviationsOCOpencyclePMParticulatematterPAHPolyaromatichydrocarbonsPCWPlantcoolingwaterPMEpalmmethylesterPaSRPartiallystirredreactorPLIFPlanarlaserinducedruorescencePoWERPower,waterextractionandrefrigerationRCCRegenerativecombinedcycleRECRecuperatorRMERapeseedoilMethylEtherRFTEregenerativefeedbackturbineengineRPFRRecirculatingPlugFlowReactorSCCSemiclosedcycleSCGTSemiclosedgasturbineSCRSelectivecatalyticreductionSCNRSelectivenon-catalyticreductionSFCSpecicfuelconsumptionSMDSautermeandiameterSOFSolubleOrganicFractionSRCSmokelessrichcombustionTCRThermo-chemicalreformingTEMTransmissionelectronmicroscopyTHCTotalhydrocarbonUCPCultranecondensationparticlecounterUHCUnburnedhydrocarbonsULSDultra-lowsulfurdieselVARSVaporabsorptionrefrigerationsystemWGCWarmgascoolerWSRWell-stirredreactor 28

PAGE 29

CHAPTER1 INTRODUCTION 1.1Background Emissionsreductionandeciencyimprovementhavebeenlon g-standinggoalsfor combustionsystemdesigners.Inthecombustionofconventi onalfuels,emissionsofCO 2 unburnedhydrocarbons(UHC),CO,CO 2 ,sootparticulates,NO x andSO x havebeen ofparticularconcernduetotheirdetrimentalimpactonhea lthandtheenvironment. GreenhousegasessuchasCO 2 ,H 2 O,CH 4 ,N 2 Oandchloro-rourocarbonshavebeenfound tobethemajorcontributorstotheglobalwarmingproblem,a ndthereisanunmistakable consensusthatthereisanurgentneedtocurtailtheanthrop ogeniccontributionofthese gases[ 1 ].AsaresultoftheKyotoProtocol,manycountriesareconsi deringemissions tradingandimpositionoftaxesonCO 2 generation.Severalcountrieshavetaken initiativestoimproveenergyeciencyandharnesspolluti on-freeenergyresourcesand technologies.Renewableenergyresourceslikewind,hydra ulicorsolarenergyarelikelyto reduceoverallemissions;however,thesecannotentirelym eetthegrowingenergydemand. Biofuelshavebeenproposedasashorttermsolutiontoheigh teningenergyandpollution crisis.BiofuelcombustionisconsideredtoreduceCO 2 impactontheenvironment,since thebiomassconsumesCO 2 initsproductioncyclebeforebeingusedasafuel.Using CO 2 sequestrationandstorageisanotherproposedCO 2 reductionconcept.Integrated gasicationcombinedcycle(IGCC)plantsredwithbiomass ,areoneexampleofreduced CO 2 emissionsandcleancombustiontechnologythatemploysgas icationofbiomassto producesyngasanditssubsequentcombustion.CO 2 separationcanbeachievedthrough installationofadditionalequipment,asinopenorsemiclo sedcombinedcyclegasturbine (CCGT)plants,orchemicallyreformedgasturbine(CRGT)pl ants,wherefuelistreated withsteamtoincreaseitshydrogencontentanditssubseque ntoxy-fuelcombustion[ 2 ]. Thermo-chemicalreforming(TCR)mayalsobeemployedbymix ingfuelwithsteamand insucientoxygen,resultinginpartialoxidationormixin gwithrecirculatedexhaust 29

PAGE 30

gases(containingsteamfromcombustionproducts).Theben etsofthisincludereduction incombustionirreversibilityandrecoveryofexhaustheat [ 2 ].OtherconceptsforCO 2 reductionincludepartialoxidationcyclesandburningcar bon-freefuels. NO x (NOandNO 2 )isanothermajorpollutantfromhightemperaturecombusti on systems,knownforitsdeleteriouseects.Itformsacidrai nandcontributestoglobal warming(throughproductionofgroundlevelozone)[ 1 ].Theformationofground-level ozoneduetoNO x isalsoknowntoaggravaterespiratoryproblems[ 1 ].LowNO x systems andultra-lowNO x systemshavebeenproposedwithemissionsbelow10ppm,ands everal NO x reductionstrategieshavebeenemployed.WunningandWun ning[ 3 ]havediscussed reductionofNOthroughseveraltechniques.ThermalNOcanb ereducedthroughrame coolingtechniques,includinginjectionofNH 3 orH 2 O(wetNO x control)orcooling throughexhaustgasrecirculation(EGR)orcoolingrodsinb urners(dryNO x control)[ 3 ]. Multistagingandusageofhighvelocityinletstreamsareem ployedtocoolfreshcharge withexhaustproducts,fordryNO x controlWunningandWunning[ 3 ].Leanpremixed technology[ 4 ]usespremixedairandfuelcombustionwithexcessairforra metemperature suppression.Thetechnology,however,suersfromproblem sofpoorcombustionstability andrashback[ 3 ].TheGERich-Quench-Leantechnology[ 5 ]usesfuel-richprimary zonecombustion,followedbyafuel-leanlowtemperatureco mbustion,forthermalNO x reduction[ 6 ].Forreductionoffuel-boundNO x ,reburningstrategiesareusedforreduction ofNOtoN 2 .OxyfuelcombustionisyetanotherdryNO x reductiontechnique,andhas beenemployedinzeroemissionssemiclosedcycleconcepts[ 2 ].However,itsuersfrom drawbacksofO 2 expense,theneedforthesystemtobeair-tight,andthatnit rogen-bound fuelscannotbeused(suchasnaturalgaswithupto14%nitrog en)[ 3 ].SecondaryNO x removalstrategiesincludeselectivecatalyticreduction (SCR)andselectivenon-catalytic reduction(SCNR).Theseareparticularlyusefulforretro ttingolderhighemissions technologies,butmaybeexpensive[ 3 ].StagedcombustionforNO x reductionmaybe appliedbyairorfuelstaging(orreburning),airstagingbe ingamoreeectiveapproach 30

PAGE 31

[ 7 ].Xuetal.[ 7 ]alsosuggestedthatNOformationinfuel-richorreburning zonesis hinderedbytheabsenceofOandtheprofusionofCH i radicals. SCRisawetNO x removalstrategyandinvolvesammoniainjectionforreduct ionof NO x toN 2 .Thereactions[ 8 ]forNO x removalaregivenbelow: 4 NO +4 NH 3 + O 2 4 N 2 +6 H 2 O 2 NO 2 +4 NH 3 + O 2 3 N 2 +6 H 2 O Particulatematter(PM)consistofverysmallcondensedpha separticlesincluding aerosols,dust,etc.dispersedintheatmosphere,andtheyi mpactthelungs(aggravate asthma,bronchialdiseases)andtheheart[ 1 ].Particulatematterlessthan10micronsand neparticlesofsizelessthan2.5micronsareofparticular concernduetotheproblems associatedwithdecreasedvisibility.[ 1 ].Thesootparticulatesaretypicallyoftheorder of0.5to50nm,andarearesultofcombustionunderlocalfuel -richconditions.These arepredominantlycomposedofcarbonintheformofpolyarom atichydrocarbons(PAH), knowntobecarcinogenic.Smallsizedparticulatesareeasi lyingestedinthehuman pulmonarysystem,andaresignicantcontributorstobronc hialdisordersandlungcancer. Thetraditionalapproachforreductionofsootemissionswa sthatofprovidingadequate time,temperatureandturbulence[ 6 ]forcombustion.Usinghydrogenfuelcombustionhas beenproposedasoneofthemeansofcurtailingsootandUHCem issions. Carbonmonoxideisextremelydangerouswhenrespiratedine xcessivequantities,since itbindswithhaemoglobinandpreventsoxygensupplyintheb lood.Itisknowntohave detrimentaleectsontheheartandthenervoussystem[ 1 ].ThereductionofCO,UHC andsootemissionsincombustionsystems,istypicallyachi evedbyincreasingtheresidence timeinsidethecombustionchamber,andavoidingcold-spot sthroughecientdesign.The CO,UHCandsootemissionsaretypicallylowerforfuel-lean combustion,forarangeof equivalenceratios.Thekeyprobleminoptimizationofemis sionsisthat,simultaneous reductionofCO,UHC,sootemissionsandNO x ,imposessevereconstraintsonthesystem variables,andmaybeextremelychallenginginconventiona lcombustionsystems[ 6 ]. 31

PAGE 32

However,semiclosedcyclescanachievesubstantiallylowe rcarbon(CO 2 ,CO,UHCand sootemissions)aswellasNO x emissions,andareakeymotivationbehindthiswork. 1.2Motivation Thedevelopmentofpowergenerationsystemsoptimizedform aximumpowerandlower overallemissionshasbeenaprimethrustinmotivatingacti veresearchintheeldof combustion.Severallowemissionsconceptshavebeenpropo sedforenergysystemsand areactivelybeinginvestigated.Anovelsemiclosedcycle( SCC)engineconceptisbeing developedbyanacademic-industrialteamleadbytheUniver sityofFlorida.Thiscycle hasbeensuccessfullydemonstratedinalaboratoryscaleen ginetestrigcalledthepower, waterextractionandrefrigeration(PoWER)engine.Thisqu ad-generation(combined heat,power,refrigerationandwaterextraction)technolo gyevolvedfromitspredecessor engine,theregenerativefeedbackturbineengine(RFTE).O verandabovethecapabilityof theRFTEengine,thePoWERengineincorporatesabottomingr efrigerationcycle,thatis, avaporabsorptionrefrigerationsystem(VARS),thatalsoa llowsfreshwaterharvesting. AschematicofthePoWERsystemisshowninFig. 1-1 .ThePoWERcycleconstitutes threemajorcomponents-microturbineblock,turbocharger ,andtheVARSunit. Thelow-pressurecompressor(LPC)andlow-pressureturbin e(LPT)constitutethe Turbocharger.Themicroturbineblockconsistsofthehighpressurecompressor(HPC), combustor,high-pressureturbine(HPT)andtherecuperato r(REC).TheVARSunit extractsheatfromtheEGRstreamintheGeneratororthehotg ascooler(HGC),andin theevaporatororthecoolgascooler(CGC).Theintercooler ,betweentheHGCandthe warmgascooler(WGC),usestheplantcoolingwater(PCW)toc ooltheEGRstreamin anintermediatestep.Therecirculatedexhaustgasespasst hroughtheHGC,WGCand theCGC,losingheatintheprocess.Forsimplicity,onlythe gasturbinerowpathhas beenshown. InordertoconductcombustionexperimentsathighEGR,ares earchrigforthe PoWERenginewasbuilt,knownastheHighRecirculationComb ustion(HiRC)facility. 32

PAGE 33

Thissystemhasamainairinlet(MAI)(notpresentinaPoWERs ystem)thatallows runningtheengineathighEGRinunboostedmode(withoutthe turbocharger).Ambient airentersthecycleatState13,rowingintotheLPC.Thecomp ressedairatState 14,adiabaticallymixeswiththerecirculatedexhauststre am(State10)atthemixing junction,theoutletofwhichmixeswiththeair(State1)thr oughtheMAI.TheMAI isgenerallyclosedfortheHiRC;however,itisusedwhenope ratingtheenginein non-boostedmode.Themixedstream(State2)isfurthercomp ressedbytheHPCto State3,whichisheatedtoState3.1intheREC,usingwastehe atfromtheexhaustgases. TheheatedvitiatedairstreamatState3.1entersthecombus tor,wherethefuelisburned underoverall-leanstoichiometry,toahightemperatureex itState4.Thehightemperature gasesareexpandedtoState5intheHPT.Thehighenthalpyoft heturbineexhaustis utilizedtoheattheHPCexitstreamintherecuperator.Inth eprocessofheatexchange, theexhaustgasescooltoState6.Alargepartoftheexhausts treamisrecirculatedback intotheengine,whiletheremainingfractionmaybeexhaust edthroughtheexhaustvalve (ExV)atState18,inordertoreducetheboostpressure.Thee xhaustatState11,is utilizedtodrivetheLPToftheturbocharger,andtherestof therecirculatedgases(State 9,EGRstream)areusedtopowerabottomingVARScycle.TheHG Cisthegenerator oftheVARSunit,whereastrongrefrigerantsolutionisheat edbytheEGRstreamto generaterefrigerantvapor.ThecooledEGRstreamatState9 .1thenentersthe(WGC), whichiscooledbythePCW.Waterfromtheexhaustproductsof combustionisharvested forenduse,whiletherestoftherecirculatedexhaustisfed intotheCGCatState9.2 wheretheliquidrefrigerantfurthercoolstheEGRstreamto State9.3.Themassrowof theexhaustgasrecirculationiscontrolledbytherecircul ationvalve(ReV). Thisstudyhasbeenmotivatedbyourinterestinmodelingofp racticalgasturbine combustors,operatingonasemiclosedcycle.TheSCCengine soperatewithpartofthe exhaustgasesbeingfedbackthroughanintermediatepressu reEGRloop,todilutethe coreenginereactantairstream.Inearlierexperimentale orts,ithasbeensuccessfully 33

PAGE 34

demonstratedthatsemiclosedcyclesmayhaveuniqueadvant agesoflowerspecicfuel consumption,highecienciesatfullandpartloads,andlow eroverallemissionlevels.The lowercombustiontemperaturesandfairlyhomogeneousdist ributionthereactioninsidethe combustorresultedinmarkedreductioninNO x levels.Inaneorttooptimizethedesign ofthisengineconguration,severaltestswereconductedw ithvaryinglevelsofEGR.At veryhighlevelsofEGR,itwasfoundthattherameluminosity wassubstantiallyreduced. Thisobservationledtothehypothesisthatramelesscombus tionisassociatednotonly withdistributedreaction,butalsowithsignicantlyredu cedsootformation. Inhisreviewpaper,Mansurov[ 9 ]hasdiscussedcoolsootingrames.Therein,he hascitedworksthathaveprovedtheexistenceoflow-temper aturethresholdsofsoot formation.Thereducedornear-zeroluminosityintheramel essregimeissuggestiveof lowsootemissionsandhenceabsenceofradiativelossassoc iatedwiththeparticulate blackbodyemissions.Thusthereducedluminosityandreduc edsootappearsconsistent withincreasingEGR,asseeninourexperiments.Severaloth erresearchgroupshave reportedsimilarphenomena,alsofrequentlyreferredtoas Mildcombustion[ 3 10 ].This combustionregimehaslowerrametemperatures,uniformrow -eldandlowerpollutant levels[ 3 11 { 14 ].However,basedontheobservationsfromotherresearcher s,thereis evidencetobelievethatsootformationandUHCemissionsma yincreasewithEGR, especiallyathighequivalenceratios.Asanexample,Conga ndDagaut[ 15 ]investigated thevitiatedcombustionofnaturalgas/syngasmixturesthr oughexperimentationand modeling.EGRwasmodeledbyvaryingCO 2 compositionintheoxidizerstream.Ignition delaysandburnedgasvelocitiesweremodeled.Intheirsens itivityanalysisstudies,itwas foundthat,CO 2 hadaninhibitingeectonfueloxidation,whichcouldimply anincreased tendencytowardssootformationwithincreasingEGR.Theco mbustionstabilityseemed todeterioratewithincreasingtherowofthecooledEGRstre am.Ithasbeenreported thatathighlevelsofdilution,lowC/Oratiosandinlettemp eratures,thereisanonsetof instabilitiesduetooscillatorycombustionbehavior[ 14 ].Marutaetal.[ 13 ]showedthatthe 34

PAGE 35

extinctionlimitsbecamebroaderwithincreasingairtempe ratureforramelesscombustion. Forairtemperatureshigherthan1300K,theextinctionlimi tsdisappeared.Theyalso showedbroadeningofthereactionzone,forthecaseoframel esscombustion.Severalsuch contradictionsexistinthegrowingbodyofliteratureonra melesscombustion.Someof thecontradictionsareduetodieringapplicationsandexp erimentalsetups.Forexample, intheirworkonasub-ppmpremixedburner,KalbandSattelma yer[ 16 ]foundthatthe presenceofradicalshadastrongeectofimprovingcombust ionstability,whenthefresh streamoffuelandairwasperiodicallyadmixedwiththecomb ustionproducts.Thisis anexampleofexhaustgasrecirculation(EGR)athightemper atures(adiabaticrame temperature).Inourmodelingstudies[ 11 17 ]wehavedemonstratedthatdependingon thetemperatureoftheEGRstream,thestabilityofcombusti onmayactuallybeimproved orreduceddependingonitsneteectontherametemperature .Throughthisstudy,we proposetomakeacomparisoninthesootproduction,combust ionstabilityandemission levelsofasemiclosedcycleoperatingatvaryinglevelsofE GRwiththebasecaseofa conventionalopen-cycleengine. Schefer[ 18 ]studiedtheeectofhydrogenenrichmentonramestability andemissions onapremixedmethane/airrame,underleanconditions.Meas urementsthroughtheOH planarlaserinducedruorescence(PLIF)techniqueindicat edashreddedappearanceclose tothestabilitylimit,andthathydrogenenrichmentincrea sedOHconcentrationand increasedthelimitsofstability.ItwasalsofoundthatNO x andCOemissionsdecreased withhydrogenaddition,aneectattributedtotheincrease intheradicalpool.The increaseinstabilitylimitsduetohydrogenwasattributed tothewiderstabilitylimitsand higherramespeedofhydrogen.Thismotivatedustoinvestig atethecorrelationofloading parameter(LP)(measureofcombustionblowoutstability)w ithOHandCOemissionsin thecurrentstudy. Duwigetal.[ 19 ]performedcombustionstabilitystudiesonramelesscombu stionin agasturbinecombustormodel,usingdetailedchemistry,co mputationalruiddynamics 35

PAGE 36

(CFD)andexperimentalapproaches.Theyobservedthatthef ormationandsubsequent disintegrationofvorticalstructuresfacilitatedhotgas entrainmentandimproved microlevelmixing.Theysuggestedthattwonon-dimensiona lnumbersKarlowitznumber Ka r = r K 2 > 1( r isthereactinglayerthicknessand K istheKolmogorov lengthscale)andignitionnumber N I = res I > 1( res istheresidencetimeand I is theignitiondelay)describetheboundariesoframelesscom bustion.Damkohlernumber hastraditionallybeenappliedasanindicatorforramestab ility[ 20 ],instabilitymodels basedonglobalchemistry[ 21 { 23 ].However,inaccountingfordetailedchemicalkinetic eects,itisincorrecttobasethechemicaltimescaleonany particularspecies,sincethe intermediatespeciespoolservestoenhanceresidualreact ivity.Hence,ignitiondelayserves asagoodmeasureforacharacteristicchemicaltimescale.T hispromptedtheexploration ofthecorrelationofLPwithignitionnumberparametersugg estedbyDuwigetal.[ 19 ]. Recently,therehasbeenagrowinginterestintheuseofbiof uelsasanenergysource [ 24 ],andissuesinvolvingtechnologyreadinessandfuelrexib ilityforcombustionof biofuels.Thestudyofalternativefuelsiscloselyinterlo ckedwithissuesofemissions. Biofuelsareadiuseenergysourceandhencearesuitablefo rdistributedgeneration [ 12 ].Also,PoWERsystemsareideallysuitedfordistributedge nerationbecause oftheirfuelrexibility,higheciencyatfullandpartload ,highcompactness,and multi-generation.Duetosuitabilityofbiofuelsforusein distributedgenerationsystems, weareinvestigatingtheiremissionsperformanceintheram elesscombustionregime inherentinthePoWERcycle.However,thereisalsoconcerno verincreasedNO x emissions forbiodieselcombustion,whileparticulatesoot,COandUH Cemissionsaregenerally lower.Inaddition,combustionofbiofuelsisofinterestsi ncecombustionofbiodiesel andotheralternativefuelshaveanincreasedtheSolubleOr ganicFraction(SOF)level ofparticulatematter,increasingtheiroveralltoxicity[ 25 ].Sidhuetal.[ 25 ]foundthat DimethylEther(DME)hadthehighestsolubleorganicfracti on(SOF)contentof71 %,followedbybiodiesel66%,compressednaturalgas(CNG)3 8%anddiesel20%. 36

PAGE 37

Theyalsopointedoutthatbiodieselexhausthadoxygenated productslikebenzoicacid, pentenoicacidandvarioussubstitutedmethylestersinadd itiontoPAHs. Inonepartofthisstudy,theuseofbiofuelsethanol,Dimeth ylEther(DME),Methyl Butanoate(MB)inthevitiatedcombustionhasbeenexplored ,andtheircombustion performancehasbeencomparedtothatofn-heptane.Mostbio fuelsareknowntohave lowercaloricvalues,higherspecicfuelconsumption(SF C)andproducehigherNO x Hence,usingbiofuelsintheinherentlylowNO x ,PoWERcyclehastheadvantageof loweringoverallemissionsratherthanpotentiallyraisin gthem.Ithasbeenshownthat useofoxygenatedfuelshasapronouncedeectofreducingso otformation[ 26 ].However, inconsistentresultshavebeenreportedforthecombustion ofbiofuelsinthevitiated combustionregime.Theemissionsperformancedependsonth esourceandtypeofbiofuel, andinvolvesinteractionofcompetingeectsoffuelproper tiesandcombustorsystem designparameters.Throughthismodelingeort,claricat ionissoughtonhowEGRcan improveoradverselyaectemissionsperformance,dependi ngontheoperationregime speciedbyparameterssuchasequivalenceratio,residenc etime,EGRlevelandEGR temperature.Thisstudyinvestigatestheproblemusingasi mpliedmodelfromthe perspectiveofdetailedchemicalkinetics,inordertoreso lvetheinconsistenciesobservedin combustionperformanceofbiofuelsinthevitiatedcombust ionregime. 1.3ResearchObjectives Preliminarystudieshavebeencarriedouttoinvestigateth ekeyissuesinvolvedin ramelesscombustion,andtodeneclearobjectivesofthepr oject.Inordertostudy theperformanceofbiofuelsinvitiatedcombustionregime, thecombustorwasmodeled usingasimpliedreactornetwork,withapartialrecircula tionoftheexhaustrow. Detailedchemicalkineticsimulationswereperformedtoco mparethepollutantsgeneration andblowoutcharacteristicsduringcombustionundervaryi ngconditionsofexhaust gasrecirculationandequivalenceratios,andacomparison wasmadewithcombustors 37

PAGE 38

operatingonaconventionalcycle.Basedontheinsightgain edfromthepreliminary studies,thefollowingobjectivesoftheresearchstudywer eset: ResearchObjective1. Comparisonofbiofuelperformanceinvitiatedcombustion {Tostudyandcomparethecombustioneciency,pollutantem issionsgeneration (intermsofCOandUHC)andrameblowoutstabilityofvarious biofuels namely,ethanol,methylbutanoate(biodieselfuelsurroga te)anddimethyl ether,withrespectton-heptane(dieselfuelsurrogate). {Toinvestigatetheperformance(emissions,combustione ciencyandstability) ofCO/H 2 syngasmixtures(25%,50%and75%H 2 ),andpresentacomparison withconventionalmethanefuel. {TostudytheeectofEGRonbiofuelemissionsandstability ,ataxedEGR streamtemperature,incomparisontothezeroEGRopencycle (OC)case. ResearchObjective2. Modelingcombustionstability {Toinvestigaterameblowoutstabilityforvitiatedcombus tionatxedfuel composition(ethanolfuel),forvaryingEGR,equivalencer atios,residencetimes andEGRstreamtemperatures. {ToinvestigatethecorrelationofloadingparameterwithO Hconcentration. {Toinvestigatethecorrelationofloadingparameterwithi gnitionnumber(ratio ofignitiondelayandresidencetime). ResearchObjective3. Modelingsootformation {Topredictthecombustioneciencyandpollutantsgenerat ion,intermsofCO, UHC,sootprecursorswithvaryingresidencetimes,equival enceratio,EGRand EGRtemperature. {TomodeltheeectofEGRontheformationsootparticulates andpresenta comparisonwiththezeroEGROCcase. {TocompareresultsofsootformationmodeldevelopedbyFre nklachand coworkers,andthatdevelopedbyRichteretal.(Howardandc oworkers). 38

PAGE 39

Figure1-1.SchematicofaPower,WaterExtractionandRefri gerationCycleEngine 39

PAGE 40

CHAPTER2 LITERATUREREVIEW 2.1SemiclosedCycleEnginesandthePower,WaterExtractio nand Refrigeration(PoWER)Engine SemiclosedCycle(SCC)enginesareanovelconceptthathave regainedattentiondue totechnologicaladvancesoverthelastcoupleofdecades.T heSCCconceptwasrst proposedbackinthe1940s.Theearlyinvestigationswereca rriedoutbyAnxionnaz [ 27 28 ],Davis[ 29 ],DeWittandBoyum[ 30 ],Gasparovic[ 31 32 ].AnSCCengineutilizes exhaustgasrecirculation(EGR)atintermediatepressure[ 33 ].Thispresentsaunique combustionenvironment,whichhasbeeninvestigatedthrou ghmodeling,foremissions reduction[ 11 12 34 { 37 ]aswellasthermodynamiccycleoptimization[ 35 38 { 41 ].Several variationsofthisconcepthavebeensuggested[ 2 ],includingthoseinvestigatedatthe UniversityofFlorida.TheRegenerativeFeedbackTurbineE ngine(RFTE)conguration wasonevariantoftheSCCconcept,withaturbocharged-inte rcooled-recuperatedcycle developedattheUniversityofFlorida(UF).Subsequently, theHighPressureRegenerative TurbineEngine(HPRTE)congurationwasdeveloped,whichw asacombinedcycle concept.InitiallyabottomingRankinecyclewasproposed, whichwaseventually substitutedbyabottomingrefrigerationcycle.Thisledto theconceptionofthePoWER enginewithcapabilitiesofquad-generation(combinedhea t,power,refrigerationandwater extraction). TheHPRTEenginewasinspiredbytheWolverineenginebuilta sabackupsystem fornuclearpropulsionsystemsforsubmarinesinnavalappl ications.Eventually,Project Wolverinewasdiscontinuedduetothesuccessofthenuclear program[ 39 ].Laganelli etal.[ 42 ],intheirwhitepaper,outlinedthebenetsoftheHPRTEsys temfornaval applications.Theysuggestedthatthecapabilityoffreshw aterextractionfromcombustion gases,reducedthermalsignature,lowerstacksizes,highe rpowerdensityandhigh part-loadeciency,madeHPRTEsystemsparticularlysuita blefornavalapplications. Inanotherstudy,Meitneretal.[ 43 ]showedthattheHPRTEengineinopencyclemode, 40

PAGE 41

producedCOconcentrationof124ppmandNO x concentrationintherangeof53to65 ppm.Insemiclosedcyclemode,theydemonstratedthat,theC Oemissionlevelswere drasticallyreducedtovaluesaslowas5ppm.Theyattribute dthereductionsinemissions tolowerexhaustrowrates,dilutionofcombustorprimaryzo newithproductgases, oxygendepletionandlowerrametemperatures. TheHPRTEwasderivedfromtheRFTEconceptbyincludingabot tomingrefrigeration cycletoaturbocharged-intercooled-recuperatedsemiclo sedcycleoftheRFTE.Inaquest tounravelthecomplexitiesofvitiatedcombustion,severa lcombustionmodelingstudies wereundertaken[ 11 23 34 36 37 44 ].MuleyandLear[ 34 ]investigatedthethermalNO x emissionsperformanceoftheSCCenginesthroughmodeling. Thecombustorprimary zonewasanalyzedusingaperfectlystirredreactor(PSR)mo del,andtheNO x chemistry wasmodeledusingtheZeldovichmechanism.ThesignicantN O x emissionsreductions achievedinSCCsystemswereattributedtoloweringofthera metemperatureandoxygen depletion,asaresultofdilutionwithanexternallycooled EGRstream.Crittendenet al.[ 23 37 ]modiedthePSRtheorytopredictthecombustionstability andemissions performanceofanengineoperatingonSCC.TheapproachofSt rehlow[ 21 ]wasextended, inhiswork,topredictcombustionperformanceforSCCengin es.Harwoodetal.[ 36 44 ] studiedthesemiclosedcyclecombustorperformanceusingr ecirculatingplugrowreactor (RFPR)model,toaccountforinternalrecirculation.Theyd emonstratedthatrame stabilitywasastrongfunctionofinternalrecirculationr owandtheresidencetimeinthe combustor.Further,itwasshownthat,theamountofinterna lrecirculationrequiredto maintainstabilityincreasedwithdecreasingtemperature oftheinletstream. ThermodynamicanalysisofSCCenginesisachallengingprob lem,andseveralstudies havebeenconductedtounderstandtheimpactofseveraldesi gnparameters[ 35 38 40 41 43 ].MacFarlaneandLear[ 38 ]investigatedthebenetsofreinjectionofextracted waterfromaRFTEsystem.Theyconcludedthatreasonableper formanceimprovements intermsofincreasedeciency,decreasedspecicfuelcons umption(SFC),andincreased 41

PAGE 42

specicpowercouldbeachievedfromRFTEenginebymeansofw aterinjection.Itwas foundthattheoptimumcongurationforthemostpracticals ystemwasachievedforthe caseofwaterextractionmatchingwaterinjection,sothatt hereisnoexternalstorage requirementandasimplercontrolschemecanbeused.Meitne retal.[ 39 ]demonstrated theHPRTESCCconceptonaturboshaftgasturbineengine.The ysuggestedthatan HPRTEenginehadbenetsofreducedairrow,lowerburnertem perature,compactness, ratSFCcurvesandlowemissions.Theiranalysispredicteda constantspecicfuel consumptionasafunctionofpowerandtheirexperimentalob servationsshoweda signicantreductioninemissions. Bozaetal.[ 45 ]performedathermodynamicanalysisofacombinedcycleSCC engine withvaporabsorptionrefrigerationasthebottomingcycle ,therststudyofthePoWER cycle.Theyinferredthatthethermaleciencyandrefriger ationratiowerestrong functionsofturbineinlettemperatureandtheambienttemp erature,respectively.Two cases,ofasmallandalargeenginewereanalyzed,bothofwhi chshowed2to5points increaseincombinedthermaleciency(inclusiveoftheref rigerationpower). NemecandLear[ 35 ]demonstratedviaadesignpointstudythatanHPRTEcyclewi th abottomingrankinecycleshowedsignicantimprovementsi neciency.Avariantof theHPRTEwithavaporabsorptionrefrigerationsystem(VAR S)bottomingcyclewas eventuallydeveloped[ 46 ].Duringthecourseofitsdevelopment,severaltechnologi cal innovationstookplace.ItwasfoundthatanSCCenginesucha stheHPRTEwascapable ofproducingfreshwater.Khanetal.[ 40 46 { 51 ]investigatedtheperformanceofthe HPRTEenginethroughmodelingandexperimentationonanHPR TEtestfacility.Itwas alsofoundthattheseenginesarecapableofrunninginramel esscombustionmode,athigh levelsofEGR.ThenewestvariantoftheSCCengine,beingdev elopedatUF,isknown asthePoWERengine.Inrecentmodelingstudies[ 41 52 ],thePoWERengineswith multi-evaporatorcongurationsoftheVARShavebeeninves tigatedforiceproduction, thatmaybeusedforload-leveling. 42

PAGE 43

Khanetal.[ 40 ]studiedthewaterextractionperformanceofPoWERcycleen gine.They concludedthatincreasingturbineinlettemperatureincre asedeciency,whiledecreased refrigerationratioandwaterextraction,ontheotherhand .Recuperatorinlettemperature hadanoptimaleectoneciencyandrefrigeration,whileit enhancedwaterextraction. Increasinglow-pressurecompressor(LPC)pressureratioa dverselyaectedeciency,while improvedrefrigerationratioandwaterextraction. CamporealeandFortunato[ 53 ]studiedasemiclosedgasturbinecycleoperating withhightemperaturedilutecombustionthroughEGR.Abott omingsteamcyclewas consideredforcombinedcyclegasturbinesystems.Theysug gestedthatsuchacycle hadtheadvantagesofreductionofcombustionexergylosses duetohighoxygenstream preheat,compactheatexchangersduetohighersystempress ures,highpartloadeciency andappropriateconditionsforCO 2 separation. FiaschiandManfrida[ 54 ]investigatedthesemiclosedgasturbine(SCGT)/regenera tive combinedcycle(RCC)concept,asapotentialzeroCO 2 emissionssystem.Thiscycle includesintercooling,aftercooling,recuperation,wate rextractionandre-injectiongas cycle,withabottomingrankinecycle.Theyconcludedthatd rasticreductionsinthecost andsizeofHRSGofthebottomrankinecyclecouldbeachieved Jordaletal.[ 55 ]modeledtheperformanceofoxyfuelcombustionwithCO 2 dilution duringhightemperaturecombustioninasemiclosedcycle,i ncludingtheeectofcooling ofturbineblades.Theyfoundadeteriorationofturbineper formanceandincrease inturbinespeedforxedgeometry.Theimprovementofecie ncyachievedthrough modicationoftheturbinebladeangles,wasindicativeoft heneedforredesigningthe turbineforoxyfuelcombustion. BollandandMathieu[ 56 ]comparedtwoCO 2 reductionstrategies,onewithoxyfuel combustion,withrecirculationofCO 2 ,andtheotherasemiclosedbraytoncyclewith CO 2 sequestration.Theyhighlightedthatthebenetoftheoxyf uelsemiclosedcyclewas that,itdoesnotrequireadditionalCO 2 separationequipment,however,anairseparation 43

PAGE 44

unit(ASU)isrequired.ItwasfoundthatthecyclewithCO 2 sequestrationhadhigher eciencyincomparisontotheoxyfuelcombustioncycle,for allrecirculationrates. 2.2FlamelessCombustion Flamelesscombustionisarecentlyobservedphenomenon,th atholdshugepotentialin improvingcombustionperformanceandemissionsreduction [ 10 ].Thereportedbenets oframelesscombustionincludeuniformtemperatureeldan dspeciesconcentration proles,signicantNO x reductions,lowersootyieldsandnoiselevels[ 3 10 { 13 57 ].A signicantlyhigherlevelofhomogeneityoftemperaturean dconcentrationprolesis achieved,suggestiveofaregimeshiftfromadiusioncontr olledregimetothatofkinetic control[ 58 ].TheabsenceoftemperaturepeaksresultsinlowerNO x ,whiletheabsenceof coolpocketsreduceszonesoffrozenreactionchemistryand highCOemissionsassociated withit.Severalnameshavebeenproposedforthephenomenon oframelesscombustion byresearchersacrosstheglobe.InJapan,thephenomenonha sbeenreferredtoasexcess enthalpycombustion,thatisnowreferredtoasHiTAC/HTAC[ 59 60 ].TheHiTAC technologyutilizesoxygendepletioninconjunctionwithh eatregenerationtoachieve ramelesscombustion.Thetermmoderateandintensivelowox ygendiluted(MILD) combustionisusedinItaly[ 14 53 ]andramelessoxidation(FLOX R r )inGermany[ 3 ]. Othernamessuchasdistributedcolorlesscombustion,volu metriccombustion,green ramecombustion,anddilutedcombustion[ 61 ]refertothesamephenomenon.Mild combustioncanbeachievedasaresultofintenseinternalmi xingandoxygendilution asincaseofFLOX R r technology[ 62 ]orthroughexternalexhaustgasrecirculation [ 11 12 17 46 53 63 { 65 ].InternaladiabaticEGRachievedthroughhotgasentrainm ent hasalsobeenutilizedintheIntegralLowNO x Injection(LNI)burnertechnology(United StatesPatent6206686,Seehttp://www.freepatentsonline .com/6206686.html)andLTCin dieselengines[ 66 ].Severalresearchers[ 7 61 67 { 71 ]havestudiedramelesscombustionin multi-stagedburnersandfoundimprovedcombustionperfor mance.TheNO x reductionsin 44

PAGE 45

ramelesscombustionhavebeenalsoinvestigatedthroughmo deling,byseveralresearchers [ 44 72 73 ]. Therearemanypredictiveindicatorsthatmaybeusedtoden etransitionfrom conventionalcombustionregimetoramelesscombustion.Ca valiereanddeJoannon [ 74 ]suggestedthatramelesscombustionconditionisachieved whentheinletreactant temperatureislargerthantheauto-ignitiontemperaturea ndtheignitiondelayislonger thanthemixingtime,sothatadistributedvolumetriccombu stionisachieved.Derudi etal.[ 75 ]usedtheconditionofdrasticreductioninNOastheconditi onforattainment oframelesscombustion.Kumaretal.[ 76 ]investigatedthehomogenizationofgradients throughtemperature-volumeandO 2 -volumecumulativedistributionfunctionsasthe criterionforattainmentofmildcombustion.AcevesandFlo wers[ 77 ]usedexaminationof sootprecursorstoidentifyregimesforramelesscombustio n.Otherresearchershavenoted amarkedreductioninOHconcentration(stronglycorrelate swithheatreleaserate[ 77 ])in ramelesscombustion[ 10 ]throughchemiluminescenceandplanarlaserinducedruore scence (PLIF)studies,whichmaybeusedasacriterionforthetrans itiontoramelessregime. Indieselinternalcombustion(IC)enginestheramelesscom bustionconceptisutilizedfor achievingnon-sootingcombustionthroughtworegimesofsm okelessrichcombustion (SRC)andmodulatedkinetics(MK)[ 77 ].Thesootformationprocessisnormally predominantinthetemperaturerangebetween1600and2600K andatequivalence ratiosgreaterthan2[ 77 ].Glassmanetal.[ 78 ]investigatedcriticalsootingtemperatures andfoundthatsootformationinceptiontypicallybeginsat around1600K,irrespective ofthefuel.Henceatemperatureboundarymaybeidentiable fortransitioningto ramelessregime.Thetransitiontoramelesscombustionisu suallyabrupt[ 11 12 17 ],and accompaniedbyasuddendropinrameluminosityandhomogeni zationofthegradientsin temperatureandconcentrationprolesinsidetheroweld. Theboundariesoftransition toramelesscombustionmaybecorrelatedwithonsetoftheso otformationprocess [ 64 ].Derudietal.[ 75 ]usedNO x emissionsbelow30ppmandCObelow50ppm,asthe 45

PAGE 46

criterionfortransitiontoramelesscombustion.TheNO x emissionsreductionsinrameless combustionaremainlyduetoreductioninthermalNO x formation.Sincethefuel-NO x routeislesssensitivetotemperature,EGRhaslittleimpac tonitsreduction[ 70 ]. Combustionprocesseshavetraditionallybeenclassiedin topreramecombustion, deragrationanddetonation[ 6 ].Preramecombustionistypicallyslowanddistributed volumetricallyandtakesupto100sfor80%completion[ 6 ].Deragrationisafaster combustionprocessoccurringataramefrontandtakesupto1 msfor80%completion [ 6 ].Thederagrationwavestypicallytravelatunder1m/s,whi ledetonationwaves travelatsupersonicvelocities[ 6 ].Gasturbinecombustiontypicallyproceedsinthe deragrativeregime.Onepossiblecategorizationmeasureo framelesscombustioncould betheswitchfromderagrativetovolumetriccombustion.Th eKlimov-Williamcriterion ( l k < L U 0 S L where l k istheKolmogorovmicroscaleand L isthelaminarrame thickness[ 79 ,p.229])mustnotbesatisedfordistributedreactiontooc cur[ 75 ].This typicallyoccursatlowDamkohlernumbers[ 80 ],withhighturbulenceintensitiesand EGR.Theexcessenthalpyratio,denedastheratio T ad T in T in T ref [ 81 ],isfoundtobelower forramelesscombustion,andmaybeusedasanindicatorforp redictingtheboundary oframelessregime.Forstability,theinlettemperaturesh ouldpreferablybehigherthan autoignitiontemperature[ 81 ]. Flamme[ 68 ]hasinvestigatedtheuseofFLOX R r andcontinuousstagedaircombustion (COSTAIR)burnertechnologiesasaNO x reductionstrategyforglassfurnaces.While astagedburnerhadaprimaryairstream,aramelessoxidatio nburnerdidnothave it.Theyconcludedthat,withrue-gasrecirculation,lowNO x emissionswerepossible evenwithhigherairpreheating.TheNO x emissionswereunder200mg/m 3 forinletair preheattemperaturesupto1000Candfurnacetemperatureof 1200Cwhilethestandard burnerproducedNO x emissionsupto1500mg/m 3 forsimilarconditions.Theuniform temperatureeldoftheseapproachesallowedincreaseoffu rnacechambertemperature. Theyalsoinvestigatedtheeectoffuelboundnitrogeninco mbustionofnaturalgaswith 46

PAGE 47

O 2 andrue-gasrecirculation.Flamme[ 82 ]exploredtheuseoframelessoxidation(FLOX) andCOSTAIRburnertechnologiesforgasturbinecombustion andcomparedthemagainst thelean-premixedcombustors. Shaddixetal.[ 83 ]studiedpremixedcombustionofmethaneandvariousmixtur esof syngas,andoxygenstreamdilutedwithCO 2 inaswirlstabilizedcombustor.Theoxy-fuel conceptisafrequentlyemployedCO 2 removalstrategy,usedinintegratedgasication combinedcycle(IGCC)plants.SignicantreductioninNO x wasfoundalthoughthe COemissionswerestillhigh.Theyreportedthattheincreas ingH 2 /COratioofsyngas improvedramestabilitysubstantially.Theyhighlightedt hatleanpremixedcombustion hadthedrawbacksofreducedramestabilityandriseinCOemi ssionsatlowequivalence ratios,whileNO x emissionsreducedasaresultofreductioninrametemperatu res.The COemissionswerehighfor > 0 : 95and < 0 : 40[ 83 ].Thecombustioneciencies droppedsignicantlyforequivalenceratiosbelow0.4,for allcases. Weberetal.[ 70 ]studiedtheuseoframelesscombustiontechnologyforcomb ustion ofgaseous,liquidaswellassolidfuelsinfurnaces.Apreco mbustoraddedbeforethe furnace,wasusedtosupplyhighairpreheat,andahotgasrec irculationstreamwas developedthroughareverserowcausedbythehighmomentumo finletfuelstream[ 70 ]. Theyfoundthatupto60-70%reductionsinNO x werepossiblebyapplicationoflowNO x burnertechnologywithfuelorairstaging.Theyattributed lowerNO x emissionsincoal combustionascomparedtonaturalgas,lightfueloilandhea vyfueloil,toNO x reburning andlongerresidencetimes.Theyfoundthatforliquidring ,highparticulateemissions weresubstantiallyreducedbyreoptimizationoftheatomiz ersforramelesscombustion. Inaddition,theyobserveduniformfurnaceheatrux,reduce dgradientsandmarked reductioninNO x levels. Aidaetal.[ 69 ]investigatedthemultistagingofleanburncombustorstob urn prevaporizedkerosene-airmixtures.Theyfoundthatveryh ighcombustioneciencies 47

PAGE 48

andultra-lowNO x couldbeachievedbymultistaging.Theyalsoobservedexist enceof NO x reburningeectsevenundermultistagedlean-leanconditi ons. Levyetal.[ 65 ]investigatedthethermodynamicsoftheFLOXCOMhightempe rature ramelesscombustionprocesswithadmixing.Itwasdemonstr atedthatforaxed combustionexittemperature,theoxygenconcentrationsre mainedclosetostoichiometric, whilea100%primaryzonecombustioneciencywasassumed. Kajitaetal.[ 71 ]showedtheuseoframelesscombustionthroughburnermulti staging fordryNO x reduction.Leanpremixedcombustion,frequentlyemployed fordryNO x reduction,hadthedrawbackofreducedramestabilityandlo wcombustioneciencyat veryleanconditions.Asaresult,theair-fuelratiohadtob emoretightlycontrolled.In themultistagedconcept,inadditiontotheeightpremixing combustors,asecondstage pilotdiusionburnerwasused,thatenhancedstabilityatv eryleanconditionsatlow loads.TheyfoundtheexistenceofNO x reductionprocessduringmultistaging. GuptaandHasegawa[ 84 ]examinedtheeectofincreasedairpreheatanddilution ontheoverallcombustionperformanceandramestructureof propaneairdiusion rames.Theirresultsindicatedthatathighlevelsofdiluti onandpreheat,theramehad reducedgradientsintemperatureandconcentrations,andl owemissionsandluminosity. Preheatingincreasedramevolume(duetolowerdiusionrat esasaresultoflower density),whileincreasingoxygenconcentration(decreas ingdilution)hadtheopposite eect.Theignitiondelaywasfoundtodecreasewithincreas ingairpreheatandincrease withtheamountofdilution.Theemissionsspectrashowedin creasingOH,CH,C 2 andH 2 Oemissionswithincreasedairpreheat.Increasingairpreh eatfacilitatedrame stabilizationdespitethereducedoxygenlevels[ 10 ].Athighpreheatandoxygendepletion, propanefuelproducedgreencoloredrameduetohighlevelso fC 2 species[ 10 ].Afurther reductioninoxygenconcentration,ledtodiscolorationof therame.Highradiativeand convectiveheatruxeswereobservedduetohighrowvelociti es,aswellasincreased 48

PAGE 49

wallheatrux[ 10 ].Theramestand-odistance,decreasedwithincreasingai rpreheat, indicatingreducedignitiondelayandimprovedramestabil ity[ 10 ]. Kumaretal.[ 76 ]investigatedscalingapproachesforamildcombustionbur ner.Itwas foundthatsecondaryairinjectionpositioningandinjecti onvelocityhadsignicantimpact onexhaustrecirculation.Theyevaluatedtraditionalscal ingmethodsofconstantvelocity, constantresidencetime,etc.,andfoundthosetobeinadequ ateformildcombustors.They foundthatconstantvelocityapproachincreasedmixingtim eandreducedmixingrate whiletheotherapproachesledtoincreasedsystempressure drop.Theyintroducedan additionalscalingcriterionbasedontheconvectivetimes cale(D/U)beinglessthan80 s Elluletal.[ 85 ]studiedtheeectoffuelcompositionsinhighlypreheated aircombustion incounterrowdiusionrames.TheyfoundthatNOformationw assuppressed,andthat NOreburningandN 2 Oroutesbecomepredominantforramelesscombustion. Mancinietal.[ 86 ]investigatednaturalgasramelesscombustioninafurnace ,withair preheatat1300C.TheyconcludedthatNO x formationwaspredominantlythroughthe thermalrouteandthatNO-reburningwasinsignicant.This isincontradictionwiththe workofSchutzetal.[ 62 ],NicolleandDagaut[ 87 ]. Adachietal.[ 67 ]investigateddilutedmulti-stagedcombustioninthree-s taged combustor.Theprimaryaxialrowstagewasusedtostabilize thecombustionof secondaryandtertiarycrossrowstages.NOfromtheprimary stagewasfoundto becompletelyoxidizedtoNO 2 inthesubsequentstages.Theyalsoreportedsubstantial reductionincombustionoscillationswithmultistaging.T heimpactofbiomasscomposition variancewasaddressedbydilutingCH 4 mixtureswithvaryingamountsofCO 2 .LowNO x andhighcombustioneciencywasobservedinthetemperatur erangebetween1500to 1700K. NicolleandDagaut[ 87 ]modeledchemicalkineticsofCH 4 ramestostudyNO-reburning phenomenoninramelesscombustion.Theydiscussedthesign icanceoftheN 2 Oroute, 49

PAGE 50

andthereductionofNObyhydrocarbonradicalsunderramele sscombustionconditions. TheyfoundpartialconversionofNOintoHCNandsubsequentl yNH 3 occurs,during autoignitionofthemixture.ThepresenceoftheexhaustH 2 Owasfoundtoinhibit prompt-NOandNNHroutes.Inthepost-ignitionphase,theth ermalandN 2 Orouteswere favored,whileNOreburningmaybeprevalentunderfuel-ric hconditions[ 87 ]. Wangetal.[ 8 ]performedtechno-economicstudiestocompareselectivec atalytic reduction(SCR),COSTAIRandFLOX R r technologiesforNO x reduction.TheCOSTAIR burnerhastwoairstreams,primaryandsecondary,whilethe FLOX R r technologyhasno primaryairsupply[ 8 ].However,bothtechnologiesrelyonentrainmentofexhaus tgases intothereactantstreamfortemperaturereduction.Theyco ncludedthattheuseofthe non-standardCOSTAIRandFLOX R r burnershadhighecienciesofNO x removal,with lowersystemcapitalcostandelectricitysellingprice. Blasiaketal.[ 81 ]investigatedramelessairandoxyfuelfurnacecombustion .They foundthatoxyfuelcombustionhadverylowNO x duetotheabsenceofN 2 .Lowerrame temperatures,largervolumeanduniformtemperaturedistr ibutionswereobservedwith oxyfuelcombustion.Asimpleradiationheattransfermodel ,predictedtheramevolumeto beaninversefourthpowerfunctionoftemperature.Flamevo lumewaspredictedusinggas samplingandcalculatingalocaloxidationmixtureratio,a ndfoundtobelargerforthe caseofoxyfuelcombustion. Therameblowoutstabilityintheregimeoframelesscombust ionhascompetingeects duetotheeectofdilutionoftheinletstreamanditstemper ature.Whiletheincreasein inlettemperaturepromotesramestability,theinletdilut ionhasjusttheoppositeeect. Theramestabilityforvitiatedcombustionmayalsobeimpro vedwithhydrogenaddition. TheworkofDerudietal.[ 75 ]discussestheuseofnon-conventionalfuelssuchascoke-o ven gas,withhydrogenpromotingramestability. Hamdietal.[ 88 ]investigatedthepollutantemissionsingasturbinesyste msathigh pressuresandtemperatures,throughdetailedkineticmode linginareactornetworkwith 50

PAGE 51

twoPSRs.Theyfoundthatatlowpressuresallthreeroutes{Z eldovich,Fennimoreand N 2 Omechanismswereofequalsignicance,whileathigherpres suresN 2 Oroutewas dominant.TheoverallNO x emissionsincreaseslightlywithincreasingpressure[ 88 ].They alsofoundaslightdropinCOemissionswithincreasedpress ures,andhighlightedthe reductionsinCO 2 emissionsasaresultofreducedexitmassrowduetoEGR. KatsukiandHasegawa[ 60 ]denedhighlypreheatedaircombustion(HPAC)as combustionwithinletreactanttemperatureaboveautoigni tiontemperature,sothat combustionproceedsspontaneously,withinthecombustion chamberevenintheabsenceof anignitionsource.Flamestabilizationthroughuseofablu bodyorrecirculationarenot necessaryforsuchacase[ 60 ].Theyreportedthepossibilityofconsiderablereduction of furnacesizeandfuelconsumptionwiththeuseofthistechno logy. Xuetal.[ 7 ]investigatedstagedcombustionforNO x reduction.Theyconcludedthat stagedcombustioninprimary,reburningandburnoutzonesi saneectivestrategyfor NO x reduction.TheyalsopointedoutthatH 2 andC 2 H 2 fuelshadgoodNOreburning eciency.CH 4 additionhadapositiveeectwhileCOadditiondidnotimpac tNO reduction.TheyfoundthattheintermediatesC,CH,CH 2 andHCCOsubstantially contributedtothereductionofNO. Characterizingtheramestructureinramelesscombustionh asbeenofinterestto researchers,andbothexperimental[ 89 ]andnumericalapproaches[ 19 61 62 90 { 94 ] havebeenapplied.Zero-dimensionalanalysisonramelessc ombustionhasbeenapplied fordetailedchemicalkineticstudies,topredictramestab ility[ 14 58 ],andemissions [ 77 87 88 ]. Fuchihataetal.[ 89 ]investigatedramestructuresforramelesscombustionusi ng lasertomography,chemiluminescence,laserdoplerveloci metry(LDV)andtemperature measurements.Theyobservedtheboundaryoftransitionbet weendistributedcombustion regimeandthatwherewrinkledramesbegintoappear.Theyno ticedthatdistributed reactionzoneoccurswhenareactionisinitiatedwithinalo wDamkohlernumberzone, 51

PAGE 52

whilepropagationofalaminarrameintoalowDamkohlernumb erzoneledtothe formationofathinreactionzone. Piepersetal.[ 93 ]performedexperimentsandcomputationalruiddynamics(C FD) modelingofnaturalgasramelesscombustioninafurnace.Th eyfoundthattheminimum safeoperationtemperatureforthefurnacewas700Cunderno rmalramemode,while 800Cforramelessmode.Thestabilitywasfoundtodeteriora te,onreconguringthe positioningofprimaryandsecondarynozzles.Theyalsohig hlightedthatasaresultof slowkineticsoframelesscombustion,the`mixedisburned' assumption,frequentlyusedin CFDmodeling,doesnotholdtrueunderlowloadconditions. Gallettietal.[ 90 ]performedCFDanalysisofaburneroperatingunderMILD combustionregime.Theresultsoftwonumericalmodels,a3D andanaxisymmetric model,werecomparedwiththeexperimentaldata.Theirresu ltssuggestedthatthe axisymmetricmodel,thatdidnotincludedetailslikethere circulationwindows,was toosimplisticandinaccurate.TheyfoundasignicantNO x reduction.Theturbulent damkohlernumberforhigherinternalrecirculation,achie vedthroughdecreasing airrowvelocity,waslowerthanthatforcombustioninramem ode[ 90 ].Hence, turbulence-chemistryinteractionwassignicantinMILDc ombustion[ 90 ]. Tabaccoetal.[ 94 ]investigatedtheramelesscombustionprocess,throughCF Dand zero-dimensionaldetailedkineticsimulations.Theyfoun dthatwiththeincreasein recirculationstreamtemperature,theignitiondelaysred uceandturbulence-kinetics interactionsbecomestronger.Thestrongturbulence-kine tics(Da=1,comparabletime scales)interactionsresultinlowerNO x andincreasedhomogeneity.Ontheotherhandat lowerrecirculationtemperatures,thereactionwaspredom inantlyfoundtobekinetically controlled[ 94 ].IncreasedCO 2 andH 2 Oinproductgases,increaseradiationabsorption assistingreactantautoignitionandpromotingthermalhom ogeneity(thermaleect)[ 94 ]. Themasseectofrecirculation(dilution)isthatofslower kineticsandweakeningofheat release[ 94 ].Thekineticeectassociatedwithrecirculationwasthat ofincreasedHand 52

PAGE 53

OHradicalswhiledecreasedOradicals,duetothepresenceo fhigherH 2 Ocontent,and alteredNO x formationchemistry[ 94 ].TheoverallNO x wasfoundtobesuppresseddue toreducedOradicalsandreducedrametemperatures.Thepre dominantNO x formation routewasfoundtobethatofnitrousoxide(N 2 O),promptNO x beingnegligible,whilethe thermalNOandNO 2 routescontributedafairlylargeproportionofthetotalNO x [ 94 ]. Schutzetal.[ 62 ]usedpartiallystirredreactor(PaSR)modeltostudytheNO x generationinaFLOX R r combustor.Theyfoundthatfortheramelessoxidationproce ss, NOproductionthroughtheroutesofZeldovich,Fennimorean dN 2 Omechanismswereof thesameorderofmagnitude.Theyalsohighlightedtheinade quacyofthePaSRand k modelstocapturetheturbulence/chemistryinteractionsw ell. Awosopeetal.[ 61 ]usedCFDmodelingoframelessoxidationtostudyitsapplic ation togasturbinecombustors.Theycomparedtheirnumericalmo delingresultstothoseofan experimentalfacilitycomprisingofapremixer,precombus torandanafter-burner.Their resultsdemonstratedreductionsinNO x andimprovedpatternfactorsingasturbines. CoelhoandPeters[ 91 ]usedEulerianparticlerameletmodel(EPFM)(anunsteady rameletapproach)topredictNOformation.Theyfoundsatis factoryagreementbetween theirmodelandexperiments,whilethesteadyrameletappro ach,ontheotherhand,was foundtooverpredicttheirmeasurements. YuanandNaruse[ 92 ]performedCFDanalysisandexperimentsonfurnacemild combustion.TheycomparedtheperformanceofdiluentsN 2 ,CO 2 ,andHeinrameless combustionandfoundCO 2 producedlowerrametemperatures,comparedtoHeandN 2 Theyruledoutthepossibilityofthedierencebeingpurely aresultofthermaleects, sinceN 2 andCO 2 hadcomparableheatcapacities.Theyquantiedhomogeniza tionof temperaturesasafunctionoftheratio T max T mean ,andfoundimprovedhomogenizationwith decreasingoxygenconcentrationinpresenceofhighlypreh eatedair.TheyusedTesner's twostepmodelforsootformation,andfoundmaximumsootcon centrationstobehigher withhigherpreheattemperatures.Athighpreheat,thesoot maximumconcentration 53

PAGE 54

washigherwithlowerO 2 %(increaseddilution)[ 92 ].TheyfoundOHconcentrations todecreasewithincreaseddilutionandairpreheat.Theyal sodeterminedtheoptimum temperaturesandoxygenconcentrationsforNO x reduction. MarekandTacina[ 95 ]investigatedeectofadiabaticEGRinarametubecombusto r andfoundimprovementsinNO x andeciency.Theyalsofoundhigherrangeof operabilityinequivalenceratio(athighereciency)with increasingEGR.Additionally, theyfoundthatNO x mayalsoincreasewithincreasingEGR,especiallyathigher equivalenceratios,andattributedthesetoheterogeneiti esofspraycombustion.The NO x dependenceonresidencetimewasreducedwithincreasingEG R[ 95 ].Moreover,they foundthatair-quenchratesandlowerrametemperatureshad astrongeectonNO x reduction.2.2.1EectofDiluentGas CO 2 hasbeenfoundtohavestrongretardingeectkinetically[ 15 15 92 96 ]onfuel oxidation.H 2 OhasapositivekineticeectduetotheincreaseinHandOHra dicals[ 39 ]. Thediluentgases,N 2 ,CO 2 andH 2 O,haveastrongthermalquenchingeect,whilethe eectofN 2 hasbeenfoundtobepurelythermal[ 15 ]. CongandDagaut[ 15 ]investigatedthevitiatedcombustionofnaturalgas/syng as mixturesthroughexperimentationandmodeling.EGRwasmod eledbyvaryingCO 2 compositioninoxidizerstream[ 15 ].Itwasfoundthroughsensitivityanalysis,thatCO 2 hadaninhibitingeectonfueloxidation. Levyetal.[ 97 ]studiedignitiondelaysfordilutedmethanemixturesinsh ocktube studiesandmodelingeorts.Theyreportedtheinsensitivi tyofignitiondelaymeasurements tothediluentcompositionbycomparingthecasesofCH 4 -N 2 andCH 4 -N 2 /CO 2 /H 2 O mixtures.They'vealsohighlightedthattheinhibitionofc ombustionbyCO 2 reported byCongandDagaut[ 15 ],Lisianskietal.[ 96 ],couldnotbeconrmedthroughtheir simulationorexperiments. 54

PAGE 55

Macadam[ 98 ]studiedtheeectofdiluentsinvitiationcombustioninje t-stirred reactor(JSR).BothN 2 andCO 2 vitiationwasfoundtoreducesooting,andCO 2 seemed toenhanceoxidationofthesootparticles(throughincreas eofOHradicalpoolbythe reaction CO 2 + H CO + OH )[ 98 ].Theeectoframetemperature,anddilutionat constantrametemperaturewerealsoinvestigated.Similar studieswereconductedby Huietal.[ 99 ],tostudytheinruenceofdilutiononramestructureandemi ssionsfroma counterrowdiusionrameofsyngas.TheyfoundincreaseinC OanddecreaseinNOwith dilutionatconstantrametemperature.TheyfoundthatN 2 andCO 2 enhanced,while H 2 OsuppressedNOemissions. WilsonandLyons[ 100 ]investigatedtherameliftoandstabilityofmethaneand ethyleneramesincorowconguration,withN 2 dilution.Theyfoundreducedjetvelocity andcorowvelocityatagivenrameheight,wereneededwhendi lutionandcorowvelocity wereincreased.Theyalsofoundthatdilutionimpactedrowc onditionsmorestronglythan thechemistry.Derudietal.[ 75 ]foundanupperlimitindilutionratioexisted,beyond whichpoorcombustioneciencyandhighCOemissionsresult ed,duetoasignicant reductioninO 2 concentration. 2.2.2RoleofAutoignitionTemperature Theimportanceofauto-ignitiontemperatureinramelessco mbustionhasbeen discussionbyJoannonetal.[ 14 58 ],KatsukiandHasegawa[ 60 ],Blasiaketal.[ 81 ], Tabaccoetal.[ 94 ].Inletreactanttemperaturesabovetheauto-ignitiontem peraturehas beenfoundtohelpstabilizeramesevenathighdilutionleve ls.Highdilutionlevelsand temperatures,andthepresenceofCO 2 andH 2 Oincreasesradiativeloss,assistsreactant autoignitionandpromoteshomogenizationoftemperature[ 94 ]. 2.2.3EectofVitiationonSootFormation Macadam[ 98 ]studiedtheeectofdiluentsinvitiatedcombustioninaJS R.BothN 2 andCO 2 vitiationwerefoundtoreducesooting,andCO 2 seemedtoenhanceoxidation 55

PAGE 56

ofthesootparticles(throughincreaseofOHradicalpoolby thereaction CO 2 + H CO + OH )[ 98 ]. Kowaliketal.[ 101 ]showedthatsootformationinliquidfueledJSRsismarkedl y dierentfromgasfueled.Theytestedspraycombustionofto lueneandisooctanefuels andfoundsootinglevelsdependentstronglyonthespraycha racteristics.Toluene producedmuchgreatersootemissionscomparedtoisooctane .Combustionofethylene andisooctane(premixedandprevaporized)fuelsintheJSRr esultedinobservablesoot formationforequivalenceratiosgreaterthan2.4(at7msre sidencetime)[ 101 ].They foundoppositetrendsinsootgenerationasfunctionsoftem perature,whenliquidand premixedcombustionmethodswereapplied.Thepremixedram esproducedlesssootat highertemperatureswhiletheliquiddropletcombustionre sultedintheoppositetrend, resultingfromunmixednesstypicalofdiusionrames[ 101 ].Thismightbesignicantin explainingresultsfromtheexperimentsonthePoWERengine ,sincethePoWERengine wasoperatedinheterogeneousvitiatedcombustionmode,wi thacorrespondingreduction intemperatureandsooting. Joannonetal.[ 58 ]studiedtheprocessofdilutecombustionofmethaneunderf uel-rich conditions.TheystudiedtheeectofC/Oratioandinlettem peratureonrameless combustion.Theyfoundthattheinletrowdilutioncouldbec orrelatedwiththereduction inacetyleneformation.Theysuggestedthatthepartialoxi dationduetotheinletoxidizer streamaswellasfromCO 2 /O 2 reductionisthecauseforlowluminosityandincreased homogeneityinramelesscombustion.Theyfoundthattheram elesscombustionissimilar toatwo-stageoxidationprocessandthattheunusualproper tiesintheregimeweredueto thestagedprocessandwell-stirringintheoxidative-pyro lyticcondition. AcevesandFlowers[ 77 ]predictedchemicalcompositionandsootprecursorsinthe processoffuel-injectionindieselenginesusingdetailed chemicalkineticsimulationsand simpliedmixing.Thesootprecursorgenerationwasdemons tratedtobeafunctionof equivalenceratioatthetimeofignition,whichisastrongf unctionofmixingtimesand 56

PAGE 57

combustiontemperature.EGRwasfoundtoincreasethesootp recursorformationto maximumandthendecrease(towardsstoichiometricmixture conditions)[ 77 ].IntheSRC, theairentrainmentintotheinjectedfuelstreamiscontrol ledtolowamountsloweringthe combustiontemperature,andhencereducingsootformed.In theMKapproachthemixing ratesareenhancedwhiletheoperatingtemperatureisreduc ed,toavoidignitionuntillow equivalenceratiosarenotachieved. YuanandNaruse[ 92 ]foundmaximumsootconcentrationstobehigherwithhigher preheattemperatures.Further,theyfoundthatathighpreh eat,thesootmaximum concentrationwashigherwithlowerO 2 %(increaseddilution). HuthandLeuckel[ 102 ]studiedsootformationinturbulentplugrowreactorusing naturalgasasaprimaryfuelandpropaneasasecondaryfuela ndforvaryinghotgas temperatures(1273-1773K),residencetimes(30-40ms)and secondaryfuelconcentrations. Thesetupwassimilartoamultistagingconceptandhenceind icativeofvitiation.The primaryzonetemperaturewasvariedbetween(1733-1253K)b ypreheatingair.They observeddecreaseinsootgrowthratesathightemperature, similartopremixedlaminar rames.Theydenedtheinceptionpointasthepointwheresoo tvolumefractionexceeds 10 10 ms 1 atasootvolumefractionof4 10 9 at40ms. Pischingeretal.[ 103 ]studiedsootformationindieselengines.Theyalsoreport ed resultsofsootformationinvitiatedenvironment(fuel-ri chnaturalgascombustion products).Theyfoundacriticalsootingtemperatureof145 0K,underhighpressure vitiatedcombustionconditions,asopposedtolowercritic altemperaturesinshocktube experiments.2.2.4StabilityinFlamelessCombustion Vitiatedcombustioninherentlyispronetoinstabilitydue todilutionofthereactants. Sincethetechnologyofvitiatedcombustionisfairlynew,c ombustionsystemsstillneed tobeoptimizedforthisregime.Severalstudieshavebeenpe rformedtounderstand combustionstabilityinvitiatedcombustion[ 3 10 11 14 17 19 21 { 23 37 44 57 65 57

PAGE 58

74 104 105 ].Someoftheseeortswerebasedonglobalreactionkinetic s,includingthe stabilitystudiesonconventionalsystemsandfuels[ 21 22 106 { 110 ]. YuanandNaruse[ 92 ]performedCFDanalysisandexperimentsonfurnacemild combustion.TheyfoundOHconcentrationstodecreasewithi ncreaseddilutionand decreasingairpreheat.SinceOHemissionshavebeenfoundt ocorrelatewithrame stability[ 18 ],theseresultswouldbeindicativeofthefunctionaldepen denceofstability ondilutionandEGRtemperature.Itwouldalsoberelevantto pointoutthat,although themotivationofthisworkhasbeenindeterminingthelimit ingEGRattheblowout condition,Derudietal.[ 75 ]havefoundthatanupperlimitindilutionratioexisted, beyondwhichpoorcombustioneciencyandhighCOemissions resulted,duetoa signicantreductioninO 2 concentration. Inwell-stirredreactor(WSR)studiesonleanblowout,Stur gessetal.[ 104 ]proposeda nitrogendiluenttechniquetodeterminetheblowoutlimits atsubatmosphericconditions (foraircraftengines).Theircorrelatedblowoutpredicti onsshowedloadingparameter(LP) variationsby1to2ordersofmagnitude,forvariousblowout equivalenceratios. Studieshavebeenperformedonclassicationoftherameles scombustionregime [ 75 105 ],andtheseinsomerespect,areassociatedwiththeblowout stabilitylimitsof theregime.Joannonetal.[ 105 ]investigatedtheclassicationofMILDcombustioninto separateregimesthroughexperimentationandmodelingofc ounterdiusionrames. Fourzoneswereidentied,namely,HomogeneousChargeDiu sionIgnition(HCDI)and auto-ignitivederagrativeregimesforhighinlettemperat ures,andignition-assisted deragrativeregimeandnocombustionregimesatlowinlette mperatures.Further classicationwasbasedonthedegreeoftemperatureriseof theinletmixture,as sub-adiabaticandsuper-adiabaticregimes. Derudietal.[ 75 ]investigatedthemildcombustionofcokeovengas(CH 4 /H 2 40/60 %)inalaboratory-scaleburner.Thetransitiontotheramel essregimewasmarkedby asuddendropinNO x andCOlevels[ 75 ].Horizontalandverticaltransitionzoneswere 58

PAGE 59

identiedonaplotofinlettemperatureanddilutionratio. Anuppertransitionlimitin temperaturewasdeterminedthatleadtoexcessiveformatio nofNO x .Theyalsofound anupperlimitindilutionratio,beyondwhichpoorcombusti oneciencyandhighCO emissionsresulted,duetoasignicantreductioninO 2 concentration. Investigationsontheacousticoscillationsinramelessco mbustionregimehavealsobeen undertaken[ 14 111 ].Leietal.[ 111 ]investigatedtheeectofdilutionoffuelwithsteam andobservedareductionincombustionoscillationsandNO x duetoreductioninrame temperaturecausedfromdilutionwithsteam.Theyalsodisc ussedthatthelinerwall temperaturemayincreasewithincreasingsteamdilution,p ossiblyduetoenhancedwall heattransfer.Joannonetal.[ 14 ]investigatedthetemperatureoscillationsduringmild combustionofmethaneforarangeofstoichiometries,inlet reactanttemperaturesand residencetimes.Modelingresultswerecomparedtoexperim entalresultsinaspherical quartzjet-stirredreactor.Thetemperatureoscillations werefoundtoincreasewith decreasingresidencetimeanddilutionratio,andwerepres entforhighenoughreactor inlettemperaturesintherangeof1000to1300K,andC/Orati oshigherthanabout 0.5.ThecompetitionbetweentheCH 3 radicalrecombinationreactionstoC 2 speciesand oxidativerouteshasbeenassociatedwiththeoscillatoryp henomenon.Adachietal.[ 67 ] reportedsubstantialreductionincombustionoscillation swithmultistaging. EortsonmodelingcombustorsusinganetworkofWSRsandPSR shavealsobeen investigated[ 112 113 ].Manyreactornetworkmodelshavebeenproposedinthepast someofthemapplicableonlytospecicgeometries[ 114 { 117 ].Thezonalconceptwas extendedfurthertothedevelopmentofdetailedcombustorr eactornetworks(CRNs) [ 118 { 121 ],withtheviewofbridgingthegapbetweenthesimplisticmo dularapproach thatignoresruidtransport,andthecomputationalruiddyn amics(CFD)approach, thatfailstoaccountfordetailedreactionkinetics.Howev er,inthiswork,wehaveuseda simplisticcontinuouslystirredtankreactor(CSTR)netwo rkmodelinordertoavoidany uncertaintiesregardingestimationofrow-splitsandinor dertoobtainfundamentaland 59

PAGE 60

sucientlygenericresults,thatmaybeapplicabletoanyap plicationorsystem(furnace, ICengine,gasturbinecombustor,etc.). Partofthecurrentstudyhasbeentodeterminetheblowoutpe rformanceofvarious biofuelsinthevitiatedcombustionregime.Anearlyeortt ostudythevitiated combustionperformanceofconventionalfuelswascarriedo utbyHarwood[ 44 ]using RecirculatingPlugFlowReactor(RPFR)andPerfectly-Stir redReactor(PSR)modelsto characterizecombustioneciencyandstabilityofcombust orsforSCCsystems.Harwood's workwasanextensionofthemodelingeortofCrittenden[ 23 ]thatwasbasedonthe approachofStrehlow[ 21 ].Theseearlyeortswereprimarilybasedonsimpliedglob al kineticrates. Themodelingapproachhasbeenrenedinthisstudy,toinclu dedetailedchemical kineticsforthealternatefuels.Whilethepresentworkpri marilyfocusedonpredicting combustionperformance,signicantexperimentalworkhas beencarriedoutbyother researchers,overtheyears.Thecombustionstabilityforc onventionalopencycleengines usingexperimentalWell-StirredReactors(WSRs)havebeen investigatedbyLongwell andLang[ 106 ],Weissetal.[ 107 ],Ballaletal.[ 108 ],ZelinaandBallal[ 109 ],Blustetal. [ 110 ],Stoueretal.[ 122 ].AsimpliedglobalreactionapproachforOpenCycle(OC) engineshasalsobeenpresentedinSpalding[ 22 ]. 2.2.5BiofuelVitiatedCombustion AshenandCushman[ 123 ]comparedthecombustionofbiodieselsurrogate(methyl butanoate),ethanolanddieselfuelthroughexperimentati onaswellasmodelingofdiesel engine.BiodieselblendsB0,B5andB20wereusedforcompari son.Thecombustion behaviorofB20blenddisplayedalowtemperatureheatrelea se(LTHR)coolrame,before themaincombustionevent[ 123 ].TheeectofEGRwasstudiedbycontrollingtheinlet temperatures[ 123 ].Itwasfoundthat,anintaketemperatureincreasereduced ignition delaywhileincreasingbiodieselpercentageincreasedign itiondelay,reducedunburned hydrocarbon(UHC)emissions(measuredasequivalentCH 4 emissions),butincreased 60

PAGE 61

NO x .Inaddition,ethanolcombustionproducedsignicantCO 2 andCOemissionsandno coolramewasobserved,unlikethecaseofbiodiesel. Zhengetal.[ 66 ]investigatedthecombustionofB100biodieselsderivedfr omsoy,canola andyellowgreaseinLTCusingEGRonadieselengine.Theyfou ndthatultra-lowNO x levelscouldbeachievedbyusingveryhighEGRlevels,andat tributedthereduction tolongermixingtimes(duetoincreasedignitiondelays)an dlowertemperatures, resultingfromvitiation.ThistendstoreduceoverallNO x .Useofbiofuels,hadcompeting eectsonignitiondelaysduetohighcetanenumber(shorten sdelay)andhighviscosity (lengthensdelay)[ 66 ].TheyfoundthatsootwasincreasedwithEGR(hightemperat ure combustionregime)initially,andthenbegantobereduceda tveryhighlevelsEGR(low temperaturecombustionregime).Itmustbenotedthatthete mperatureofEGRstream hasasignicantimpactonresultstoo,however,isfrequent lynotreported. Agarwaletal.[ 124 ]suggestedthatthebiodieselsproducedlowerparticulate matter (PM)duetotheirloweraromaticcontent,higheroxygencont entandshort-chainparan hydrocarbons,incomparisontodieselfuel.Theyperformed experimentsonadiesel engineandfoundthatNO x reducedwithEGRand,theamountofreductionwashigherat higherloads.Theyfound15%EGRproducedlowestNO x andthatthesmokeemission increasedwithEGRandload.Thethermaleciencyincreased withEGRatlowerloads butremainedunaectedathigherloads. 2.3CombustionofAlternateFuels Abriefbackgroundonthecombustionperformanceofvarious biofuelshasbeen presentedinthissection.Severalinconsistencieshavebe enobservedintheresultsfrom dierentresearcheorts,forbiofuelcombustion.2.3.1Syngas Previousstudiesonsyngashaveraisedconcernsoverthecom positionalvariabilityof syngasanditsimpactonturbinedesign.Syngasisalsoknown toproducehighNO x emissionsathighH 2 /COratios,owingtopotentiallyhigherassociatedtempera turesof 61

PAGE 62

combustion,dependingonthemethodofcontrol.Inletcompo sitiondilution,especially withN 2 ,isoftenresortedtolimitthecombustiontemperaturestoa cceptablelimits. Hence,operatingsyngasonasemiclosedcycleisofparticul arinterest,sincetheEGR wouldresultinloweroverallcombustiontemperaturesandl owerNO x levels. 2.3.2Biodiesel Biodieselfuel(BDF)consistsofmonoestersproducedthrou ghtransestericationof vegetableoils.Thetransestericationprocessinvolvesc onversionoftriglyceridesto glycerolandmethylesteruponreactionwithmethanol,inth epresenceofanacidoralkali catalyst.Alkalicatalysesisfasterandusedcommercially fortransesterication[ 124 125 ]. Biodieselfuelshavehigherrashpointandthereforearesaf ertohandle[ 124 ].However, biodieselfuelshaveknownissueswithcorrossion,fuelbui ldupandcoldstartproblems. Bolszoetal.[ 126 ],BolszoandMcDonell[ 127 ]investigatedtheatomization,vaporization, combustionandemissionsofbiofuels.Theyfoundanincreas einNO x andCOemissions resultingfrompooratomizationandvaporizationcharacte risticsofbiodiesels,comparedto dieselfuel.Theyhavereportedinconsistenciesinliterat ureregardingbiofuelcombustion performance[ 126 ].Theyalsosuggestedtheuseofethanol/biodieselblendst oimprove ignitioncharacteristics,attheexpenseofheatingvalueo fthefuel. Muncriefetal.[ 128 ]studiedtheeectofcombiningtheuseofbiodieselandEGRi n ICengines.TheyconcludedthatasignicantreductioninPa rticulateMatter(PM) andNO x wasachievable.TheyreportedthatNO x emissionsdependedonthesourceof biodieselandthathighestNO x levelswereobtainedwiththeuseofultra-lowsulfurdiesel (ULSD),followedbysoy-basedB100fuel,whilecottenseedbasedB100fuelhadlowest NO x levels.Assuggestedbyreferencestherein,NO x emissionswerehigherforfuelsthat wereunsaturatedtoahigherdegree,likecanolaorsoybased biodiesel.Theyfoundan increaseinPMwithEGR,unlikewhathasbeenobservedbyElli setal.[ 12 ]. Metcalfeetal.[ 129 ]comparedthecombustionperformanceoftwobiodieselsurr ogate fuels,frequentlyusedformodelingbiodiesel,namelyethy lpropanoate(C 2 H 5 COOC 2 H 5 ) 62

PAGE 63

andmethylbutanoate(CH 3 CH 2 CH 2 COOCH 3 ),bothwithmolecularformulaC 5 H 10 O 2 Shocktubeexperimentswereperformedfor =0.25-1.5,temperaturerangeof1100-1670 Kandatpressuresof1.0-4.0atm.,tostudytheignitiondela ytimesofthefuels[ 129 ]. Theyfoundethylpropanoateignitesfasterthanmethylbuta noate. 2.3.3PalmMethylEster(PME) Hashimotoetal.[ 130 ]studiedthecombustionofpalmmethylester(PME)ingas turbineengine.PMEisanimportantalternativefuelsince, palmtreeshavehighyieldin comparisontoothercrops[ 130 ].However,palmoilhasalowpourpointandhenceneeds heatingoffuellinestolowerviscositybeforefuelinjecti on,inordertomatchthatofdiesel fuelusedforcomparison[ 130 ].Thesautermeandiameter(SMD)forPMEwasfound tobeslightlylowerthanthatofdieselfuel,forthesameinj ectionpressure[ 130 ].They suggestedthattheabsenceofaromaticcontentandtheprese nceofoxygencontentinthe fuelresultedinlowersootformationinPME,incomparisont odieselfuel.TheCOand totalhydrocarbon(THC)emissionswerefoundtobelowertha n2ppm(correctedto16% O 2 ),forbothfuels.NO x levelsofPMEwerefoundtobelowerthanthatofdieselfuel, andwerefoundtoincreasewithdecreasingpressureofatomi zation,becauseofincreasein SMDresultinginincreasedevaporationtimeandenhancedde greeofdifussion-controlled envelope-combustionofdroplets[ 130 ].Localhightemperaturesatnearstoichiometric conditionsincreaseNO x levels.Hence,lowerNO x ofPMEcouldbeattributedtolower SMDofPMEcomparedtodiesel,forthesameinletkinematicvi scosity[ 130 ]. 2.3.4RapeseedMethylEther(RME) Tsolakisetal.[ 131 ]investigatedtheperformanceofRME/dieselblendsincomp ression ignition(CI)dieselengineswithEGR.Theyfoundthataninc reaseinthepercentageof RMEreducessmoke,UHCandCOemissionsbutincreasesNO x emissions.Theyfound thatRMEhadaneectofreducingignitiondelay,increasing thefuelburninginthe premixedphase,andconsequentlyraisingthepressureandt emperature,andhenceNO x Itwaslesscompressible,andhencetheinjectionpressurew ashigherforRMEblends, 63

PAGE 64

increasingcombustionpressure,temperatureandNO x [ 131 ].Sootreductionwasachieved predominantlyduetohighoxygencontentoffuel,sincetheo xygencontentinthefuel-rich premixedreactionzoneswasincreased.Additionally,RMEh assignicantlylowersulphur content[ 131 ].EGRhadaneectofloweringtheoverallcombustiontemper atureand increasingtheignitiondelay,thusreducingNO x Dagautetal.[ 132 ]modeledthecombustionofRMEinajet-stirredreactorand suggestedn-hexadecaneasamodelfuelsurrogateforRME.Th emodelingcouldnot, however,capturetheearlyCO 2 production,andover-predictedlargeolens.The agreementbetweenexperimentsandmodelingwasreasonably goodotherwise. Karabektas[ 133 ]studiedtheeectofRMEbiodieselandturbocharging,ondi esel engineperformance.Theyattributedtheimprovementinbra kethermaleciency(BTE) ofbiodieseltotheoxygencontentofthefuel,anditshigher cetanenumber.LowerCO emissionsofbiodieselweremainlyanoutcomeoflowerC/Hra tio,higheroxygencontent andhighercetanenumber(whichresultinlowerignitiondel ays)[ 133 ].NO x emissions werefoundtobehigherandwereattributedtotheoxygencont entofbiodiesel. 2.3.5DimethylEther(DME) DMEhaswiderammabilitylimitsandemissionpropertiessim ilartothoseofnatural gas[ 134 ].IthaslowC/Hratioandhashighoxygencontent,isknownto benon-toxic, haslowauto-ignitiontemperature,highcetanenumber,hig hervolatilityandhassoot-free combustion[ 134 ].However,ithasdrawbacksofhavingalowerlubricity,and decreased heatingvalueduetoitsoxygencontent[ 134 ]. Leeetal.[ 135 ]investigatedthecombustionofDMEinamodelcombustortes tfacility. Acomparisonofcombustionperformancewithrespecttometh anewaspresented.They foundthatDMEpresentedlessercombustioninstability,ho wever,ithadatendencyfor rashback.Theyattributedtheenhancedstability(lesserc ombustionoscillations)ofDME comparedtomethane,toitsgreatermolecularmassthatshap estheramelengthwise,and 64

PAGE 65

toitslowerheatreleaserate.TheCOemissionsforDMEwerem uchlower,whiletheNO x emissionswerecomparabletothoseofmethane. Intheirstudy,Sidhuetal.[ 25 ]foundthatDMEwasfoundtohaveanignitiondelaya factorofthreelowerthanthatofcompressednaturalgas(CN G).Biodiesel,ontheother hand,hadafactoroftwotimeslargerignitiondelaythandie selundersimilarconditions, despiteitshighercetanerating[ 25 ].Theysuggestedthatthehigherviscosityofbiodiesel resultedinlargerbiodieseldroplets,contributingtolar gerignitiondelays. 2.3.6Ethanol AshenandCushman[ 123 ]reportedintheirstudy,thatethanolcombustionproduced predominantlyCO 2 andCOemissionsintheexhaust.Themechanismdevelopedby Marinov[ 136 ]hasbeenusedtomodelthecombustionofthefuel. 2.4SootFormation 2.4.1SootStructure,PropertiesandHealthconcerns Sootparticlesconsistlargelyofcarbonatomswithabout13%hydrogen[ 137 ],and maycontainotherelementslikeoxygen[ 138 ].Thephysicalstructureofsootconsists ofplateletsgroupedtoformcrystalites,whichfurthergro uptoformparticles(5-50 nm)[ 137 ].Theseparticlesthenformclusters(0.6-0.8microns)off ractalnatureby agglomeration[ 137 ].Thespherulesofsoot(primarysootparticles)inagglomo rates (secondarysootparticles)areof10to80nmsize,althoughm ostparticlesfallbetween thesizerangeof15-50nm[ 139 ].Theremaybeasmanyas10 5 to10 6 carbonatomsina primarysootparticle[ 139 ]. Sootisacommercialproductusedincopymachinesandasanac tivellerinrubber products[ 9 ].Theradiationfromsootparticlesisprimarilyresponsib leforramespreadin resandcontributestocombustioneciencyinfurnaces[ 140 ].Theradiantsootparticles intherametendtoheatupthecombustorlinerandbladesofth egasturbine[ 141 { 143 ]. Theparticulatesfromthecombustionofcertainfuelshaveb eenfoundtobelinkedtotheir highlevelofsulphurcontent[ 144 ]. 65

PAGE 66

Thehealthandenvironmentaleectsofsootparticulatemat terhavebeenextensively reviewed[ 144 { 147 ].Theparticulatematterisgenerallyclassiedasultran eparticles (lessthanPM0.1),neparticles(lessthanPM1.0)andcours eparticles(greaterthan PM1.0).ThePM10particlesaectupperrespiratorysystem, whiletheneandultrane particlesaectthelowerrespiratorysystemandmayalsoen teralveoli,causingchronic respiratorydisorders[ 144 ].Thepolyaromatichydrocarbon(PAH)moleculesconsistin g ofupto16carbonatoms(pyrene,A4)areprecursorstosootfo rmation.Thesehave beenhypothesizedtodimerizeandformtherstsootnuclei[ 148 { 150 ]andthenfurther growthroughcoagulationandsurfacegrowth.Thecarcinoge niccharacterofthesoot particulateshasbeenattributedtotheconstituentPAHmol eculesaswellastheadsorbed polycyclicorganicmatter(POM)[ 144 ]. Hopke[ 147 ]havediscussedthechangingtrendsinairpollutioneecte dfromchanging globaleconomicandpoliticalscenario's.Further,theydi scussedthethreatfacedinover populated\megacities"likeDhaka,BeijingandKarachi.Th eypresentedairpollution dataofseveralcountriesandcompareditagainsttheUSregu latorystandards.The regulatorystandardsofEnvironmentalProtectionAgency( EPA)[ 1 ]generallycontrol particlesofsizePM10andPM2.5.TheUSNationalAmbientAir QualityStandard (NAAQS)forPM2.5is15 gm 3 (annualaverage)and35 gm 3 (24houraverage), whileforPM10is150 gm 3 (24houraverage)[ 147 ].Thedatafromseveralcountries wascomparedtotheseaveragesanditwasfoundthatamajorit yofthesecountries didnotmeettheUSstandard,evenatthecurrentpollutionle vels.Withtheshiftof manufacturingindustriesfromthewesttoeast,thegrowtho fthedevelopingeconomies intheeastwouldeventuallyberestrictedbytheneedtomeet theincreasinglystringent pollutionstandards[ 147 ].Inanotherrecentstudy,itwasconcludedthatthePMhavea signicantgreehousewarmingeectduetotheirlongreside ncetimeandbroadabsorption spectrum,andthattheirreductionwouldhaveahugeimpacto ncurbingtheglobal warmingproblemintheshortterm[ 144 147 ]. 66

PAGE 67

Dellingeretal.[ 146 ],intheirsummaryofthe10 th internationalcongressonthehealth eectsofcombustionbyproductsreportedthat,70%oftheae rosolsintheatmosphereare releasedfromcombustionand50%areprimarypollutantsfro mthesourcesofcombustion. Theysuggestedthatcombustionsourcesreleasedsignican tnano-organiccarbon(NOC), whichareparticlesofsizesoftheorderof2nm,togetherwit hsootparticlesofsizesinthe rangeof20-50nm.TheNOChavebeenfoundtobehydrophilic,s howinglittletendency tocoagulateandmayberelatedtothetoxicityofsootpartic les[ 146 ].TheseNOCmay beresponsiblefortheoxidativestress,knowntohaveadver sehealtheects[ 146 ].Some oftheirhealthimpactsincludedepositioninlowerrespira torytractandalveoli,increased risktolungcancer,reducedlungfunctioninchildren,chro nicobstructivepulmonary disease(COPD),arrhythmia,atherosclerosis,neurodegen erationofthecentralnervous system,etc.PM2.5havebeenrelatedtoincreasedmortality rates,cardiovascularand respiratorydysfunction,COPD,asthma,pneumonia,etc[ 146 ].OtherreportedPMrelated healtheectsincludelunginrammation,irritationandmay causemyocardialinfraction, tachycardia,increasedbloodpressureandanaemia[ 145 ].Mostairpollutantsincluding PMincreasefreeradicals,normallygeneratedinresponset oextraneouseectsbythe defensesystemsoforganisms.Thisresultsinoxidativestr esslinkedtoseveraldiseaseslike chronicinrammatorydiseases,cataract,centralnervouss ystem(CNS)disorders,etc.In addition,thesemayalsocausehaemotologicalproblems(CO ,benzene)andcancer[ 145 ]. 2.4.2FactorsAectingSootFormation Sootformationisacomplexphenomenonanddependsonsevera lcombustion parameterssuchasfuelstructureandtype,pressure,tempe rature,equivalenceratio, EGR,etc.[ 137 142 143 151 { 153 ].Theeectoffuelstructureonsootformationhas traditionallybeenquantiedusingexperimentalapproach esandparameterssuchassmoke height,luminositynumber(LN),Tesnersootingindex(TSI) [ 154 ],diusionrameblackness parameter(DFBP)[ 155 ]. 67

PAGE 68

Therelativesootingtendencyoffuelmolecularstructureh asbeenextensivelystudied forpremixedanddiusionrames((a)Premixedrames:acetyl ene < alkenes < isoalkanes < n-alkanes < monocyclicaromatichydocarbons < naphthalenes.(b)Diusionrames: parans < monoolens < diolens < acetylenes < benzenes < naphthalenes.[ 154 ]). Aminetal.[ 156 ]performedCFDstudiesandinvestigatedpressureeectson soot formationandNO x .Theyconcludedthatpressurehadasignicanteectonsoot formationduetoareductioninoxidationfromreducedOHcon centrationsathigher pressure,aswellashigherassociatedtemperatures.Furth er,ithasbeenfoundthat pressurehasatendencyofshiftingthesoot-bellmaximumto wardshighertemperatures [ 157 ].Thiseectwasfoundtobestrongeratlowerpressures.Fre nklachetal.[ 157 ] explainedthisphenomenonasresultingfromthepressure-d ependentunimolecular fragmentationroute(slowroute)ofPAH,thatcompeteswith thecondensationofPAH intoPeri-CondensedAromaticHydrocarbons(PCAH)andsubs equentlysoot(fastroute). TheapplicabilityofthemodeldevelopedbyFrenklachandWa ng[ 149 ]wasextendedto highpressureramesbyKazakovetal.[ 158 ].Itwasfoundthatpressurehadatendencyto increaseacetyleneconcentration,andtherefore,surface growth.Brownetal.[ 159 ]modeled theeectofpressureoncombustionofethylenefuelinaPSR. Theyfoundashiftinthe peakofthesoot-belltowardshighertemperatures,accompa niedbyhighersootvolume fractionsathigherpressures. Swirlalsohasaneectonsootformation,andhasbeenfoundt odecreasethesooting levels[ 160 ].Theformationofsoottypicallyhasabell-shapedtempera turedependency andformsatequivalenceratiosgreaterthanabouttwo(crit icalequivalenceratiosarein therangeof1.8to2.1)andtemperaturesbetween1600and260 0K[ 141 ].Thecritical sootinceptiontemperatureisabout1600Kandisknowntobei ndependentoffueltype [ 78 ]. Warnatzetal.[ 161 ]haveshownthetrendofincreasingsootformationwithdecr easing hydrogencontentinthefuel.Thesootformationincreasesw ithpressure,C/Hratio,C/O 68

PAGE 69

ratio,whiledisplayingbell-shapedtemperaturedependen ce.Thisbellshapeddependency isduetotheabsenceofpyrolyticradicalproductsatlowert emperaturesandincreased oxidationathighertemperatures,boundingthesootformat ionprocesstowithin1000 to2000K[ 161 ].ThetemperaturerangeforsootformationaccordingtoWag ner[ 162 ]is around1300to2800K. Forpressureatomizersthemainzoneforsootformationisth eoxygendecientfuel sprayzone[ 142 ].Thesootemissionlevelscanbereducedbyallowingmoreti me, temperatureandturbulence(threeT's)inacombustor[ 161 ],whichontheotherhand resultsinhigherthermalNOproduction.Therefore,thesim ultaneousoptimization particulatecarbon,CO,UHCandNO x emissionscanbechallenging. Theeectofequivalenceratiohasalsobeeninvestigated[ 163 ].Theoddcarbonatom andaromaticspeciesexhibitlargersensitivitytoequival enceratios,incomparisonto thealiphaticmolecules[ 163 ].Moreover,thesootsensitivitytoequivalenceratiowas comparabletothatofbenzene[ 163 ].Theonsetofsootinceptionoccursatanequivalence ratioofabout2,correspondingtoC/Oof0.5[ 161 162 ].Theincipientparticlesare thoughttobeofthesizerangebetween300to1000amuandhave aneectivediameter ofabout1.5nm[ 139 ].Wagner[ 162 ]haspresentedthedataforsootinceptionC/Oratios forvariousexperimentalcongurations,includingbunsen rames,stirredreactorsandrat rames.Basedontheirglobalreaction(asgivenbelow),thec riticalequivalenceratiosmay beobtainedusingequation 2{1 andequation 2{2 C x H y + ( O 2 +3 : 76 N 2 ) 2 CO + y 2 H 2 +( x 2 ) C +3 : 76 N 2 = stoich = 1 x + y 4 (2{1) Therefore, 69

PAGE 70

= 1 x + y 4 = 2 C=O x x + y 4 (2{2) HenceforC 2 H 4 ,withcritical =0 : 60[ 162 ], =1 : 80.Increasingtemperatureincreases thethresholdvalueoftheequivalenceratioforsootformat ion[ 162 ].Akeychallengein sootmodelingisthefactthattheobservationofthePAHspec iesislimitedtosizesless than300amu(usinggaschromatography),whilethesmallest observablesootparticlesizes areoftheorderof1.5nm[ 139 ].Indiusionrames,theeectofdilutionoffuelorair streamsisthatofreductioninthesootingtendency[ 162 ].ThereviewarticlebyMansurov [ 9 ]discussedthepresenceoftwothreshholdsforsootformati onatdierenttemperatures forvariousfuel-airramesatdierentpressures.2.4.3SootFormationStudies Macadam[ 98 ],Kowaliketal.[ 101 ],Stoueretal.[ 122 ],Brownetal.[ 159 ],Manzello etal.[ 164 ]havestudiedsootformationinWSRs.ModelingofsootinWSR /PSR/PFR orothercongurations(bothmodelsandexperiments)haveb eenperformedby[ 58 77 101 122 141 159 164 ].Otherstudiesonsootformationandsootincludethoseof [ 102 103 139 156 165 { 169 ].Similarzero-dimensionaldetailedchemicalkineticstu dies havebeenappliedforramelesscombustion,topredictrames tability[ 14 58 ],and emissions[ 77 87 88 ]. ColketIIIetal.[ 141 ]modeledthesootformationprocessinaWSRforethyleneand ethylene-ethanolmixtures.Theyalsodiscussedtheeecto fheatlossandthesootbell, intheircomparisonoftheexperimentaldatawiththemodel. Theyconcludedthat theadditionofethanolhadaneectofincreasingsoot,duet oloweringoftherame temperature(onhightemperaturesideofthesootbell). SongandZhong[ 166 ]modeledsootformationindieselenginecombustionusingC FD modeling.Theyusedthereducedmechanismofn-heptanepyro lysisandsootformation, combinedwiththemethodofmomentstopredictsootparticle sizedistribution.The overallstoichiometrywasfuel-lean,andtheyfoundthatth eoxidationratesweremuch 70

PAGE 71

largerthanthesootgrowthrates,resultinginshortlifeti mesofthesootparticles.The PAHformationrateswereverylow.Themodelwasfoundtounde rpredictthePAH concentrationsandsootnumberdensity. Brundishetal.[ 167 ]measuredthesootparticlesizedistributioninacombusto rusing scanningmobilityparticlesizing(SMPS).Theyusedasmoke meterandanoptical nephelometerforsootmassmeasurements,andanSMPStoobta insootparticlesize distributionandsootvolumefractiondata.Thesootnumber densitieswerefoundtovary between4.4e7to1.08e6andtheparticlesizesvariedbetwee n60to200nm.Acomparison oftheresultsfromSMPSandsootmassmeasurementsprovided anestimationforsoot density.Theyobtainedadensityestimateof132to3025kg/m 3 ascomparedtovaluesof 1500kg/m 3 and1850kg/m 3 proposedbyotherresearchers.TheFrenklachparticlegrow th modelusesavalueof1800kg/m 3 [ 149 ]. Manzelloetal.[ 164 ]measuredsootinglevelsinaWSR/PFRsetup,usingdierent ial mobilityanalyzer(DMA),ultranecondensationparticlec ounter(UCPC)andtransmission electronmicroscopy(TEM).ThetemperatureintheWSRwasco ntrolledbyaddinginerts (N 2 )totheWSR.HigherN 2 rowsuppressedthepeaksootnumberdensityandreduced theaverageparticlediametersfrom10nmtoabout3nminthet emperaturerangeof 1723-1558K.Theyfoundacriticalsootinceptiontemperatu reofabout1558K. Stoueretal.[ 122 ],Colketetal.[ 170 ]studiedtheeectoffueladditivesonan nheptane/toluenerameinaWSR.Thefueladditivecompounds nitromethane,nitroethane, nitropropane,cyclohenanoane,pyridineandquinolinewer etested.Pyridineandquinoline reducedsootthroughreductionofacetyleneconcentration whiletheotheradditiveswere oxydizingtype,andreducedsootbyvirtueoftheiroxygenco ntent.Theeectofsoot bellwasalsodiscussed.Theyfoundquinolinetobeapooradd itive,whilenitroalkanes, althougheectiveinreducingsoot,increasedNO x .Cyclohexanonewasfoundtoreduce sootoveralltestedequivalenceratios,andhadnoimpacton NO x levels.Theauthorsalso emphasizedthattheeectofadditiveswassystemdependent 71

PAGE 72

Frenklachetal.[ 171 ]modeledthedetailedsootformationchemistryandparticl egrowth andoxidationusingthemethodofmoments.Shocktubepyroly sisstudywasconducted onthefuels1,3-butadiene,benzeneandethene,inordertoc haracterizefueleectson sootformation.Theyfoundthatfuelstructureaectedsoot formationthroughgeneration ofintermediatesthatmayparticipateinthechemistrytofo rmPAHmoleculesand,by aectinggenerationofHatoms.Themodelingofaromaticand aliphaticfuelsrevealed signicantinteractionofaromaticringwithacetylene.Re sultsfromalcohol-benezene mixturecombustionrevealedreductionofsootformationpr obablyduetoreductionofH atomsandincreaseinC 2 H 2 KronholmandHoward[ 168 ]studiedtheparticlegrowthprocessinPFRatequivalence ratioof2.2,rametemperaturesof1520and1620K,atapressu reof1atm.Theyfound oscillatorybehaviorofthesurfacegrowthrateconstantan dproposedthatsuchvariation couldbereconciledbyassumingPAHtobethepredominantgro wthspecies.Theyalso rejectedthepossibilityofsharpdeclineinapparentC 2 H 2 -sootreactivityonreductioninH concentrationsasproposedbyotherresearchers. Brownetal.[ 159 ]modeledsootformationofethylene-ArrameinaPSRusing CHEMKINIII.Theypredictedthevariationsootingcharacte risticswithequivalence ratio,temperatureandpressure.Theyfoundthatwithincre asingpressurethedegreeof couplingofthegas-phaseandsurfacechemistryincreased. Balthasaretal.[ 165 ]modeledtheeectofnitemixingratesonsootformationus ing PaSPFRmodel.Tuovinen[ 169 ]studiedCOformationfromsootandsootparticlegrowth inhotgaslayerusingPSRmodelinChemkin.Thesimulationsw ererunforequivalence ratios0.5-4.0andresidencetimesof0.25-10s.Itwasfound thatthereactionofsootand CO 2 increasedCOgenerationandreducedsootvolumefraction. 2.4.4SootFormationModels Therehasbeensignicantprogressintheeldofsootformat ionoverthepastfew decades.Severalreviewpapers[ 9 138 172 { 177 ]havesummarizedthemostcriticalwork 72

PAGE 73

inthisarea.Severalempiricalandsemi-empiricalapproac hestomodelingsootformation havebeenappliedinthepast[ 137 142 143 152 153 ].Soothasalsobeenmodeled usingCFDstudiesandphenomenologicalmodelsintheworkso f[ 88 156 160 166 ]. Phenomenologically,sootformationinvolvesparticleinc eption,surfacegrowth,coagulation andaggregationprocesses[ 161 ]. Severalsootformationmodelshavebeenproposedinthepast .Someofthenames givenincludeatomiccarbontheory,C 2 theory,C 3 theory,acetylenetheory,polymerization theory,Boudouardreactiontheory,polyacetylenetheory, polyaromatics,etc.[ 9 138 ]. Proposedintermediatesinthesootformationprocessinclu de,ions,PAH,polyacetylinic chains((C 2 H 2 ) n ),polyenes,fullerenes,andneutralradicals[ 178 ].Polyacetelynetheory hasthedrawbackthattherearrangementofpolyacetelynest oaromaticringstructures wouldhaveamuchlongertimescale[ 138 ].Amechanismbasedonpolyenepolymerization reactionshasbeenproposedbyKrestinin[ 179 ].Someofthedetailedmodelsforsoot formationproposedincludethosebyCalcote[ 138 ],Frenklachandcoworkers[ 148 { 150 ]and, Howardandcoworkers[ 180 { 184 ].Thesootaerosoldynamicshavebeenmodeledusing sectioningapproach[ 180 185 186 ]andthemethodofmoments[ 149 187 { 189 ]. HigheraromaticchainsuptoC 60 havebeenmodeledbyRichtermechanism[ 180 { 183 ]. Theclosedringaromaticstructuresaremorestableandabun dantinsootparticles,and areknownasFullerenes[ 9 ].Theopenchainringaromaticstructures,ontheotherhand arenotstable.FullerenemolecularspeciesC 60 ,C 70 andheavierfullerencesuptoC 116 havebeenidentiedinrames[ 9 ].Amorphouscarbon,throughinternalredistribution ofsolid-statecarbon,particledepositionandgasphasere actions,transitionsintoan orderedFullerenestructure[ 9 ].TheCorannulenesubsystemhasbeenfoundinalmost allfullerenes,andthoughtbetheprecursortoFullerenes[ 9 ].Thehydrogenabstraction acetyleneaddition(HACA)mechanismwithcorannulenehasb eenproposedasthegrowth mechanismforfullerenes[ 9 ].Richteretal.[ 180 { 183 ]havemodeledthedetailedchemistry ofsootformationthroughreactionsofPAHs. 73

PAGE 74

ThemodelproposedbyCalcote[ 138 ]wasbasedonionicinteractionsandisnolonger employedinstudies.Calcote[ 138 ]intheircriticalreviewonsootformationprocess presentedseveralsupportingevidencesforsootformation toberesultingfromionic mechanisms.Theysuggestedthationswerepresentinsigni cantconcentrationsinrames toserveasprecursorsforsootformation.Calcote[ 138 ]reviewedtheeectofelectricelds, chemicaladditives,electronaddition,etc.onsootformat ionandcitedpertinentliterature thatdiscussedroleofionicmechanismsinsootformation.F orexample,experimental workhasshownthattheconcentrationofpositivelycharged particlespeaksbeforethe sootinceptioninaburnerrame.Electriceldsmaybeusedto controlsootgrowthby controllingtheresidencetimeofthechargedsootparticle swithintheramezone[ 138 ]. Ithasalsobeenshownthatthermalionizationcannotaccoun tforthepresenceoflarge molecularweightionsinrames[ 138 ]. Thesootformationprocessinthisstudy,hasbeenmodeledus ingdetailedchemical kineticmechanismforsootformationbyFrenklachandcowor kers[ 148 ].Thesoot formationmodelformulatedbyFrenklachandcoworkers[ 148 { 150 ],isamulti-stepprocess startingwithgasphasekinetics. Briery,thesootgrowthmechanismingasphasemayprocedeth rougheitheralow temperaturerouteorahightemperatureroute[ 149 ].Thealiphaticmoleculesundergo dehydrogenationandsequentialpolymerization(HACAreac tions,propargylcombinations, etc)resultinginformationofunsaturatedhydrocarbons,c yclicalmoleculesandeventually aromaticmolecules[ 157 ].Furthergrowthofaromaticmoleculesmayoccurthrough ring-ringcondensationreactions(formationofbiphenyl[ 148 ])orHACAmechanism.The ring-ringcondensationreactionsareknowntoquicklyequi librateandbulkofthearomatic growthprocedesthroughacetyleneadditionroute[ 149 ].Thegrowthofaromaticmolecules beyondacriticalsizeeventuallyleadstonucleation/ince ptionofthree-dimensionalsoot particles,furthermodeledbyaparticlegrowthandoxidati onmodel.Theseparticles arelargeenoughtobeheldtogetherbyVan-der-waalsforces [ 161 ].Thesesootparticles 74

PAGE 75

furtherundergocoagulationandaggregation,resultingin particlegrowth.Simultaneously, theparticlesmayoxidizeorpolymerizeatthesurfaceresul tinginreductionorgrowthof particlesizeandmass.Theentireprocesshasbeenmodeledu singthemomentsapproach [ 187 188 190 ],tocharacterizetheparticlesizedistribution.Thesoot numberdensities, sootvolumefractionandevenopticalpropertiesmaybeesti matedthroughthismodel.Xi andZhong[ 139 ]alsodiscussedthecarbonizationofmaturesootparticles thatresultsin atransitionofamorphoussootparticlestoamoregraphitic structure.Thefuturemodels forsootformation,mayalsobeabletopredicttheparticlem orphologyinadditiontothe otherobservables. Severalmechanismshavebeenproposedtocharacterizethes ootgrowthprocess. Substantialmassadditionoccursthroughthesootgrowthpr ocesswithoutaectingthe numberdensity[ 139 ].Ontheotherhand,theparticlecoagulationprocessincre asesthe sootvolumefractionanddecreasesthenumberdensity,with outaectingthetotalmass [ 139 ].Maussetal.[ 191 ]proposedamodiedHACAschemeforsurfacegrowththatalso accountedforthereductionofsurfacegrowthathightemper atures.Intheirmodelingof laminarpremixedrames,theyshowedthatthedecreaseofrad icalswithdecreasingrame temperatureresultedinreducedsootformationrates.They concludedthattheHACA mechanismwascapableofdescribingthedecayofsootgrowth rate,asinterpretedfrom thephenomenologicalmodel,accurately. Detailsaboutthedetailedkineticsandparticulategrowth andoxidationmodelsmaybe foundintheworkofFrenklachandcoworkers[ 148 { 150 158 171 187 { 190 192 { 201 ].The Frenklachchemicalkineticmechanismconsistsofaromatic moleculesuptofourbenzene ringstructures:benzene(A 1 ),naphthalene(A 2 ),acenephrene(A 3 )andpyrene(A 4 ). SomeoftheimportantintermediatePAHmoleculesincludena phthalene,acenaphthalene, phenanthrene,pyreneandcoronene[ 138 ]. 75

PAGE 76

2.4.5SootFormationChemistry ThechemicalmechanismofsootformationproposedbyAppele tal.[ 148 ]includes initiationreactions,pyrolysisreactions,reactionsfor formationofrstaromaticring,and aromaticgrowthandoxidationviaHACA[ 148 ].Themechanismisbasedonacetylene (even-carbon)channels,incontrasttothemechanism(base donodd-carbonchemistry) proposedbyMarinovetal.[ 148 ]. WangandFrenklach[ 195 ]presentedadetailedchemicalkineticmodel(99species,5 27 reactions)forPAHformationinacetyleneandethylenerame s.Thereactionsoflighter species(C 1 /C 2 Species,H/Oreactions)wastakenfromGRI-Mech1.2.Theyal soadded reactionstoaccountforaromaticringformationthroughth epropargylchannel,the formationofwhichoccurredbyreactionsofC 1 andC 2 molecules[ 195 ].The1,3-butadiene (1,3-C 4 H 6 )andvinylacetylene(C 4 H 4 )formationreactionsproceedthroughreactionsof twoC 2 species(vinyl,acetylene,etc.)orviareactionsofC 1 andC 3 radicals[ 195 ].The ratecoecientfortheabstractionof1,3-C 4 H 6 wasestimated,sinceexperimentaldata wasn'tavailable. C 2 H 3 + C 2 H 2 C 4 H 4 + H C 3 H 3 + CH 2 C 4 H 4 + H C 2 H 4 + C 2 H 3 1 ; 3 C 4 H 6 + H C 3 H 3 + CH 3 1 ; 2 C 4 H 6 1 ; 2 C 4 H 6 + H 1 ; 3 C 4 H 6 + H Formationofthebenzenering,proceededthroughthefollow ingdominantroutes: 76

PAGE 77

n C 4 H 3 + C 2 H 2 n C 6 H 5 phenyl n C 4 H 5 + C 2 H 2 n C 6 H 7 benzene + H Additionalroutesfromn-C 4 H 5 ,n-C 4 H 3 (andresonantlystabilizedisomersi-C 4 H 3 and i-C 4 H 5 )andpropargyl(C 3 H 3 )radicalswerealsoincluded[ 195 ].Thesereactionsresultin theformationofnon-aromaticC 6 H 6 adductthateventuallyisomerizestobenzene,through internalrearrangements.Incomparison,theC 3 H 3 reactionsrequirelessintermediatesteps toformbenzene,andhenceareofhighersignicance.Howeve r,WangandFrenklach[ 195 ] didhighlightthepossibilityofintermediatestepssuchas theformationofnon-aromatic C 6 H 6 adductwithpropargylroutealso,therebytreatingtheprop argylradicalsonlyas globalsteps.Theyalsoconcludedthattheproductionofaro maticsthroughtheC 4 H 3 and C 4 H 5 channelscouldbecomparabletothatfromtheC 3 H 2 channel,dependingonthe rameconditions. n C 4 H 3 + H i C 4 H 3 + H n C 4 H 5 + C 2 H 2 i C 4 H 5 + H C 3 H 3 + C 3 H 3 benzene C 3 H 3 + C 3 H 3 phenyl + H Further,theyhighlightedtheuncertaintiesintheoxidati onreactionsofphenyl/PAH. ThePAHgrowthprocessfollowedtheHACAmechanism[ 195 ].Reactionswithethynyl radical(C 2 H),andtheoxidationchannelsofOH,OandO 2 werealsoadded.Comparison 77

PAGE 78

withexperimentalresultsindicatedunderpredictionofHa ndCO,anoverpredictionof OH,whiletheimportantintermediateswerepredictedwell. ThePAHformationand growthreactionshavebeenfoundtobehighlyreversible,wi ththeformationofstable intermediateslikepyreneandacenaphthaleneproceedingt hroughnearlyirreversible reactions[ 148 ].Thereversibilityofthereactionsteps,hastheeectofb uildinga thermodynamicbarrier,whiletheirreversiblestepshelpt hereactionprogressforward [ 148 ]. RichterandHoward[ 182 202 ]havediscussedtheimportanceofPAHformationroutes throughtherecombinationofpropargyl(C 3 H 3 )andcyclopentadienyl(C 5 H 5 )radicals.It hasalsobeenshownthatHACAmechanismalone,maynotbeadeq uateforpredictionof theratesofformationofsomeofthePAHmolecules[ 181 ]. Thepropargylrecombinationrouteisthemostdominantpath waytoformation ofbenzene[ 148 ].Appeletal.[ 148 ]alsoaddedthereactionstepforadditionofvinyl acetylenetoaromaticradicals.Formationofnaphthalene( A 2 )proceedspredominantly throughthisroute,whilephenanthrene(A 3 )generationmostlyfollowsthering-ring condensationroute(formationofbiphenyl)[ 148 ]. 2.4.6SootNucleation Thesootnucleationphenomenonisnotwellunderstood.Seve ralchemicalspecies havebeenproposedaspotentialsootprecursors.Sootincep tionisbelievedtostartat molecularweightsof500to2000amu(approx.C 40 orlargermolecules)[ 161 ].Schuetzand Frenklach[ 203 ]presentedconvincingargumentsaboutthepossibilityofp yrenemolecules beingprecursorstosootnucleation.Intheircomputationa lstudies,theyfoundthat dimerizationofpyreneresultedinstableintermediatesth atmaybeconsideredassoot nuclei.Glassmanetal.[ 78 ]foundthatsootinceptionbeganaround1600Kandthatthe inceptiontemperaturewasindependentofthefuel.Manzell oetal.[ 164 ]foundacritical sootinceptiontemperatureofabout1558KintheirWSRexper iments.Frenklachand Wang[ 150 ]modeledthenucleationprocessusingmethodoflinearlump ingofaninnite 78

PAGE 79

HACAreactionset.Frenklach[ 190 ]assumedthattherstmoleculeinthepolymerization blockhad24carbonatoms. Milleretal.[ 178 ]showedthatdimerconcentrationsaretypicallylow,inord erfor thesetoproducerequisitenumberofnucleationsites.Howe ver,theirstudywasbased onestimationofequilibriumconstants,towhichFrenklach [ 150 ]arguedthatlow dimerconcentrationscouldstillbejustiedsince,thePAH polymerizationprocessis predominantlygovernedbychemicalkinetics,andmaystill proceedthroughformationof quasi-steadystatestabledimerintermediatesthroughirr eversiblereactionsinachemical kineticsequence. DobbinsandSubramaniasivam[ 204 ]observedsingularpolydispereseincipientsoot microparticles(upto15nm)inlaminarpremixedramesusing TEM.Thesewerefoundto bemorematuresootparticles;dierentfromtheincipientP AHmoleculesproposedby otherresearchers.Theysuggestedthattheseincipientpar ticlesundergodehydrogenation (annealing)atthehightemperatureramefront,toformgrap hiticsootparticles.They hypothesizedthatthesenearlytransparentparticlestoel ectronbeam,werepredominantly madeupofPAHfragments,andthata(liquid-like)coagulati onprocessresultedintheir subsequentgrowth. Glassmanetal.[ 78 ]performedexperimentsoncriticalsootingtemperaturesb y controllingsootinceptionbydilutionwithN 2 .Theyfoundthethresholdtobearound 1600Kirrespectiveoffueltypeanddiluentused.Theyalsod iscussedthattheoxidation ratesarenearlyzerobelow1300K,andthereforesuggesteda noxidativeregimebetween 1300-1600Kforasasootreductionstrategy.2.4.7SootGrowthandOxidation Over75%ofsootmassadditionresultsfromsootsurfacechem istry[ 9 161 ].The particlesizeis,therefore,astrongfunctionofequivalen ceratio,unliketheparticle numberdensity[ 162 ].However,frequentlythisprocessismodeledusingphenom enological 79

PAGE 80

approachequation 2{3 .Othermoredetailedmodelsforsootparticlegrowthhaveal so beenproposed[ 205 ]. df v dt = k sg ( f v 1 f v )(2{3) Appeletal.[ 148 ]presentedrevisedsubmodelsforgasphasechemistry,coag ulationand surfacegrowth(toaccuratelymodelthedecayofsurfacerea ctivity)andvalidatedthe modelsagainstexperimentalresultsforC 2 hydrocarbons.Themodelwasadaptedfroma previouslydevelopedmodelofWangandFrenklach[ 195 ]. Sootoxidationresultsinreductionofsootsizethroughrea ctionwithoxidizersOHand O 2 .TheconcentrationofOradicalsisverylow,andhenceisnot consideredasakey oxidizingspecies.Theprocessofsootoxidation,mayoccur simultaneouslyasinpremixed systems(WSRs,premixedrames,etc.),orinasubsequentpro cessasindiusionrames [ 139 ].Severaloxidationmodelshavebeenproposed[ 139 ],includingthosebyMarshand Kuo,Neohetal.[ 206 ]and,NagleandStrickland-Constable. Neohetal.[ 206 ]suggestedthattheoxidationmodelofNagleandStrickland -Constable maynotbeabletocorrectlypredictoxidationdueadditiona loxidizers(OH,O,etc.), besidesO 2 .Intheirpredictions,thesootoxidationratewasoveresti matedwhenthe oxidationbyOH,infuel-richconditions,wasneglected.Ne ohetal.[ 206 ]studiedthe burnoutofsoot-ladengasesinanoxidativeenvironmentofa secondaryrame.In theirexperiments,therameenvironmentwascontrolledbyv aryingtheamountsof N 2 /CO 2 /CH 4 /O 2 gasesinjectedintothesystem.Thecollisioneciencies(r atioofthe experimentalreactionratestothosedeterminedfromcolli siontheory)werecalculated forOH,O 2 ,O,CO 2 ,H 2 O.TheydiscussedtherelativeoxidativeeectsofO 2 ,OHandO andsuggestedthatOHwasthedominantoxidizerinfuel-rich conditionswhileO 2 beinga signicantcontributerinfuel-leanconditions.Thesigni canceofOwasfoundtobeonly secondaryinfuel-richconditions. 80

PAGE 81

CHAPTER3 DESCRIPTIONOFMODEL 3.1SootFormationModel Currently,oneofthemostwidelyacceptedmodelsofsootfor mationisthedetailed sootformationmodelproposedbyFrenklachandcoworkers[ 149 ],andasummaryof theirmodelhasbeendescribedinthissection.Themodelinc ludesadetailedchemical kineticmechanismforgasphasechemistryofsootformation .Thegasphasechemistry alsoincludesthepolyaromatichydrocarbons(PAH)molecul egrowthupto4aromatic ringpyrene(A 4 )molecule.TomodelplanarPAHandsubsequent3-dimensiona lsoot particulategrowth,acoupledaerosoldynamicsmodelwaspr oposed.Thismodelincludes submodelsforPAHplanargrowthchemistrysimpliedbychem icallumpingtechniques [ 190 207 ],particlenucleation[ 150 ],PAHcondensation,surfacegrowthandoxidation usingsimpliedhydrogen-abstraction-acetylene-additi on(HACA)mechanism,particle coagulation,andparticleaggregation.Thesesubmodelswe recoupledintoamethodof moments[ 187 { 189 ],thatutilizessectioningapproach,commontoaerosolmod els.The gure 3-1 outlinesthebasicprocessesofsootformation. Figure3-1.SootFormation-ParticulateModel 81

PAGE 82

Figure3-2.SootFormation-GasPhaseKinetics Insummary,thedetailedsootformationmodelmaybebroadly categorizedintothe followingsubmodels:1.detailedgas-phasekinetics2.planargrowthofPAH3.particlenucleation4.paticlecoagulation5.paticleaggregation(neglected)6.surfacegrowthandoxidation-PAHcondensation,HACAche mistry 3.1.1MethodofMoments Thismethod[ 149 187 { 189 208 ]allowscharacterizingtheparticledistribution functionsusingsmallernumberofequationstosolveforits moments.Themethodof momentswithinterpolativeclosure(MOMIC)describedinFr enklach[ 187 ]hasbeen developedfromtheMethodIIforpolydispersedistribution s,describedinFrenklachand Harris[ 188 ].Ther th momentofadistributionwithN i particleswithsizem i ,maybe writtenasequation 3{1 .Further,thesizemoments r maybecalculatedbynormalizing 82

PAGE 83

theconcentrationsmoments M r (normalizedto M 0 ).Theparticulatesootmodelcalculates theaveragesootdiameter,sootvolumefraction,sootsurfa cearea,fromcalculationof momentsofthesootparticledistribution,determinedbysu mmingupthecontributionsof moments,asgivenbyequation 3{2 fromeachofthesubmodels. M r = X i m ri N i (3{1) dM 0 dt = R 0 G 0 + W 0 (3{2) dM 1 dt = R 1 + G 1 + W 1 dM r dt = R r + G r + W r where R r G r and W r arethemomentratesofformationduetonucleation, coagulationandsurfacegrowthrespectively.Thebasicass umptionisthattheratesof allthethreeprocessesareadditive[ 187 ].Theindividualratesofeachoftheseprocesses hasbeenderivedinsubsequentsections.Therate W 0 isassumedtobezero,sincethe surfacechemistryisnotexpectedtoaectthenumberdensit ies.Ontheotherhand,the coagulativerstmomentproductionrate G 1 ,istreatedaszero.Thepossibleexplanation beingthat,ifacontrolvolumeisdrawnaroundthesootparti cles,thecoagulative processes,wouldoperateonlywithinthecontrolvolume,no netmasstransferoccurs acrossthecontrolvolume.Inotherwordscoagulationwould onlyredistributethenumber densitieswithintheexistingsizeclasses,butnoaddition almasstransferwilloccuracross theboundarybetweentheparticlesandthegasmixture. Theconsumptionofgasphasespeciesinthegrowthandoxidat ionofsootparticlesis accountedforbycorrectingtheproductiontermsofthegasp hasespecies.Themoments areintegrated,togetherwithotherstatevariablesofther eactor.Theconservation equationsforspeciesandmomentbalancesaregivenbyequat ions 3{3 and 3{4 ,respectively. 83

PAGE 84

Thecorrectiontermforspeciesproductionrate,asgivenby equation 3{5 needstobe applied,inordertoaccountforthespeciesconsumptioninp articulatechemistry. dm j dt CV =(_ m in y j;in m ex y j;ex )+ M j V R j (3{3) dM r dt CV = 1 V R m in in M r;in m ex ex M r;ex + Q r (3{4) Q r istheformationrate( cm 3 s 1 )ofther th moment M r ( cm 3 ).Ifthearraysize ofspeciesproductionrates_ j ismodiedtoincludethemomentformationrates,and thespeciesproductionratesarecorrectedforsootproduct ion,theresultingsystemof equations[ 208 ],isgivenbyequation 3{5 throughequation 3{8 j 0 =_ j +_ w j corr ; j =1 ;:::;n j (3{5) j + r +1 0 = Q r ; r =0 ;:::;n r 1(3{6) dm j dt CV =(_ m in y j;in m ex y j;ex )+ M j V R 0 j ; j =0 ;:::;n j (3{7) dM l dt CV = 1 V R (_ m in M l;in m ex M l;ex )+ Q l ; l =( n j +1) ;::;n j + n r (3{8) Hence,theoutputofsootparticlegrowthmodelyieldsthemo mentproduction rates( Q r )andthespeciescorrectionterms(_ corr j )forspeciesA 4 ,C 2 H 2 ,CO,H,H 2 H 2 O,O 2 andOH.Thesewereobtainedfromthesootparticulatecode[ 208 ].Thespecies conservationequationsintheCanteracontinuouslystirre dtankreactor(CSTR)code,had tobemodiedtocouplethesolutionofthesootmomentsateac htimestep.Aderived classwaswritteninC++,thatmodiedthefunctionalityoft he Reactor classinthe originalCanteradistribution.Thisclassallowedthemodi edconservationequations 84

PAGE 85

tobesolvedbycouplingthemomentproductionandspeciesco rrectiontermsobtained fromsimultaneoussolutionofthefortransootcode[ 208 ],usingexterncommands.Aless accurateapproach[ 158 ]sometimesusedinthepastistosolvesootmomentsdecouple d fromtheCSTR/PSRsolution.3.1.2GasPhaseKinetics Thegasphasechemistryofsootformationhasbeendescribed inAppeletal.[ 148 ], FrenklachandWang[ 149 ],WangandFrenklach[ 195 ],Frenklach[ 197 ],andhasbeen adaptedfromaGRI-Mech1.2,forfuelpyrolysisreactions.A ppeletal.[ 148 ]proposed anewchemicalkineticmechanismwithmodicationstothato fWangandFrenklach [ 195 ].Thestartingpointofformationoftherstaromaticringi svinyladditionto acetylenethroughformationofn-C 4 H 5 ,incaseoflowtemperaturereactionsorn-C 4 H 3 incaseofhightemperaturereactions.Anotherroutetoform ationofbenzenering, isthecombinationofpropargylradicals(C 3 H 3 )[ 149 ].TheformationofhigherPAH moleculesproceedesthroughthe hydrogenabstractionandacetyleneaddition(HACA reactions) .Foraromaticfuels,thecyclizationreactionsofthearoma ticringsbecome predominant.Suchsystems,however,relaxbacktotheHACAm echanism,oncethe acetyleneconcentrationincreasestoacriticalvalue[ 149 ].Intheearlierversionofthe sootmodels,thespeciesA 6 -A 8 weretreatedusingchemicallumpingwhileparticulatesoot wasdenedascumulativearomaticcontentbeyondbenzo[ghi ]pyrelene(A 6 )[ 193 ].Inthe modeldescribedin[ 149 ],thePAHgrowthbeyondacepyrenewasmodeledusingchemica l lumping.Subsequentnucleationprocesswasmodeledasresu ltingfromthecoalescence ofthesePAHmolecules[ 149 ].ThecollisionsleadtodimerationofthePAHmolecules, causingparticulatenucleation.Incurrentversionofthem odel,nucleationwastreatedas resultingfromdimerizationoftwopyrenemolecules[ 208 ].Thereisnoexplicittreatment forPAHplanargrowthbeyondpyrene,whilegrowthuptopyren eisincludedingas phasechemistrysubmodel.However,themomentcontributio nsfromthereducedHACA mechanismforsurfacegrowthmaybederivedthroughthechem icallumpingtechnique. 85

PAGE 86

3.1.3PAHPlanarGrowth ThePAHmoleculesundergoinnitesetofcyclizationreacti onsandgrowinsize. Thisinnitereactionmechanismismodeledusinganiterea ctionset,usingmethodof linearlumpingthatyieldsmomentratesduetoplanargrowth ofPAH.Frenklachand coworkers[ 149 189 207 ]proposedanHACAmechanismoflumpedPAHspecies,to modeltheinnitePAHgrowth.Subsequently,areducedHACAm echanismwasutilized formodelingsurfacechemistrybyafewHACAreactions,aswe llasoxidationwithO 2 andOHspecies[ 148 158 ].AdierentialequationsetformomentratesofthePAH distributionfunction,maybeusedtocharacterizeproles fromthePAHpolymerization reactions.The r th concentrationmomentisgivenbyequation 3{9 whilethesizemoment isgivenbyequation 3{10 M r PAH = 1 X i = i 0 m i r N i PAH (3{9) r PAH = M r PAH M 0 PAH (3{10) where m i givesthenumberofcarbonatomsinthesizeclass i .Itisassumedthatonly Catomscontributetothemassofaparticle,whilethatfromH atomsisneglected. Thegeneralschemeoflinearlylumpingasequentialreactio nmechanismhasbeen showningure 3-3 .ItalsoshowshowtheHACAreactionmechanismdescribedin[ 149 ] maybereducedtoasequentialreactionset,assumingallrea ctionsarerstorderand dependonlyontheconcentrationsofpolymerizingaromatic hydrocarbons.Therate constantshowever,maybefunctionsoftheconcentrationso fnon-polymerizinggas-phase species. Forexample,considerreaction L 2 inthereactionblockdescribedin[ 149 ],givenby A i + C 2 H 2 n A i C 2 H + H .Thisreactionmaybeconsideredtobearstorderreaction A i n A i C 2 H withrateconstant k L 2 = k L 2 ( A;E;n )[ C 2 H 2 ]and k L 2 = k L 2 ( A;E;n )[ H ] 86

PAGE 87

A i A i C 2 H A i C 2 H A i A i + 1 A i + 2 A i + 2 A i + 2 C 2 H k 0 S 11 S 12 S 13 S 14 S 15 k 0 S 21 S 22 S 23 S 24 S 25 Blo ck 1 Blo ck 2 Genera l Schem e fo r Lin e arly L um p in g o f an in fini te -step sequenti a l re act i on me chani s m Re p li c ation Blo c k fo r Lin e ar Lu mpin g of Re actio ns of PA H S j 1 S j 2 S j 3 S j 4 S j 5 k 0 S .. S .. S .. S .. S .. Blo ck k ... .... .... Blo ck .. Figure3-3.LinearlyLumpedSequentialReactionsofPAHMol ecule Let i;j betheblock(orreactionclass)andspeciesindex-variable s,respectively, while r and s betheindicesofreactionsandspeciesineachblock.Thetot alnumberof reactionsinablockis n r andtotalnumberofspeciesinablockis n s .Let S j represent allpossiblereactingspeciesofsizeclass j while S is representspecies s inblock i .Hence, S is = S s +( i 1) n s Thegoalistoestimatetherateofproductionofindividuals peciesinthe innitesequencebyinformationinonereplicatingblock,b ybreakingdownthereaction sequenceintoidenticalnitenumberofreactionclasses[ 149 ].Consider,theproduction ratesofallPAHbeyondtheincipientspecies A i 0 ,whichmaybecalculatedusingequations 3{11 through 3{15 .Therateofformationoftheincipientspecies( k 0 )iscalculatedfrom thegasphasespeciesproductionsratesofsootprecursors, availablefromkinetics-coupled CSTRsolution. d [ S 1 ] dt = k 0 k 1 [ S 1 ]+ k 1 [ S 2 ](3{11) d [ S 2 ] dt = k 1 [ S 1 ] ( k 1 + k 2 )[ S 2 ]+ k 2 [ S 3 ](3{12) 87

PAGE 88

d [ S 3 ] dt = k 2 [ S 2 ] ( k 2 + k 3 )[ S 3 ]+ k 3 [ S 4 ](3{13) d [ S 4 ] dt = k 3 [ S 3 ] ( k 3 + k 4 )[ S 4 ]+ k 5 [ S 5 ](3{14) Generalizingfor j> 1, d [ S j ] dt = k ( j 1) [ S ( j 1) ] ( k ( j 1) + k j )[ S j ]+ k j [ S ( j +1) ](3{15) S = 0BBBBBBBBBBBBBBBBBBBBBBBB@ 1 S 1 S 2 S 3 S 4 S 5 S j ...... 1CCCCCCCCCCCCCCCCCCCCCCCCA K = 0BBBBBBBBBBBBBBBBBBBB@ 100000 k 0 k 1 k 1 000 0 k 1 ( k 1 + k 2 ) k 2 00 0000000 0000000 0000 k j 1 k ( j 1) + k j k j 000000 ... ... ... ... ... ... ... 1CCCCCCCCCCCCCCCCCCCCA 88

PAGE 89

Hencethesequenceofinnitereactionsofspecies S j maybegivenbyequation 3{16 Theproductionratesformomentsforeachofthespecies S j maybefoundfrom equation 3{17 .Equivalentresultsmaybeobtainbyperformingblock-wise summations basedonsizeclass,whilethereactionsmaybegroupedintor eactionclasses,suchthat reactionsofaparticularclassmayhavesimilarrateconsta nts[ 149 ].Theequationsfor suchblock-wisesummationsaregivenbyequation 3{18 andequation 3{19 d [ S ] dt = K [ S ](3{16) dM r dt = d P 1j =1 m j r [ S j ] dt (3{17) =( m 1 r k 0 m 1 r k 1 [ S 1 ]+ m 1 r k 2 [ S 2 ])+ m 2 r ( k 1 [ S 1 ] ( k 1 + k 2 )[ S 2 ]+ k 2 [ S 3 ]) + m 3 r ( k 2 [ S 2 ] ( k 2 + k 3 )[ S 3 ]+ k 3 [ S 4 ])+ ::: + m j r ( k ( j 1) [ S ( j 1) ] ( k ( j 1) + k j )[ S j ]+ k j [ S ( j +1) ])+ ::: = m 1 r k 0 + 1 X j =1 ( m j +1 r m j r )( k j [ S j ] k j [ S j +1 ]) d [ S i;s ] dt = k i; ( s 1) [ S i; ( s 1) ] ( k i; ( s 1) + k i;s )[ S s ]+ k i; s [ S i; ( s +1) ]; s =1 ; 2 ::: ( N s 1)(3{18) dM r dt = m 1 r k 0 + 1 X i =1 ( n s 1) X s =1 ( m i;s +1 r m i;s r )( k i;s [ S i;s ] k i; s [ S i;s +1 ])(3{19) Nowconsider, ( m i;s +1 r m i;s r )=[ m i;s +( m ) s ] r ( m i;s ) r = 1 X p =1 ( rC p )( m ) s p ( m i;s ) ( r p ) +( m i;s ) r ( m i;s ) r 89

PAGE 90

Hence,thesimplicationgivenbyequation 3{20 isobtained,where m i;s mayfurther becalculatedfromequation 3{21 ,ifthethemassaddition(( m ) B )ineachreplicating blockisknown.Itisimportanttonotethatthezero th moment M 0 = k 0 andthatonly when m j +1 6 = m j themomentproductiontermisnon-zero.Hence,themomentpr oduction ratesareaectedonlyby\mass-addition"reactions.Furth ermore,mass-additionsoccur onlyviachangesincarbonatoms,whileHatommassesareigno red[ 149 ].Ifthereare l a mass-additionreactionsinareplicatingblock,themass-a ddition( m ) B ,maybegivenby equation 3{22 ,where m l = m i;s l +1 m i;s l .Theunderlyingassumptionisthat,thetotal massadditionineachreplicatingblock(( m ) B )isconstant. ( m i;s +1 r m i;s r )= 1 X p =1 ( rC p )( m ) s p ( m i;s ) ( r p ) (3{20) m i;s +1 = m 1 ;s +1 +( i 1)( m ) B (3{21) ( m ) B = l a X l =1 m l = N s l X s =1 m i;s l +1 m i;s l = m i +1 ;s m i;s (3{22) Theterm( m i;s +1 r m i;s r ),mayfurtherbeexpressedbyequation 3{23 ,whichupon substitutioninequation 3{19 ,givesequation 3{24 .Sinceonlymass-additionreactions participateinmomentgeneration,equation 3{24 mayberewrittenasequation 3{25 where s l isthespeciesindexsuchthatmass-additiontakesplacebet weenspecies S s l and S s l +1 .Theequation 3{25 mayberewrittenasequation 3{26 byusingthevariable transformation i = b +1. 90

PAGE 91

m i;s +1 r m i;s r =( m 1 ;s +1 +( i 1)( m ) B ) r ( m 1 ;s +( i 1)( m ) B ) r (3{23) =[( m 1 ;s +1 ( m ) B )+ i ( m ) B ] r [( m 1 ;s ( m ) B )+ i ( m ) B ] r = r X p =0 ( rC p )( m 1 ;s +1 ( m ) B ) r p ( i ( m ) B ) p r X p =0 ( rC p )[( m 1 ;s ( m ) B ) r p ( i ( m ) B ]) p = r X p =0 [( m 1 ;s +1 ( m ) B ) r p ( m 1 ;s ( m ) B ) r p ]( i ( m ) B ) p dM r dt = m 1 r k 0 + 1 X i =1 ( N s 1) X s =1 r X p =0 ( rC p ) ( m 1 ;s +1 ( m ) B ) r p ( m 1 ;s ( m ) B ) r p ( i ( m ) B ) p ( k i;s [ S i;s ] k i; s [ S i;s +1 ])(3{24) dM r dt = m 1 r k 0 + 1 X i =1 l a X l =1 r X p =0 ( rC p ) ( m 1 ;l +1 ( m ) B ) r p ( m 1 ;l ( m ) B ) r p ( i ( m ) B ) p ( k i;s l [ S i;s l ] k i; s l [ S i;s l +1 ])(3{25) dM r dt = m 1 r k 0 + l a X l =1 r X p =0 ( rC p ) ( m 1 ;l +1 ( m ) B ) r p ( m 1 ;l ( m ) B ) r p ( m ) B p 1 X b =0 ( b +1) p ( k b +1 ;s l [ S b +1 ;s l ] k b +1 ; s l [ S b +1 ;s l +1 ])(3{26) Deningnewvariables, [ n ] m l and R l [ n ] ,givenbyequation 3{27 andequation 3{28 respectively,allowsthesimplicationofthemomentequat iontoequation 3{29 .Note thatthereactionratevariables k b +1 ;s arefunctionsofnon-polymerizingspecies.This mathematicalmanipulationisrequiredduetotheassumptio nthatalltheequations 91

PAGE 92

arerstorderanddependonlyontheconcentrationsofpolym erizingPAHmolecules. However,eachreplicatingblock,hasthesamebasicreactio nmechanism,since,every higherorderPAHmolecule,reactsinasimilarfashionasthe speciesintherstreplicating block.Hence, k b +1 ;s = k b +2 ;s = k b + x;s ; x =1 ::: 1 .Furthermore,fromtheassumptions ofchemicalsimilarity,thespecies S b +1 ;s canberelatedtothespecies S 1 ;s intherst replicatingblockasfollows:[ PAH ] b =[ PAH ] i 0 +3 b where i 0 denesthesizeofthespecies S 1 ;s .Forexampleif, S 1 ;s = A i ,then S b +1 ;s = A i +3 b [ 149 ]. [ n ] m l =[( m 1 ;l +1 ( m ) B ) n ( m 1 ;l ( m ) B ) n ](3{27) R l [ n ] = 1 X b =0 ( b +1) p k b +1 ;s l [ S b +1 ;s l ] k b +1 ; s l S b +1 ;s l +1 (3{28) = k b +1 ;s l 1 X b =0 (( b +1) p [ S b +1 ;s l ]) k b +1 ; s l 1 X b =0 ( b +1) p S b +1 ;s l +1 dM r dt = m 1 r k 0 + r X p =0 ( rC p )( m ) B p l a X l =1 [ r p ] m l R l [ p ] (3{29) Theequationsoftheconcentrationmomentsassummarizedin [ 149 ],aregivenby equations 3{30 through 3{37 .Equation 3{36 givesanexampleofnetrux R l [0] .Forthe caseof R l [ j ] ,thezero th lumpedspeciesmomentmustbereplacedbythej th lumpedspecies momentgivenbyequation 3{37 dM 0 PAH dt = r 0 (3{30) dM 1 PAH dt = m 0 r 0 + l c X l =1 m l r R l [0] (3{31) dM 2 PAH dt = m 0 2 r 0 + l c X l =1 [2] m l R l [0] +2 c m l c X l =1 m l R l [1] (3{32) 92

PAGE 93

dM r PAH dt = m 0 r r 0 + r 1 X j =0 rC j ( c m ) j l c X l =1 r j m l R l [ j ] (3{33) [ p ] m l =( m l c m ) p ( m l 1 c m ) p (3{34) for l =1 ; 2 :::;l c c m = l c X l =1 m l = l c X l =1 ( m l m l 1 )= m i + l c m i (3{35) R l [0] = k L 1 [ C 2 H 2 ] 1 X i =0 [ A i 0 +3 i ] k L 1 [ H ] 1 X i =0 [ A i 0 +3 i C 2 H ](3{36) S [ PAH ] i 0 j = 1 X i =0 ( i +1) j [ PAH ] i 0 +3 i (3{37) 3.1.4PaticleCoagulation Frenklach[ 189 ]describedthecoagulationmodelwiththeassumptionthatt he collisionfrequencywasindependentoftheparticlesize.T hismethodologyextendedthe workofSmoluchowskiformonodisperseinitialsizedistrib ution.Thisapproachfollows adiscretizationapproachwithaself-preservingparticle sizedistributionassumption, meaningthatanasymptoticsolutionexistedforthereduced particlesizedistribution[ 189 ]. Althoughtheparticlesizedistributionatthetimeofincep tionmaynotbeself-preserving [ 150 ],thisassumptionisfundamentaltothesolutionapproach. Theinitialparticle-size distributionneedstobenarrowinorderforasolutiontoexi st,sincethetimerequired toattainconvergenceforapolydispersedistributionisaf unctionofthevarianceof initialsizedistribution[ 189 ].Theparticlecoagulationmodelusedinthisstudyisthe onepresentedinFrenklach[ 187 ],whichhasbeendevelopedfromMethodIIdescribedin FrenklachandHarris[ 188 ].ThecoagulationrateisassumedtodependontheKnudsen number.DependingonthevalueoftheKnudsennumber,thecoa gulationprocessis 93

PAGE 94

+ PAH PAH Soo t Incep t i on : PAH-PAH C ond en s a tion R ea ct i ons m i m j m i j + Soo t Co agul ation : Particl e Co ale scen s e R ea ct i ons m i m j m i j m i + j j < i j > 0 m j + Figure3-4.ProcessofSootCoagulationclassiedintothreeseparateregimes,thecontinuumregim e,transitionregimeandthe free-molecularregime.Thedescriptionoftheprocessinth eseregimeshasbeendiscussed insections 3.1.4.1 through 3.1.4.3 ,whilefurtherdetailsmaybefoundin[ 187 ].Further, thecoagulationprocessisassumedtoberesultingfromcoal escentcollisionsinvolving dimerizationreactions,suchthatthesphericalshapeofth eparticlesispreserved[ 189 ]. Theprocessofcoagulationhasbeenillustratedingure 3-4 3.1.4.1Continuumregime Theprocessofcoagulation[ 187 ]issaidtobeincontinuumregimewhen Kn<< 1, thatis,whenparticlesizesarelarge.Forthisregime,thec ollisioncoecientisgivenby equation 3{38 C istheCunninghamslipcorrectionfactorand K c =2 k B T= 3 .The correctionfactor C isafunctionofKnudsennumberandisgivenby C =1+1 : 257 Kn Substitutingtheseexpressionsinequation 3{38 ,givesequation 3{39 i;j c = K c C i m i 1 3 + C j m j 1 3 m i 1 3 + m j 1 3 (3{38) 94

PAGE 95

i;j c = K c 1 m i 1 3 + 1 m j 1 3 m i 1 3 + m j 1 3 +1 : 257 K c Kn i m i 1 3 + Kn j m j 1 3 m i 1 3 + m j 1 3 (3{39) TheKnudsennumber( Kn )mayberelatedtothemassofthesootparticle byequation 3{40 ,where K c 0 =2 : 514 6 1 3 .Substitutingtheseexpressionsinto equation 3{39 givesequation 3{41 Kn i = d i = 2 = 1 2 6 m i 1 3 =2 6 m i 1 3 = Kc 0 1 : 257 m 1 3 i (3{40) i;j c = K c 1 m i 1 3 + 1 m j 1 3 m i 1 3 + m j 1 3 + K c K c 0 1 m i 2 3 + 1 m j 2 3 m i 1 3 + m j 1 3 (3{41) Ifwesubstitutetheaboveexpressionsof i;j intheexpressionsforcoagulationrates, wecanobtainanexpressionoftheratesintermsofthefracti onalmomentsofthesoot particles,asgivenbyequation 3{42 Considerrstthecoagulationrateofformationofzero th moment.Itsexpressionis givenasequation 3{42 ,whichmayfurtherbesimpliedtoequation 3{43 .Substituting normalizedsize-moments r = M r M 0 inequation 3{43 givesequation 3{44 G 0 c = dM 0 dt = 1 2 1 X i =1 1 X j =1 i;j N i N j (3{42) = 1 2 1 X i =1 1 X j =1 K c 1 m i 1 3 + 1 m j 1 3 m i 1 3 + m j 1 3 + K c K c 0 1 m i 2 3 + 1 m j 2 3 m i 1 3 + m j 1 3 # N i N j = 1 2 K c 1 X i =1 1 X j =1 h 2+2 m i 1 3 m j 1 3 + K c 0 2 m i 1 3 +2 m i 1 3 m j 2 3 i N i N j 95

PAGE 96

G 0 c = 1 2 K c h 2 M 0 2 + M 1 3 M 1 3 + M 1 3 M 1 3 + K c 0 2 M 1 3 M 0 +2 M 1 3 M 2 3 i (3{43) = K c h M 0 2 + M 1 3 M 1 3 + K c 0 M 1 3 M 0 + M 1 3 M 2 3 i G 0 c = K c M 0 2 h 1+ 1 3 1 3 + K c 0 1 3 + 1 3 2 3 i (3{44) Similarly,the r th coagulationmomentratemaybeexpressedbyequation 3{45 Substitutingthedenitionofconcentrationmomentandsiz emomentsintoequation 3{45 givesequation 3{46 andequation 3{47 ,respectively. G r c = dM r dt = 1 2 ( 1 X i =1 1 X j =1 r 1 X k =2 [ K c [ 1+ m i 1 3 m j 1 3 + m j 1 3 m i 1 3 +1 + K c 0 ( m i 1 3 + m i 1 3 m j 2 3 + m j 1 3 m i 2 3 + m j 1 3 )]]( rC k ) m i r m j r k N i N j )(3{45) G r c = 1 2 K c r 1 X k =2 [( M k M r k + M k + 1 3 M r k 1 3 + M k 1 3 M r k + 1 3 + M k M r k )+ K c 0 ( M k 1 3 M r k + M k + 1 3 M r k 2 3 + M k 2 3 M r k + 1 3 + M k + 1 3 M r k 1 3 )](3{46) G r c = 1 2 K c ( M 0 ) 2 r 1 X k =2 [ 2 k r k + k + 1 3 r k 1 3 + k 1 3 r k + 1 3 + K c 0 ( k 1 3 r k + k + 1 3 r k 2 3 + k 2 3 r k + 1 3 + k + 1 3 r k 1 3 )](3{47) 3.1.4.2Free-molecularregime Theexpressionforcollisioncoecientinfree-molecularr egime( Kn>> 1)isgiven byequation 3{48 ,where K f maybeobtainedfromequation 3{49 .Substitutingthese equationsforthefree-molecularregimeintothezero th coagulationmomentrateequation 96

PAGE 97

givesequation 3{50 .Inequation 3{50 ,duetothe p m i + m j term,itisimpossibleto writetheaboveexpressionfor G 0 f intermsofthefractionalmoments.Instead,anew functioncalledthe\gridfunction"wasdened,whichcanbe describedintermsof momentsandcanbeinterpolatedforgridfunctionvalues.Th egridfunction( f l )isgiven byequation 3{51 andequation 3{52 .Thegridfunctionsmayfurtherbeinterpolated between,thegridfunctionvaluesfor f 0 f 1 and f 2 ,whichmaybeevaluatedfromequations 3{53 through 3{55 .Similarly,theexpressionforthe r th coagulationmomentrateisgiven byequation 3{56 .Furtherdetailsmaybefoundin[ 187 ]. i;j f = K f s 1 m i + 1 m j m i 1 3 + m j 1 3 2 (3{48) where, K f = s 6 k B T 3 4 1 6 (3{49) G 0 f = dM 0 dt = 1 2 1 X i =1 1 X j =1 i;j N i N j (3{50) = 1 2 1 X i =1 1 X j =1 [ K f s 1 m i + 1 m j ( m i 1 3 + m j 1 3 ) 2 ] N i N j = 1 2 K f 1 X i =1 1 X j =1 [ p m i + m j m i 1 2 m j 1 2 ( m i 1 3 + m j 1 3 ) 2 ] N i N j f l = 1 X i =1 1 X j =1 [( m i + m j ) l m i 1 2 m j 1 2 ( m i 1 3 + m j 1 3 ) 2 ] N i N j (3{51) f l ( x;y ) = 1 X i =1 1 X j =1 [( m i + m j ) l m i x m j y ( m i 1 3 + m j 1 3 ) 2 ] N i N j (3{52) 97

PAGE 98

f 0 = 1 X i =1 1 X j =1 [ m i 1 2 m j 1 2 ( m i 2 3 +2 m i 1 3 m j 1 3 + m j 2 3 )] N i N j (3{53) = 1 X i =1 1 X j =1 ( m i 7 6 m j 1 2 +2 m i 5 6 m j 5 6 + m i 1 2 m j 7 6 ) N i N j =2[ M 7 6 M 1 2 + M 5 6 M 5 6 ] f 1 = 1 X i =1 1 X j =1 [( m i + m j ) m i 1 2 m j 1 2 ( m i 2 3 +2 m i 1 3 m j 1 3 + m j 2 3 )] N i N j (3{54) = 1 X i =1 1 X j =1 ( m i + m j )( m i 7 6 m j 1 2 +2 m i 5 6 m j 5 6 + m i 1 2 m j 7 6 ) N i N j = 1 X i =1 1 X j =1 [( m i 13 6 m j 1 2 +2 m i 11 6 m j 5 6 + m i 3 2 m j 7 6 )+ (( m i 7 6 m j 3 2 +2 m i 5 6 m j 11 6 + m i 1 2 m j 13 6 )] N i N j =2[ M 13 6 M 1 2 + M 3 2 M 7 6 +2 M 5 6 M 11 6 ] f 2 = 1 X i =1 1 X j =1 [( m i + m j ) 2 m i 1 2 m j 1 2 ( m i 2 3 +2 m i 1 3 m j 1 3 + m j 2 3 )] N i N j (3{55) = 1 X i =1 1 X j =1 ( m i + m j )( m i 7 6 m j 1 2 +2 m i 5 6 m j 5 6 + m i 1 2 m j 7 6 ) N i N j = 1 X i =1 1 X j =1 ( m i + m j )[( m i 13 6 m j 1 2 +2 m i 11 6 m j 5 6 + m i 3 2 m j 7 6 )+ (( m i 7 6 m j 3 2 +2 m i 5 6 m j 11 6 + m i 1 2 m j 13 6 )] N i N j =2[ M 19 6 M 1 2 + M 5 2 M 7 6 +2 M 5 6 M 17 6 + M 13 6 M 3 2 + M 11 6 M 11 6 ] 98

PAGE 99

G r f = dM r dt = 1 2 ( 1 X i =1 1 X j =1 r 1 X k =2 i;j f ( rC k ) m i r m j r k N i N j )(3{56) = 1 2 ( 1 X i =1 1 X j =1 r 1 X k =2 [ K f p m i + m j m i 1 2 m j 1 2 ( m i 1 3 + m j 1 3 ) 2 ]( rC k ) m i r m j r k N i N j ) = 1 2 r 1 X k =2 ( rC k ) f 1 2 k 1 2 ;r k 1 2 N i N j 3.1.4.3Transitionregime Theexpressionofratesforthetransitionregime[ 187 ]maybeobtainedfromthe geometricmeanofthelimitingcasesofcontinuumandfree-m olecularregimes,givenby equation 3{57 G tr = G r f G r c G r f + G r c (3{57) 3.1.4.4Interpolativeclosure Thefractionalmomentsintheexpressionsformomentratesw ereinterpolated betweenthewholeordermomentsobtainedfromthemomentrat eexpressions G 0 ;G 1 :::G r Thepositive-sideinterpolationusesthelagrangeinterpo lationtechniquebetweenthe wholeordersize-moments.Thegeneral n th orderinterpolationatpoint x betweengrid points x i forknown y i = y ( x i )isgivenbyequation 3{58 ,wherethepolynomials l i ( x ), maybeevaluatedfromequation 3{59 .Theinterpolationof ( p )followsthelogarithmic lagrangeinterpolationgivenbyequations 3{60 through 3{62 P n ( x )= n X i =1 l i ( x ) y i (3{58) l i ( x )= n Y j =1( j 6 = i ) x x j x i x j (3{59) 99

PAGE 100

m i m j m i + N j N j j > 0 + Soo t Agg reg ation : Mul t i -Particl e F ra ct a l A ggre ga t i on Re actio ns Figure3-5.ProcessofSootAggregation log ( p )= L P ( log ( 0 ) ;log ( 1 ) :::log ( r max )); p> 0(3{60) log ( p )= P n ( p )= n X i =1 l i ( p )[ log ( i )](3{61) p =10 P n ( p ) = n Y i =1 [ i ] l i ( p ) (3{62) Thegridfunctions( f l and f ( x;y ) l )maybeinterpolatedusingasimilarapproach. Furtherdetailsoninterpolativeclosuremaybefoundin[ 149 187 ]. 3.1.5PaticleAggregation Theaggregationofparticlesthroughnon-coalescentoragg lomerativecollisionsresults intheformationofself-similarfractalstructures,assho wningure 3-5 .Thispartof submodelwasomittedfromtheanalysis,sinceaggregationw ouldonlyoccurathighsoot concentrations,however,weareinterestedinmodelingthe ramelesscombustionregime, withlowsootconcentrations.Detailsonthefractalaggreg ationsubmodelmaybefoundin theworkof[ 165 196 198 { 200 ]. 3.1.6SurfaceChemistry-GrowthandOxidation ThesootsurfacechemistryismodeledbyareducedHACAmecha nism.Theprocess hasbeendepictedingure 3-6 .ThereactiondatafortheHACAmechanismhasbeen 100

PAGE 101

Active si t e X i m i 1 m i = m i 1 + D m m i + D m, C 2 H 2 D m, OH/O 2 Active si t e X i m i j m i + PAH PAH Su r face C on den sation Re act i on s H eterog eno us Surface R eactio ns + PAH c c m i + j m j j < i j > 0 m j Figure3-6.ProcessofSootSurfaceReactionstakenfromAppeletal.[ 148 ].Appeletal.[ 148 ]proposedchangestothereducedHACA mechanismforsootsurfacegrowthsubmodelofKazakovetal. [ 158 ].Thesubmodel utilizesthetechniqueoflinearlumpingdescribedin[ 190 207 ].Themethodofmoments [ 149 187 { 189 ]isusedtoderiveinformationregardingtheparticle-size distribution.The relevantprocedureandequationshavebeendiscussedindet ailinsection 3.1.3 The parameter(representssootparticlesurfaceageing/annea ling),inthesurface growthmodelwasinitiallyconsideredtobeaconstant[ 150 ].Later,itwasmodied toincludeafunctionaldependenceonthemaximumrametempe rature[ 158 ].Appel etal.[ 148 ]furtherincludeddependenceofthe ontheparticlesize 1 andlocalrame temperatureT,asgivenbyequation 3{63 ,wheretheconstantsaandbweregivenby a =12 : 65 0 : 00563 T and b = 1 : 38+0 : 00068 T [ 148 ]. = tanh a log ( 1 ) + b (3{63) Thereducedsetofheterogeneoussurfacereactions(HACAme chanism)betweenthe sootparticlesandgasphasespeciesmaybefoundinAppeleta l.[ 148 ],Revzanetal. [ 208 ].Thesereactionsaddorremovemassfromthesootsurface,r esultingingrowthor oxidationoftheparticle.Asaresult,thesizeclassofanis olatedsootparticlemayshift 101

PAGE 102

downorupbythedierence m inthesizebinofthePSDF.Inthesurfacereactions, C s [ i ] H representstheactivearm-chairsitethatmayundergosurfa cereactionswhile C s [ i ] representstheactiveradicalsiteonthesurfaceofasootpa rticle.Notethatthe surfacereactions,dependontheconcentrationsofthegasphasespeciesandthetotal activereactionsites[ X i site ]onasootparticle.Theaveragesurfacedensityofthereact ion sitesmaybegivenbyequation 3{68 Thesootmass,surfaceareaandvolumefractionarerelatedt othesizemoments [ 208 ].Theequationsforthesootmass,surfacearea s andsootvolumefraction f v may becalculatedfromequations 3{64 3{65 and 3{67 ,respectively.Fromequation 3{64 for sootmass,theparticlediameter( d i ),maybedeterminedasgivenbyequation 3{66 ,where C d = h 6 m C i 1 3 .Substitutingtheexpressionfor d i fromequation 3{66 inequation 3{65 allowscalculatingtheexpressionfor s intermsofthemoments. m i m C = 6 d i 3 (3{64) s = X i N i s i = X i N i d i 2 = h d i 2 i N tot (3{65) = C d 2 2 3 M 0 d i = 6 m i m C 1 3 = m 1 3 i C d (3{66) f v = X i N i v i = 6 X i N i D i 3 = 6 h D i 3 i N tot (3{67) = 6 C d 3 1 M 0 102

PAGE 103

< [ X i site ] > = s N A = P 1i =1 X i site s i N i P 1i =1 s i N i (3{68) s isthenumberofactivesitesonthesootparticlesurface, isthesterichindrance factorand s isthesurfacearea.Theconcentrationfor C s [ i ] canbedeterminedfrom applicationofquasi-steady-stateapproximation(QSSA). Thetotalactivesites( X i site )availableareasumofthesites C s [ i ] H aswellas theradicalsites C s [ i ] .Theaverageovertheentiredistributionofparticleswith number density N i andsurfacearea s i ,givestheaveragedactivesitecoverage( < [ X i site ] > ),as givenbyequation 3{68 .Theconcentrationof C s [ i ] H isgivenbyequation 3{69 ,while equation 3{71 allowscalculationof C s [ i ] .Theradicalsitesareassumedtobeextremely reactiveandhence,quasi-equilibrated.Hence,therateco nstant k s = k s j [ X j ][ X i site ]where X j isthegasphasespeciesreactingwiththeactivereaction-s ite X i site ofthe i th soot particle.Thereactionbeingthe j th HACAreactionproceedingatarate k s j .Thevalueof m is2foracetylenereactions,-2forreactionsofO 2 and-1forOHreactions. h C s [ i ] H i = s N A s (3{69) d h C s [ i ] i dt = k S 1 h C s [ i ] H i [ H ] k S 1 h C s [ i ] i [ H 2 ]+ k S 2 h C s [ i ] H i [ OH ] k S 2 h C s [ i ] i [ H 2 O ] k S 3 h C s [ i ] i [ H ] k S 4 h C s [ i ] i [ C 2 H 2 ] k S 5 h C s [ i ] i [ O 2 ]=0(3{70) h C s [ i ] i = h C s [ i ] H i ( k S 1 [ H ]+ k S 2 [ OH ]) k S 1 [ H 2 ]+ k S 2 [ H 2 O ]+ k S 3 [ H ]+ k S 4 [ C 2 H 2 ]+ k S 5 [ O 2 ] = k ratio h C s [ i ] H i (3{71) where k ratio = ( k S 1 [ H ]+ k S 2 [ OH ]) k S 1 [ H 2 ]+ k S 2 [ H 2 O ]+ k S 3 [ H ]+ k S 4 [ C 2 H 2 ]+ k S 5 [ O 2 ] 103

PAGE 104

Themodelformomentcontributionsfromthesurfacegrowthp rocesshasbeen discussedin[ 149 187 ].Thesurfacegrowthprocessisasequentialgrowthprocess ,such thattheparticlesizeclassmayshiftbetweenadjacentbins basedonthemass-additionor removalbyspeciedmonomerunits.Theequationsforsuchas urfacegrowthbetweensize classesi-1,iandi+1,asdescribedin[ 187 ],aregivenbyequation 3{73 andequation 3{74 Therateconstant( k s )usedintheseequationsmaybeevaluatedusingequation 3{72 Substitutingthedenitionofrateconstant( k s )fromequation 3{72 andsurfacearea ( s )fromequation 3{65 intoequation 3{74 ,allowscalculationofthesurfacegrowth momentproductionratecontributionsfromtheindividuals pecies.Themomentrate contributionsforsurfacegrowthfromspecies C 2 H 2 OH and O 2 .maybeevaluatedbased onequation 3{75 throughequation 3{77 k s = k s j [ X j ][ X i site ]= k s j [ X j ] s N A s (3{72) dN 1 dt = k s N 1 s 1 m (3{73) dN i dt = k s m ( N i 1 s i 1 N i s i ); i =2 ; 3 ::: W r = dM s r dt (3{74) = k s m 1 X i =2 ( m i 1 + m ) r ( N i 1 s i 1 m ri N i s i ) # ; i =2 ; 3 ::: = k s m 1 X i =1 r 1 X k =0 ( rC k ) m i k m r k N i s i +( m i ) r ( m ) 0 N i s i 1 X i =1 ( m i r N i s i ) # = k s m 1 X i =1 r 1 X k =0 ( rC k ) m i k m r k N i s i ; i =1 ; 2 ; 3 ::: 104

PAGE 105

W r C 2 H 2 = k S 4 [ C 2 H 2 ] s h C s [ i ] i C d 2 m r 1 X k =0 ( rC k ) M l + 2 3 soot (2) r k ; i =1 ; 2 ; 3 ::: (3{75) W r O 2 = k S 5 [ O 2 ] s h C s [ i ] i C d 2 m r 1 X k =0 ( rC k ) M l + 2 3 soot ( 2) r k ; i =1 ; 2 ; 3 ::: (3{76) W r OH = k S 6 r OH q k B T 2 m OH [ OH ] C d 2 m r 1 X k =0 ( rC k ) M l + 2 3 soot ( 1) r k ; i =1 ; 2 ; 3 ::: (3{77) Thecollisionsaleciency( r OH )whichistheratiooftheexperimentaltocollisional theoryrateconstantforOHreactions,wastakenfromNeohet al.[ 206 ].Thesurface processesdonotchangethenumberdensity,butonlyaectth eparticlesize,hence,the zero th momentisnotaected.Asaconsequenceofthesurfacereacti ons,someofthe surfacespeciesareconsumed,andhencecorrectiontermsne edtobeaddedtothegas phasespeciesbalanceequations,ateverytimestep.Thecor rectiontermsforthespecies productionratesarecalculatedfromthetotalmassgrowtho fsootparticlesduetosurface growth(whichisafunctionof W 1 ).OnlyspeciesC 2 H 2 ,OHandO 2 contributedtomass additionorremovalfromthesootparticles.Anadditionalt ermaccountsforproduction ratesofspeciesinvolvedinthechemistrythataccountsfor depletionorappearanceof h C s [ i ] i and h C s [ i ] H i sites.Thesereactionsdonotcontributetowardsmassaddit ion orremovalfromsootparticles,especiallysincemassofHat omshasbeenneglectedin theanalysis,andtherefore,needtobeaccountedforbyanad ditionalchemicalkinetic productionrateterm.Deningtwoconstants C a and C r givenby C a = s s C 2 d M 0 and C r = k ratio C a ,allowssimplicationoftheequations.Asanexample,cons iderthe productiontermforH 2 O, 105

PAGE 106

_ corr: H 2 O = k S 2 f [ OH ] h C s [ i ] H i k S 2 r [ H 2 O ] h C s [ i ] i =( k S 2 f [ OH ] k S 2 r [ H 2 O ] k ratio ) s N A C d 2 2 3 M 0 =( k S 2 f C a k S 2 r C r ) 2 3 N A Thesecorrectiontermsfortheindividualspecieshavebeen summarizedinequations 3{78 through 3{84 .Althoughtheformalismoftheapproachforsurfacegrowthh asbeen presentedin[ 149 ],theHACAmechanismitselfwasmodiedrstinKazakovetal .[ 158 ], andtheninAppeletal.[ 148 ].Thecorrectiontermsandthekineticparametersforthe HACAreactionspresentedhere(includingthereactionrate sforthereversereactions) havebeentakenfromRevzanetal.[ 208 ]. corr: C 2 H 2 = W 1 C 2 H 2 2 N A (3{78) corr: O 2 = W 1 O 2 2 N A (3{79) corr: OH =( k S 2 r [ H 2 O ] C r k S 2 f [ OH ] C a ) 2 3 + W 1 OH N A (3{80) corr: H =(( k S 1 r [ H 2 ]+ k S 4 f [ C 2 H 2 ]) C r k S 1 f [ H ] C a k S 3 f [ H ] C r ) 2 3 (3{81) corr: H 2 =( k S 1 f [ H ] C a k S 1 r [ H 2 ] C r ) 2 3 N A (3{82) corr: H 2 O =( k S 2 f [ OH ] C a k S 2 r [ H 2 O ] C r ) 2 3 N A (3{83) corr: CO = W 1 O 2 + W 1 OH N A (3{84) 106

PAGE 107

3.1.6.1PAHcondensation Asub-groupofsurfacereactionsoccurringsimultaneously ,arethoseofPAHmolecules condensingonthesurfaceofsootparticles.Thereactionsc hemehasbeendepictedin gure 3-6 .Thecondensationprocesscanbewrittenintermsofsmoluch owskiequationfor particlecoagulation,givenbyequation 3{85 .Theequationsgiveninthissectionhavebeen takenfromFrenklachandWang[ 149 ],Balthasar[ 209 ]. dN 1 dt = 1 X j =1 i;j N 1 N j P (3{85) dN i dt = 1 2 i 1 X j =1 j;i j N j P N i j S 1 X j =1 i;j N i S N j P ; i =2 ; 3 ::: where N i P and N j S representthenumberdensitiesofPAHandsootparticles respectively.Theaboveequationcanfurtherbeexpressedi ntermsofmomentsasin equation 3{86 ,followingprocedureoutlinedinsection 3.1.4 .Thecollisionfrequency forsurfacecondensation( i;j sc )canbecalculatedfromequation 3{87 ,whilethemass andradiusofsootparticlesmaybecalculatedfromequation 3{88 andequation 3{89 respectively.ThereducedmassforthecollisionofPAHmole culesandsootparticleis calculatedfromequation 3{90 ,basedontheassumptionthat m i S >>m j P dM r dt = 1 2 1 X i =1 1 X j =1 r 1 X k =2 ( rC k ) m i r m j r k N i S N j P ; i =1 ; 2 ; 3 ::: (3{86) i;j sc =2 : 2 s k B T 2 i;j ( d i S + d j P ) 2 (3{87) wherethePAHmoleculediameterisgivenby d j P = d A q m j P 2 3 .Thewidthofan aromaticring( d A )isgivenby d A = p 3 L C C where L C C istheC-CbondlengthinPAH molecules,withavalueof1.39nm[ 149 ]. 107

PAGE 108

m i m C = 4 3 r i 3 (3{88) r 2 i = 3( im 1 )( m C ) 4 2 3 (3{89) i;j = 1 m i S + 1 m j P = 1 m j P (3{90) Newconstant C h and C s aredenedforconvenience,byequations 3{91 and 3{92 Substitutingtheresultsfromtheseequationsandequation 3{90 inequation 3{87 givesequation 3{93 .Substitutingtheexpressionfor i;j sc inthemomentequationgives equation 3{94 ,whichmayfurtherbesimpliedtoequation 3{95 C h = m C 1 3 (3{91) C s = d A r 2 3 (3{92) i;j sc =2 : 2 r k B T 2 m C m j P 1 2 h C h m i S 1 3 + C s m j P 1 2 i 2 =2 : 2 r k B T 2 m C h C s 2 m j P 1 2 m i S 2 3 +2 C s C h m j P 0 m i S 1 3 + C h 2 m j P 1 2 m i S 0 i (3{93) dM r sc dt = 1 2 1 X i =1 1 X j =1 r 1 X k =2 i;j sc ( rC k ) m i r m j r k N i S N j P (3{94) W r sc = dM r dt sc = 1 2 r 1 X k =2 i;j sc ( rC k ) m i r m j r k N i S N j P (3{95) = 1 2 r k B T 2 r 1 X k =2 ( rC k ) h C s 2 M r k 1 2 P M k + 2 3 S +2 C s C h M r k P M k + 1 3 S + C h 2 M r k + 1 2 P M k S i 108

PAGE 109

3.1.6.2Particlenucleation Previously,particlenucleationwasmodeledbyassumingco llisionbetweenPAH molecules(dimerization),largeraromaticsorbenzenepar ticles[ 148 { 150 ].Theinnite reactionsequencewasbrokendownintopolymerizationbloc ksconnectedviairreversible reactions[ 207 ].FrenklachandWang[ 150 ]describedthenucleationprocessusingmethod oflinearlumpingofthesequentialgrowthHACAreactionmec hanism.Thereaction setwasapproximatedasaunimolecularreactionsequenceof fusedringmolecules.The subsequentdimerizationoftheseringstructuresinfreemo lecularcoagulationprocesses resultsinfurthergrowthofthesemolecules.Post-dimeriz ationthree-dimensionalclusters weretreatedassootparticles.Theyalsodiscussedthatthe sizediscrepancybetween PAHmoleculesandthesootparticles,wasanindicationofth eimportanceofcoagulation processinparticlegrowthprocessandthatnucleationalon e,couldnotexplainsucha discrepancy.Moreover,theHACAsequentialgrowthmodelfo rnucleationfailedtopredict theconcentrationsofsootnucleiobservedintheexperimen ts[ 203 ].Subsequently,the coagulationoftwopyrenemoleculesindimerizationreacti onswasstudiedSchuetzand Frenklach[ 203 ],anditwasfoundthatcollisionofthesepyrenemoleculesi sstabilizedby amechanismofenergystorageininternalrotors,resulting insucientlylonglifetimes fornucleationtooccur[ 203 ].Therefore,thisstudyusestheapproachofdimerization ofpyrenemoleculestomodelnucleation[ 208 ].Thesmoluchowski'sequationforsucha coagulationprocessresultinginnucleationisgivenbyequ ation 3{96 dN i dt = 1 2 i 1 X j =1 i;i j N i P N i j P 1 X j =1 i;j N i P N j P (3{96) Note,fromgure 3-4 that,therstandsecondtermsofthesmoluchowski'sequati on representtheproductionandconsumptionofnuclei,respec tively.Thenuclei,onceformed, mayreactwithothersootparticlestoformlargersootparti cles.Theconsumptionterm ofthenucleusofsize i 0 ,signiesthisphenomenon.However,thisisessentiallyth egrowth process,andhasalreadybeenaccountedforincoagulationm odel.Hence,thesecondterm 109

PAGE 110

maybeneglectedforthenucleationprocess.Thesmoluchows ki'sequationfornucleation, reducestoequation 3{97 .Multiplyingthisequationwith m i P r givesequation 3{98 whichisthemomentequationfornucleation. dN nuc i dt = 1 2 1 X i =1 i 1 X j =1 i;i j N i P N i j P (3{97) R r = dM i dt = 1 2 1 X i =1 1 X j =1 ( m i + m j ) r i;j N i P N j P (3{98) Thecollisionfrequencyfornucleation i;j nuc isalsoafunctionof Kn .Forsmall particlessuchasPAHmolecules( Kn>> 1),theexpressionsforfree-molecularregime maybeapplied,withtheassumptionthattheybelongtothesa mesizeclass( i = j = i 0 ). Applyingtheseassumptionsgivestheexpressionfor i;j nuc giveninequation 3{99 .The constant K f maybecalculatedfrom K f = q 6 k B T 3 4 1 6 .Substitutingtheseexpressions intoequation 3{98 givesequation 3{100 andequation 3{101 forther th andthezero th momentratesfornucleation,respectively.Theequationsd escribedinthissectionmaybe befoundin[ 208 ]. i;j nuc = K f s 1 m i + 1 m j ( m i 1 3 + m j 1 3 ) 2 (3{99) = K f s 2 m i 0 (4 m i 0 2 3 )=4 p 2 K f m i 0 1 6 R r f = dM r nuc dt = 1 2 1 X i =1 1 X j =1 4 p 2 m i 1 6 ( m i + m j ) r N i P N j P (3{100) =2 p 2(2 m i 0 ) r + 1 6 N i 0 P 2 =(2 m i 0 ) r R 0 f =(2 m i 0 ) R r 1 f 110

PAGE 111

R 0 f =2 p 2(2 m i 0 ) 1 6 N i 0 P 2 (3{101) 3.1.7PredictionofSmokeNumber(SN) TheexpressionsforcorrelationofSmokeNumberandsootmas sdensity( m s = f v in mgm 3 )weretakenfromColketIIIetal.[ 141 ].Thevolumewasbasedonconversion toatmosphericpressureand298K[ 141 ].TheSNcanbeevaluatedfromequation 3{102 andequation 3{103 ,if m s isknown.Similarly, m s couldbecalculatedfromSNusing equation 3{104 andequation 3{105 SN = 1 : 8743 m 2s +12 : 117 m s ; m s < 2 : 5(3{102) SN = 12 : 513 m 0 : 4313 s ; m s > 2 : 5(3{103) m s =3 : 232 1 1 SN 19 : 58 0 : 5 # ; m s < 18 : 7(3{104) m s =0 : 002751 SN 2 : 319 ; m s > 18 : 7(3{105) 3.2CombustorCSTRNetworkModel 3.2.1Software Inordertomodelthedetailedgasphasekinetics,thesemicl osedcycle(SCC)gas turbinecombustorhasbeenmodeledusingaCSTRnetwork.The codewaswrittenusing opensourcesoftware Canterav.1.7 [ 210 ]usingitsC++interface,anditwascompiled using g++compilerv.4.2.3 ona Linux(Ubuntudistribution) machine.Themachine hardwarecongurationwasa DellDimensionC521systemwithaAMDAthlon(tm)64X2 DualCoreProcessor4400+,2GBRAM 111

PAGE 112

3.2.2Model AnetworkofCSTRs,asshowningure 3-8 ,wasusedtomodelthecombustor.The modelconsistedoftwoCSTRs,thecombustorandtherow-spli t.Fourreservoirs,namely, fuelinlet,airinlet,igniterandexhaust,formedthesourc esorsinksforthermo-ruid interactionswiththecombustorandrow-splitreactors.Th estatesofthereservoirswere time-invariantandhence,thesewereusedtospecifytheent rystatesoftheruidstreams andtheexitpressure.Themodelusedmassrowregulators(MF Rs)inordertosimulate constantmassrowconditionsfortheinletruidstreams,irr espectiveofthecombustor transients.Theexhaustgasfromthecombustorreactoraswe llastherow-splitreactor wasfeddownstreamthroughvalvesv 1 andv 2 ,respectively.Theexitmassrowrate throughthevalveswascontrolledbythedierentialpressu rebetweenthereactorand thedownstreamcondition.Itmaybenotedthat,unlessthepr essureiskeptconstant acrossthevalves,therowofEGRstreammaynotoccurduetoad versepressuregradient. Additionally,inapracticalsystem,theEGRstreammayunde rgoasequenceofprocesses suchasexpansion,recuperation,waterextraction,compre ssion,etc. 3.2.3GoverningEquations Forthegivenreactornetwork,thedetailedkineticsaresol vedinparallelwiththerow simulations.Theschematicfortheinputstothemodelisdep ictedinthegure 3-9 .Fora singlestreamtransientCSTRnetworkmodelwithonlyonestr eameach,forinletandexit, thegoverningconservationequations[ 210 ]maybewrittenas: SpeciesConservation: dm j dt CV =_ m in Y j;in m ex Y j;ex + M j V R j + A R s j (3{106) TotalMassConservation: d ( V R ) dt CV =_ m in m ex (3{107) EnergyConservation: 112

PAGE 113

dU dt CV =_ m in h in m ex h ex P dV R dt + A R Q int (3{108) Adiscussiononthegoverningequationsofasinglestreamtr ansientCSTRhasbeen presentedinTosun[ 211 ].However,thecanterasoftwareassumesamulti-streamtra nsient CSTRmodel.Thederivationofthegoverningequationsofthe transientCSTRmodelhas beendiscussedinsection A 3.2.4ModelInputs Foreachreservoirandreactor,a mechanismle hadtobespecied.These Cantera mechanism(CTI)les wereequivalenttothe Chemkin inputles.However,asingle Cantera mechanismlecontainsalltheinformationforthereaction mechanismand species,while Chemkin requiresindividuallesforthekinetic,thermodynamican d transportdata. Thestatevariables,namely,pressure,temperatureandcom positionwereneededasuser inputsforthereservoirstatesandreactorinitialstates. ThemassrowratesfortheMFRs andvalveK v valueswereprovideddeterminingthereactorpressureands toichiometry. Alinearvalveresponsetopressuredropwasused,asspecie dbydefaultinCantera combustioncode.ThevalueofK v wassethighenoughsothatthereactorpressure remainsconstant.Forthisstudy,avalueof0.5wasusedforK v forbothvalves. Theresidencetimewascontrolledbyscalingthetotalinlet massrowrate,which requiredanaccurateestimateofthehotgasdensity.Hence, thesimulationswererepeated untilagreementinpredictedandactualresidencetimesand densitieswasachieved. Theglobalvitiatedcombustionequationforbiofuelswasus edtocalculatemassrow ratesfortheinletreactantstreams.Thereactionfor < 1isgivenas: 113

PAGE 114

( ) C x H y O z + x + y 4 z 2 [ O 2 +(3 : 76) N 2 ]+ ( R n ) h ( )( x ) CO 2 +( ) y 2 H 2 O +(3 : 76) x + y 4 z 2 N 2 +(1 ) x + y 4 z 2 O 2 i (1+ R n ) h ( )( x ) CO 2 +( ) y 2 H 2 O +(3 : 76) x + y 4 z 2 N 2 +(1 ) x + y 4 z 2 O 2 i Theexhaustgasrecirculation(EGR)streammassrowrateand temperaturewerealso specied.Themassrowrateofthefuelinletwasgivenbyequa tion 3{109 .Anigniter reservoirwasaddedinordertoarticiallyigniteacombust iblefuel-airmixture.Ashort durationgaussianpulseofhightemperaturehydrogenatoms maybeusedforthepurpose. Themassrowrateofigniterinletisgivenbyequation 3{110 .However,forthisstudy, theignitermassrowratewassettozero.Themassrowrateofa irinletwasgivenby equation 3{111 m 1 =_ m F = M F n F (3{109) m 2 = Gaussian ( t 0 ;A I ; I )(3{110) m 3 =_ m A =4 : 76 M A x + y 4 z 2 n F = m 1 (3{111) Themass-basedrecirculationratio R m (orEGR)wasaspeciedparameterandwas denedastheratioofmassrowrateoftherecirculationstre am,tothatofthefreshair stream.Hence,themassrowrateofexhaustgasrecirculatio nstreamwascalculatedby equation 3{112 .Thetotalmassrowratewascalculatedbasedonequation 3{113 asthe sumofthemassrowratesoffuel,igniter,airandEGRstreams .Theignitermassrow rate,isnon-zeroforonlyashortdurationoftime,toinitia tecombustion.Theresidence timewasfoundusingequation 3{114 114

PAGE 115

_ m 4 =_ m R = R m m 3 (3{112) m T =_ m 1 +_ m 2 +_ m 3 +_ m 4 (3{113) res = V R m T (3{114) Thescalingofthemassrowratesinordertoachievetherequi redresidencetimewas achievedthroughthedeterminationof_ n F givenbyequation 3{115 ,basedonpredictedhot gasdensity .Thevalveexitmassrowrate_ m valve ,wascalculatedusingequation 3{116 n F = V R res M F +4 : 76 M A (1+ R m ) x + y 4 z 2 (3{115) m valve = K v P valve (3{116) 3.2.5PerformanceParameters Thecombustionperformanceparameterswerecalculatedbas edonequation 3{117 throughequation 3{119 .Theemissionindex(EI)ofaspeciesisdenedasthemass productionrateofthespeciesperunitmassrowrateoffuel[ 212 ],asgivenbyequation 3{117 assumingthat Y R;j = Y ex;j .Itmayfurtherbenotedthattheemissionindex,inourwork wasdenedbasedonthenetspeciesoutrow.Formostpollutan tspeciesofinterest,this dierencemaybeanon-issue,since Y F;j and Y A;j maybesafelyassumedtobezero. However,forfuelslikesyngasandacetylene Y F;j maynotbezeroforspecieslikeacetylene, H 2 andCO,butwehavedisregardedthese.Thedenitionofecie ncyasspeciedby equation 3{118 ,isbasedontheworkofZelinaetal.[ 109 ].Theloadingparameterisa measureofblowoutstabilityandwascalculatedbasedonequ ation 3{119 .Theconstant n hadavalueof2.Furtherdetailsabout LP maybefoundin[ 6 106 107 109 ]. 115

PAGE 116

EI j = m T Y ex;j m F Y F;j m A Y A;j m R Y R;j m F (3{117) = m F ( Y ex;j Y F;j )+_ m A ( Y ex;j Y A;j ) m F c =1 EI UHC +0 : 232 EI CO 1000 (3{118) LP = n T V R P n (3{119) 3.3ModelingApproach Theactualgasturbineemissions,eciencyandstabilityin volveacomplexinterplay ofseveralfactorslikephysicalpropertiesoffuel,suchas ignitiondelay,cetanenumber, viscosityandthecombustionsystemdesignitself.Thesema yfurtheraecttheatomization quality,evaporationrate,etc.andultimately,thecombus tionparametersofourinterest. Wehaveattemptedtolookatthecombustionperformancefrom thepointofview ofpurechemicalkinetics,isolatingtheinteractioneect softheparametersobserved inaphysicalsystem.Theprimaryzoneofthecombustorwasmo deledasaconstant pressureCSTR,withaconstantsupplyoftheinletchargefro mfuel,airandEGR streams,atpredeterminedratesandstates.Theairstreamt emperature(T A )andEGR temperature(EGRT)werealwayssetequal(unlessexplicitl yspecied),whilefuelstream temperature(T F )wasxedat300K.Themodelassumesinstantaneousmixingan d spatialhomogeneityofgasmixtureinsidetheCSTR.Thesimp licityofthemodelyields anintuitiveunderstandingoftheprocessesaectingcombu stionperformance.TheC++ interfaceofCantera[ 210 ]combustioncodewasusedformodeling.Thecombustion performanceofvariousbiofuelsnamely,ethanol,dimethyl etherandmethylbutanoate whichisabiodieselsurrogatehasbeencomparedagainsttha tofn-heptane,which isasurrogatefordiesel.Inaseparatestudy,thecombustio nperformanceofsyngas 116

PAGE 117

wascomparedtothatofmethane.Themechanismsusedforetha nol,dimethylether, methylbutanoateandn-heptaneweretakenfromtheLawrence LivermoreNational Laboratory(LLNL)website[ 213 ].Furtherstudiesonblowoutstabilitywereperformed forstoichiometriccombustionofethanolandvaryingEGRle velsandEGRtemperatures. Forthestudyofsootformation,acetylenefuelwaschosen,s inceitschemistryisfairly wellknown.ThemechanismdevelopedbyFrenklachandcowork erswasusedforsyngas, methaneandacetylenefuels.Table 3-1 summarizesthemechanismsusedtomodelthe combustionkineticsforeachofthefuels. Inthisstudy,combustionblowoutstabilityhasbeenassume dtoberesultingfrom kineticprocesses,andmeasuredintermsoftheloadingpara meter(LP)givenby equation 3{119 .Combustionstabilityisdenedasthemeasureofresistanc etoarame blowoutcondition,whichoccurswhentheheatreleaseratef romthereactionisinsucient toraisethetemperatureofthereactantstothesteadystate reactiontemperature.As aresult,asustainedhightemperaturerameisnotattainabl e,andtheCSTRexitstate eventuallybecomesidenticaltotheinletreactantstate.T hetermcombustionstability, asusedinthisworkshouldnotbeconfusedwiththatusedinth econtextofpressure ructuationsandunsteadyheatrelease,asexperiencedinan unsteadycombustionprocess. ThetotalinletmassrowrateataxedequivalenceratioandE GRwasincreaseduntilthe rametemperaturedroppedtothatofthemixedunreactedinle tstream,indicatingabsence ofareaction.Themaximummassrowrateatwhichasteadystat erameconditionwas achieved,wastreatedastheblowoutlimitforagivenequiva lenceratioandEGRlevel. Specicconstraintsappliedtoindividualcasestudieshav ebeendiscussedinsections 3.3.1 through 3.3.7 Theapplicabilityofthemodelisgenerallylimitedtoreaso nablylowpressures,since athigherpressures,theblowoutstabilitymaynotbekineti callycontrolledandsoot formationathigherpressuresmaybesignicantlyinruence dbyfractalaggregationeects. 117

PAGE 118

3.3.1CaseStudy1:SyngasandMethaneFuelSimulations Thecombustorpressureandvolumewerexedat202.65kPa(2. 0atm)and250e-6 m 3 (0.25Liter),respectively.Thecombustorwasinitialized toatemperatureof2000 Kandthecompositionwassettothemajorproductspeciesofc ombustion,tosimulate asteadyramewithatemperatureabovetheauto-ignitiontem peratureoftheinlet reactants.Theemissions(atresidencetimeof5.0ms)andbl owoutlimitsforthree dierentmixturesofsyngaswerecompared,namely,Syn75(C O:0.25,H 2 :0.75),Syn50 (CO:0.5,H 2 :0.5)andSyn25(CO:0.75,H 2 :0.25).Simulationswereperformedforeach ofthesefuelcompositionsovertheequivalenceratiosrang ingfrom0.4to3.0,andthree dierentlevelsofEGRwithadiabaticrecirculationatreci rculationratiosof0.0,1.0 and2.0.Inasubsequentpartofthiswork,threelevelsofnon -adiabaticEGRwitha recirculationtemperatureof1200Kweretested.Theairstr eamtemperature(T A )was maintainedconstantat400K.Theconstantresidencetimesi mulationswererunfor5.0 ms.3.3.2CaseStudy2:Biofuelsandn-HeptaneFuelSimulations Thecombustionperformanceofvariousbiofuelsnamely,eth anol,dimethylether andmethylbutanoatewhichisabiodieselsurrogatehasbeen comparedagainstthatof n-heptane,whichisasurrogatefordiesel.Intherstsetof runs,theblowoutperformance ofthefuelshasbeeninvestigated,forequivalenceratiosv aryingfrom0.4to1.0atintervals of0.1,andEGRlevelsof0.0,1.0and2.0.TheEGRstreamwasco oledtoatemperature of700K,andthecombustorpressureandvolumewerexedat20 2.65kPa(2.0atm)and 0.00258m 3 (15.71in 3 ),respectively.Thecombustorstatewasinitializedtoaco nstant CSTRpressureof202.65Pa(2atm),atemperatureof2000Kand thecomposition wassettothemajorproductspeciesofcombustion,tosimula teasteadyramewitha temperatureabovetheauto-ignitiontemperatureoftheinl etreactants.Forsimulationsat xedresidencetime,thevalueof10.0mswasused. 118

PAGE 119

3.3.3CaseStudy3:StudyofEGRTemperature/EGRlevelonEth anol BlowoutLimits Theblowoutperformanceofethanolfuelwastestedforequiv alenceratioof1.0and EGRtemperaturesrangingfrom400to1200Katintervalsof20 0K.Comparisonof theserunshasbeenpresentedforEGRlevelsof0.0,0.5,1.0, 1.5and2.0.Thecombustor pressureandvolumewereheldxedat202.65kPa(2atm)and0. 00258m 3 (15.71in 3 ), respectively.Theblowoutstabilitywasquantiedinterms ofLPgivenbyequation 3{119 TheeectofreductioninloadingparameterasafunctionofE GRandEGRtemperature havebeenplotted.Theresultswerenormalizedtothecorres pondingvaluesfortheopen cycle(EGR=0.0)case.Theeectofpresssurewasalsoinvest igated,byvaryingthe pressuresfrom0.5to2.0atm,instepsof0.5atm. Inasubsequentstudy,ignitiondelayswerecalculatedfors toichiometriccombustion ofethanolinaconstantpressurebatchreactor,atvaryingE GRlevelsandreactantinlet temperatures.Theignitiondelaywasfurtherusedforestim ationoftheignitionnumber. Further,correlationsofvariationsofLPwithOHemissioni ndiceshavebeenpresented. 3.3.4CaseStudy4:AStudyoftheEectofEGRTemperature/EG R level/ResidenceTime/EquivalenceRatioonSootFormation (AcetyleneFuelCombustion) Thesootformationprocesswasmodeledusingdetailedchemi calkineticmechanism forsootformationbyFrenklachandcoworkers(Appeletal.[ 148 ]).Forthisstudy, particulategrowthmodelforsootformationwasomittedand onlydetailedkineticmodel wasusedtostudytheprocessesinvolvedingasphasechemist ryleadingtosootnucleation. TheeectofEGR,EGRtemperature,equivalenceratioandres idencetimeonsoot formationwasinvestigated.Thecombustorpressureandvol umewereheldxedat202.65 kPa(2atm)and0.00258m 3 (15.71in 3 ),respectively.Thefollowingcaseswereanalyzed: 1.case1:EGR=0.0-5.0, =2 : 0, res =1 : 0,EGRT=1000K 2.case2:EGRT=400-1200K,EGR=0.0and2.0, =2 : 0, res =1 : 0 3.case3: =0.5-5.0,EGR=0.0and2.0, res =1 : 0,EGRT=1000K 119

PAGE 120

4.case4: res =0.1-10.0, =2 : 0,EGR=0.0and2.0,EGRT=1000K 3.3.5CaseStudy5:SootParticleGrowthandOxidationinVit iated CombustionofAcetyleneFuel ThesootformationdynamicswassimulatedinaCSTRnetworkm odel,withmodications totheCanteraCSTRmodeltoincludethekineticcouplingofh eterogeneoussoot growthprocesses,usingthemethodofmoments.Themethodof momentsapproachfor particulatesootmodeling,andthedetailedchemistryofso otformationused,wasbased ontheworkbyFrenklachandcoworkers[ 148 { 150 158 171 187 { 190 192 { 194 196 { 201 ]. Thedetailedchemicalkineticmechanismisbasedon[ 148 ].Thefortrancodewiththe comprehensiveparticulatemodelused[ 149 208 ]includedsubmodelsforsurfacechemistry [ 148 ],nucleation[ 150 ],andcoagulation[ 149 188 189 ].Theconservationequationsin Canterazero-dimensionalreactorweremodiedtoincludet hecorrectiontermsfromthe sootcode.AninterfacingclasswritteninC++,allowedsimu ltaneouscalculationsofthe sootaerosoldynamicsproblemfromthefortransootcode(us inganexterncall)andthe nativeC++zero-dimensionalreactorclass.Themodelvalid ationwasperformedagainst theresultsfromBrownetal.[ 159 ]usingdetailedchemistryfromWangandFrenklach [ 195 ],particulatemodelfromRevzanetal.[ 208 ]andsurfacechemistrysubmodelfrom Kazakovetal.[ 158 ].Inordertoobtainanunderstandingoftheuncertaintiesi nthe predictions,theresultsfromthismodelwerecomparedtoth esootformationmodel developedbyRichteretal.[ 183 ],Ergutetal.[ 184 ],Richteretal.[ 214 ].Theirmodel usedthesectioningapproach,inwhichthechemistryof20di scretesootparticlebins wasmodeled.Thebinswithasux\J",indicatedradicalbins .Therstfewbinswere treatedassootprecursors,whilebinsbeyondbin4weretrea tedassootparticles.The advantageofthesectioningapproachwasthatthebinscould betreatedasindividual species,andthechemistrycouldbeintegratedwiththegasp hasemechanism.The thermodynamicparametersforthebinswerenotspecied;in steadtheforwardandreverse reactionswereindependentlyspecied. 120

PAGE 121

Theconditionsofsimulationwere =2.0,P=202.65kPa(2atm),EGR=0.0and 2.0.Thesimulationswererunataresidencetime( res )of1.0mswithspeciedrame temperaturesbetween1500-2000K.Thesootvolumefraction ,averagesootdiameter, sootsurfacearea,sootparticlemassdensity,andsmokenum berweremodeled.The simulationswereperformedundersimilarconditionsforRi chtersootmechanism.Thebin andtheradicalbinmassfractions(beyondbin4)wereusedto determinethesootvolume fractions.Thesmokenumber(SN)wasmodeledusingcorrelat ionsdevelopedbyColketIII etal.[ 141 ]. 3.3.6CaseStudy6:ModelingofSemi-ClosedCycleEngineCha racteristics usingaKineticallyEquivalentUnvitiatedOpen-CycleCSTR System Thefocusofthisstudywastodemonstratetheuseofanapproa chtomodela semi-closedcycle(SCC)engineusingakineticallyequival entunvitiatedopencycle (OC)model.Thisapproachhastheadvantageofbeingabletor eproducecharacteristic curvesofamorecomplexsystemusingasimplermodel,andall owstheuseofavailable dataintheliteratureonOCenginesforequivalentSCCengin es.Thisapproachisof signicanceinreducingtheneedforfurtherexperimentati ononsemiclosedcycles.The steadystateCSTRconditionsforbothOCandSCCcaseswereco mpared,todetermine the\matchingconditions"forthetwomodels.Thetemperatu resuppressioneectof vitiationinSCCCSTR,wasmodeledbyanequivalentlowerinl etairtemperature(cases AandB)enteringanadiabaticOCCSTR,oraheatlossterm(cas esCandD)ina non-adiabaticCSTR.Theresidencetimeeecthasbeenmodel edusingappropriate massrowscalingcalculations.Asetofnumericalsimulatio nswiththeequivalentCSTR networkmodelswereperformed,tovalidatetheresultsofth etheoreticalanalysis. ThematchingconditionsforakineticallyequivalentOC,ar ebasedonamatchinthe exitenthalpiesandspeciesproductionrates.Fourequival entsystemswereanalyzed: CaseA: AnadiabaticOCsystemequivalenttoanSCCsystem,withlowe rreactantinlet temperaturesandlargerresidencetimewithmatchedfreshr eactantrows. 121

PAGE 122

CaseB: AnadiabaticOCsystemequivalenttoanSCCsystem,withlowe rreactantinlet temperaturesandmatchedresidencetime. CaseC: Anon-adiabaticOCsystemequivalenttoanSCCsystem,witha positiveheat lossandlargerresidencetimewithmatchedfreshreactantr ows. CaseD: Anon-adiabaticOCsystemequivalenttoanSCCsystem,witha positiveheat lossandmatchedresidencetime. a ) O pen C yc l e C S T R b ) S e m i C l o s ed C yc l e C S T R h A h F hi hi A hi hi h A mA hi h F mF B hi hi h A mA hi h F mF C Figure3-7.EquivalentSCCsystems.(A)OCandSCC(withnonadiabaticEGR)CSTR systems,(B)EquivalentSCCsystemwithadiabaticEGRandre ducedairinlet temperature,(C)Equivalentnon-adiabaticOCsystem. Inthepreliminarytestsimulations,itwasfoundthatthemo delingbasedonequivalent lowerinletairtemperature,wasseverelyconstrainedbyth elowestairtemperature(300 K)thatcouldreliablybeused.Ontheotherhand,theuseofah eatruxtermtomodel 122

PAGE 123

rametemperaturesuppression,hadnosuchconstraint.Them assrowscalingbasedon matchingtheresidencetimeoffreshreactants(casesAandC insection 4.7.3 )wasfound tobelesscomplicatedanapproachthanthematchingofthere sidencetimebasedon thetotalmassrow.Therefore,theuseofanequivalentOCsys temforpredictingthe chemicalkinetictrendsofaSCCsystemwasdemonstrated,ba sedontheapproachin caseC.Simulationswererunfor =1.0,and res =10ms(forSCCsystem),EGRT =600,800,1000and1200K,EGR=0.0and2.0.ForanSCCsystemw iththese conditions,anequivalentOCsystemwasmodeledusingthema ssrowscalingbased oncase3section 4.7.3 .Anappropriateheatlosstermwascalculatedtomatchtheex it temperaturesfortheSCCandtheOCsystems.Theplotsoframe temperatureand emissionindicesforunburnedhydrocarbons(UHC),COandOH wereplotted. 3.3.7CaseStudy7:ASensitivityAnalysisStudyonCombusti on PerformanceofEthanolFuelClosetoBlowout Asensitivityanalysisstudywascarriedoutforethanolfue lclosetoblowout.The simulationconditionswere =1.0,EGRT=1000K,andEGR=0.0,2.0.Therelative andabsolutetolerancesforthesensitivitysolutionwere0 .005.Thekeyassumptionbehind thisanalysisisthatthesensitivityoftheblowoutlimitis astrongfunctionoftheOH concentrationsensitivity.Themostinruentialreactions aectingtheOHconcentrations wereidentied. Figure3-8.Semi-ClosedCycleCSTRNetworkModel 123

PAGE 124

i R ea ct i on s j S p ec i es a ij S j b ij S j k f, i k r i K P i ( A b, E ) K i n et i c D at a k( T ) = A T b e x p (E / R T ) K P i = k r i / k f, i S p ec i es D at a R eact i on D at a Th e r m od yn am i c D at a T r ansp o r t D at a C p i h i S i C S T R M O D E L V i sc o c i t y D i ff usi on coef f i c i e nt s, c o nduct i v i t y [][]' j nj = ij b j r i j nj = ij a j fi iS k S k = r1 1 jdSj dtdcj dti niijri I npu t S t a t e E x i t S t a t eN o t U sed Figure3-9.ModelInputsTable3-1.Combustionmechanisms FuelSpeciesReactionsRef. Ethanol57383Marinov[ 136 ] DME79660Fischeretal.[ 215 ] Curranetal.[ 216 ] Kaiseretal.[ 217 ] MB2641966Fisheretal.[ 218 ] n-Heptane(reduced)1601540Seiseretal.[ 219 ] Methane101544Appeletal.[ 148 220 ] Syngas101544Appeletal.[ 148 220 ] Acetylene101544Appeletal.[ 148 220 ] Acetylene2966663Richteretal.[ 183 214 ],Ergutetal.[ 184 ] 124

PAGE 125

CHAPTER4 RESULTSANDDISCUSSION 4.1CaseStudy1:SyngasandMethaneFuelSimulations 4.1.1EectofFuelCompositionandEquivalenceRatioatEGR =0 Theimpactofvariabilityinfuelcompositiononcombustion performancewasstudied forarangeofequivalenceratiosfrom0.4to3.0.Figure 4-1 depictsthevariationinsteady statecontinuouslystirredtankreactor(CSTR)temperatur easafunctionofequivalence ratioforMethaneandthreecompositionsofsyngastested,S yn75(CO:0.25,H 2 :0.75), Syn50(CO:0.5,H 2 :0.5)andSyn25(CO:0.75,H 2 :0.25).Themaximumtemperaturewas foundtooccurunderfuel-richconditionsforsyngasmixtur esandwasnearstoichiometric conditionsforMethane,fortherangeofconditionstested. Thecombustionofsyngas mixturesproducedhigherrametemperaturesascomparedwit hMethanefuel.Hence, NO x productionratescouldbehigherforsyngas.However,opera tingthesefuelsina semiclosedcycle(SCC),whichhastheeectofloweringrame temperatures,islikelyto considerablymitigatethiseect. Thecombustioneciencyandloadingparameterforthesyste m,werecalculatedas describedinZelinaandBallal[ 109 ].Thecombustioneciencywasfoundtodropsteadily forsyngasmixtureswithrisingequivalenceratios,asshow ninFigure 4-2 .TheSyn25 mixturehadgenerallyhigherecienciesovermostequivale nceratiosascomparedtothe othertwomixtures,Syn75beingthelowestamongstsyngasmi xtures.Thecombustion eciencyofMethanewassignicantlylower,especiallyont hefuel-richside. TheemissionindexforCOincreasedbyaboutafactorof2betw eenequivalenceratios of0.4to3.0,ascanbeseeninFigure 4-3 .Underleanconditions,Methaneproduced highestemissionindex(EI)forCOandforsyngasmixtures,S yn25wasfoundto producethemaximumEICO,followedbySyn50andSyn75,Syn75 beingthelowest. Atequivalenceratiosbetween1to2,COemissionsforMethan ewereconsiderablyhigher thansyngas.TheplotsforEIUHCversusequivalenceratioar eshowninFigure 4-4 125

PAGE 126

Syn75syngasmixturehadUHCemissionscomparabletoMethan eonthefuel-leanside, whilemuchlowerEIUHConthefuel-richside.TheUHCemissio nsforSyn25werefound tobelowest,whileSyn75hadhighestEIUHCnexttoMethane.T heUHCemissionsfor Methanewereaboutanorderofmagnitudehigherthananyofth esyngasmixturesonthe fuel-richside. Figure 4-5 depictstheEIUHCversuscombustioneciencytrends.TheUH C emissionschangedbyaboutanorderofmagnitudewithachang eineciencyfromover 0.99tobelow0.70,dependingonthefuel.Theplotshowsthat UHCemissionsmay decreasedespiteareductionineciency,indicatingaposs ibleincreaseinCOemissions correspondingtotheeciencyloss. Syn25 Syn50 Syn75 Methane EGR=0.0, res =5.0ms,P=202.65kPa(2atm) T A =400K EquivalenceRatio( )Temperature[K]3.0 2.5 2.0 1.5 1.0 0.5 0.0 2400230022002100200019001800170016001500 Figure4-1.ComparisonoframetemperaturesforfuelsMetha ne,Syn75,Syn50andSyn25 asafunctionofequivalenceratioatEGR=0.0, res =5.0ms,P=202.65 kPa(2atm),T A =400K. Thedependenceofcombustioneciencyonsteadystatecombu stionrametemperatures hasbeenpresentedinFigure 4-6 .Thelean-sidedataresultedinhigherecienciesas expected.TheecienciesweremaximumforSyn25,followedb ySyn50andSyn75,inthat order.Adrasticfall-oofeciencywasobservedforMethan emixture,whiletheeciency ofallsyngasmixtureswashigherthan0.7,forequivalencer atiosbetween0.4and3. 126

PAGE 127

Syn25 Syn50 Syn75 Methane EGR=0.0, res =5.0ms T A =400K,P=202.65kPa(2atm) Equivalenceratio( )Combustioneciency( c )3.0 2.5 2.0 1.5 1.0 0.5 0.0 1.000.900.800.700.600.500.40 Figure4-2.ComparisonofcombustioneciencyforfuelsMet hane,Syn75,Syn50and Syn25asafunctionofequivalenceratioatEGR=0.0, res =5.0ms,P= 202.65kPa(2atm),T A =400K. Syn25 Syn50 Syn75 Methane EGR=0.0, res =5.0ms,P=202.65kPa(2atm) T A =400K EquivalenceRatio( )EICO[g/kg]3.0 2.5 2.0 1.5 1.0 0.5 0.0 1000 100 10 Figure4-3.ComparisonofCOemissionindexforfuelsMethan e,Syn75,Syn50andSyn25 asafunctionofequivalenceratioatEGR=0.0, res =5.0ms,P=202.65 kPa(2atm),T A =400K. 127

PAGE 128

Syn25 Syn50 Syn75 Methane EGR=0.0, res =5.0ms,P=202.65kPa(2atm) T A =400K EquivalenceRatio( )EIUHC[g/kg]3.0 2.5 2.0 1.5 1.0 0.5 0.0 1e+031e+02 10 1 Figure4-4.ComparisonofUHCemissionindexforfuelsMetha ne,Syn75,Syn50and Syn25asafunctionofequivalenceratioatEGR=0.0, res =5.0ms,P= 202.65kPa(2atm),T A =400K. Syn25 Syn50 Syn75 Methane EGR=0.0, res =5.0ms,P=202.65kPa(2atm) T A =400K Combustioneciency( c )EIUHC[g/kg]1.00 0.90 0.80 0.70 0.60 0.50 0.40 1e+031e+02 10 1 Figure4-5.VariationofUHCemissionindexforfuelsMethan e,Syn75,Syn50andSyn25 asafunctionofcombustioneciencyatEGR=0.0, res =5.0ms,P= 202.65kPa(2atm), =0.4,0.5,0.6,0.7,0.8,0.9,1.0(fuel-leanside)and =1.2,1.6,2.0,3.0(fuel-richside),T A =400K. 128

PAGE 129

Syn25 Syn50 Syn75 Methane T A =400K,EGR=0.0, res =5.0ms P=202.65kPa(2atm) FlameTemperature[K]Combustioneciency( c )2400 2300 2200 2100 2000 1900 1800 1700 1600 1500 1.000.900.800.700.600.500.40 Figure4-6.VariationofcombustioneciencyforfuelsMeth ane,Syn75,Syn50andSyn25 asafunctionoframetemperatureatEGR=0.0, res =5.0ms,P=202.65 kPa(2atm), =0.4,0.6,0.8,1.0(fuel-leanside)and =1.2,1.6,2.0,3.0 (fuel-richside),T A =400K. 129

PAGE 130

4.1.2EectofAdiabaticExhaustGasRecirculation(EGR) TheeectofEGRunderconditionsofadiabaticrecirculatio nonthecombustion eciencywasstudied.Undertheseconditions,theinletrec irculationstream(Stream 4inFigure 3-8 )intothecombustorhasnearlythesametemperatureastheco mbustor exittemperature.Theonlydierenceintemperatureswould arisefromthefactthat thecombustorexhaustwasstillallowedtoreactintheFlowS plitreactorforasmall niteresidencetime.Figure 4-7 depictsthetemperature-eciencydependenceforsyngas mixturesforEGRof0.0and2.0,respectively.Themodelresu ltshavebeenplottedat equivalenceratiosof0.4,0.6,0.8,1.0,1.2,1.6,2.0and3. 0.Therametemperaturesand eciencyincreasedwithincreasingEGRto2.0.Thereisasig nicantdropinrich-side eciencyofallsyngasmixturesatEGRof2.0.Syn3hadthelow esteciencyonthe rich-sidewhileSyn2hadthehighest,atalmostallequivale nceratios,fortheEGRlevels tested.Moreover,theeectofvariationsincompositionso fthemixtureswasmore pronouncedonthefuel-richsideofthetemperature-ecien cyloop. Syn25 Syn50 Syn75 EGR=0.0, res =5.0ms T A =400K,P=202.65kPa(2atm) FlameTemperature[K]Combustioneciency( c )2400 2300 2200 2100 2000 1900 1800 1700 1600 1500 1.000.950.900.850.800.750.70 Syn25 Syn50 Syn75 EGR=2.0,Adiabatic, res =5.0ms T A =400K,P=202.65kPa(2atm) FlameTemperature[K]Combustioneciency( c )2500 2400 2300 2200 2100 2000 1900 1800 1700 1600 1500 1.000.950.900.850.800.750.70 (a)EGR=0.0(b)EGR=2.0 Figure4-7.VariationofcombustioneciencyforfuelsSyn7 5,Syn50andSyn25asa functionoframetemperatureatEGR=0.0and2.0, res =5.0ms,P= 202.65kPa(2atm),T A =400K. TheeectofadiabaticEGRonthecombustionperformanceoft hethreefuelswas testedforthreedierentequivalenceratios.Figures 4-8 to 4-11 depicttheeectofEGR at =0 : 4.Figures 4-12 to 4-15 wereplottedfor =1 : 0whileFigures 4-16 to 4-19 130

PAGE 131

correspondto =1 : 6.Figure 4-8 showstheeectofadiabaticEGRontemperature.The combustorrametemperaturesincreasedwithincreaseinadi abaticEGR.Theeectof compositionvariancesontemperaturewasfoundtoincrease withincreasinglevelsofEGR at =0 : 4.TheeectofEGRoncombustioneciencyisshownin 4-9 .At =0 : 4,the combustionecienciesforSyn25werefoundtobemaximumove ralllevelsofEGRtested, whilethoseforSyn75werelowest.WithrisinglevelsofEGR, thedierenceincombustion ecienciesofthethreecompositionswasalsofoundtoincre ase,whileallthemixtures showedageneralupwardtrendineciencyat =0 : 4. Syn25 Syn50 Syn75 =0.4, res =5.0ms,P=202.65kPa(2atm), T A =400K EGRFlametemperature[K]2.0 1.5 1.0 0.5 0.0 15701560155015401530152015101500 Figure4-8.VariationoframetemperatureforfuelsSyn75,S yn50andSyn25asafunction ofadiabaticEGRat =0.4, res =5.0ms,P=202.65kPa(2atm),T A = 400K. TheeectofEGR(at =0 : 4)onEIUHCandEICO,canbeseeninFigures 4-10 and 4-11 ,respectively.TheEIUHCwasmaximumforSyn75whileEICOwa smaximum forSun25forallEGR.TheUHCandCOemissionswererelativel ylowforallmixtures anddecreasedwithEGR.At =1 : 0,therametemperaturesandcombustioneciency increased,whileemissionindicesofCOandUHCdecreasedwi thadiabaticEGR.The Syn75mixturehadthemaximumtemperaturewhileSyn25hadth elowest.Ascanbeseen inFigures 4-12 to 4-15 ,theeectofcompositionvariabilityonCOemissions,temp erature andeciencyreducedslightlyat =1 : 0,withincreasingEGR.At =1 : 6,wider dierencesinparameterscouldbeobserved,asdepictedbyF igures 4-16 to 4-19 .All 131

PAGE 132

mixturesshowedanincreaseintemperaturewithadiabaticE GR,thatforSyn25being maximum.WithincreasingEGR,allmixturesshowedsmallinc reasesineciencyanda slightreductioninEICOandUHC. Syn25 Syn50 Syn75 =0.4, res =5.0ms,P=202.65kPa(2atm), T A =400K EGRCombustioneciency( c )2.0 1.5 1.0 0.5 0.0 0.9900.9880.9860.9840.9820.9800.9780.9760.9740.9720.970 Figure4-9.VariationofcombustioneciencyforfuelsSyn7 5,Syn50andSyn25asa functionofadiabaticEGRat =0.4, res =5.0ms,P=202.65kPa(2atm), T A =400K. Syn25 Syn50 Syn75 =0.4, res =5.0ms,P=202.65kPa(2atm), T A =400K EGREIUHC[g/kg]2.0 1.5 1.0 0.5 0.0 1e+02 10 1 Figure4-10.VariationofUHCemissionindexforfuelsSyn75 ,Syn50andSyn25asa functionofadiabaticEGRat =0.4, res =5.0ms,P=202.65kPa(2 atm),T A =400K. 132

PAGE 133

Syn25 Syn50 Syn75 =0.4, res =5.0ms,P=202.65kPa(2atm) T A =400K EGREICO[g/kg]2.0 1.5 1.0 0.5 0.0 40353025201510 Figure4-11.VariationofCOemissionindexforfuelsSyn75, Syn50andSyn25asa functionofadiabaticEGRat =0.4, res =5.0ms,P=202.65kPa(2 atm),T A =400K. Syn25 Syn50 Syn75 =1.0, res =5.0ms,P=202.65kPa(2atm) T A =400K EGRFlametemperature[K]2.0 1.5 1.0 0.5 0.0 2420240023802360234023202300 Figure4-12.VariationoframetemperatureforfuelsSyn75, Syn50andSyn25asa functionofadiabaticEGRat =1.0, res =5.0ms,P=202.65kPa(2 atm),T A =400K. 133

PAGE 134

Syn25 Syn50 Syn75 =1.0, res =5.0ms,P=202.65kPa(2atm), T A =400K EGRCombustioneciency( c )2.0 1.5 1.0 0.5 0.0 0.9500.9400.9300.9200.9100.9000.8900.880 Figure4-13.VariationofcombustioneciencyforfuelsSyn 75,Syn50andSyn25asa functionofadiabaticEGRat =1.0, res =5.0ms,P=202.65kPa(2 atm),T A =400K. Syn25 Syn50 Syn75 =1.0, res =5.0ms,P=202.65kPa(2atm), T A =400K EGREIUHC[g/kg]2.0 1.5 1.0 0.5 0.0 70605040302010 Figure4-14.VariationofUHCemissionindexforfuelsSyn75 ,Syn50andSyn25asa functionofadiabaticEGRat =1.0, res =5.0ms,P=202.65kPa(2 atm),T A =400K. 134

PAGE 135

Syn25 Syn50 Syn75 =1.0, res =5.0ms,P=202.65kPa(2atm), T A =400K EGREICO[g/kg]2.0 1.5 1.0 0.5 0.0 190185180175170165160155150 Figure4-15.VariationofCOemissionindexforfuelsSyn75, Syn50andSyn25asa functionofadiabaticEGRat =1.0, res =5.0ms,P=202.65kPa(2 atm),T A =400K. Syn25 Syn50 Syn75 =1.6, res =5.0ms,P=202.65kPa(2atm) T A =400K EGRFlametemperature[K]2.0 1.5 1.0 0.5 0.0 230022902280227022602250224022302220 Figure4-16.VariationoframetemperatureforfuelsSyn75, Syn50andSyn25asa functionofadiabaticEGRat =1.6, res =5.0ms,P=202.65kPa(2 atm),T A =400K. 135

PAGE 136

Syn25 Syn50 Syn75 =1.6, res =5.0ms P=202.65kPa(2atm) T A =400K EGRCombustioneciency( c )2.0 1.5 1.0 0.5 0.0 0.8900.8800.8700.8600.8500.8400.8300.8200.8100.800 Figure4-17.VariationoframetemperatureforfuelsSyn75, Syn50andSyn25asa functionofadiabaticEGRat =1.6, res =5.0ms,P=202.65kPa(2 atm),T A =400K. Syn25 Syn50 Syn75 =1.6, res =5.0ms P=202.65kPa(2atm)T A =400K EGREIUHC[g/kg]2.0 1.5 1.0 0.5 0.0 70605040302010 0 Figure4-18.VariationofUHCemissionindexforfuelsSyn75 ,Syn50andSyn25asa functionofadiabaticEGRat =1.6, res =5.0ms,P=202.65kPa(2 atm),T A =400K. 136

PAGE 137

Syn25 Syn50 Syn75 =1.6, res =5.0ms,P=202.65kPa(2atm), T A =400K EGREICO[g/kg]2.0 1.5 1.0 0.5 0.0 560540520500480460440 Figure4-19.VariationofCOemissionindexforfuelsSyn75, Syn50andSyn25asa functionofadiabaticEGRat =1.6, res =5.0ms,P=202.65kPa(2 atm),T A =400K. 137

PAGE 138

4.1.3EectofNon-AdiabaticExhaustGasRecirculation Attherowsplitreactor,awallwithaspeciedoverallheatt ransfercoecientwas addedtocontroltheEGRinlettemperatureforStream4toasp eciedvalueof1200 K.ThetargetEGRinlettemperaturesforStream4wereachiev edbysettingarbitrarily highheatruxvaluesbyassigningveryhighvaluesoftheover allheattransfercoecient, wallsurfaceareaandthetemperatureofthecold-sideruidt o1200K.Theeectof non-adiabaticEGRwithaninlettemperatureof1200Khasbee nstudied,andthe resultshavebeenpresentedingures 4-20 through 4-26 forSyn25fuel.Clearly,ascan beobservedfromgure 4-20 ,withincreasingEGR,therewasadropincombustion temperatureatallequivalenceratios.Figure 4-21 showsaplotofcombustioneciency versusequivalenceratio.IncreasingEGRreducedCOandUHC emissionindicesslightly whileincreasedcombustioneciency,asshowninFigures 4-22 and 4-23 .Dependingon theEGRlevel,aCOemissionsminimumemergesatanequivalen ceratioofabout0.6. EGR=0.5 EGR=0.2 EGR=0.0 P=202.65kPa(2atm)Syn25,EGRT=1200K, res =5.0ms T A =400K EquivalenceRatio( )Flametemperature[K]3.0 2.5 2.0 1.5 1.0 0.5 0.0 24002300220021002000190018001700160015001400 Figure4-20.ComparisonoframetemperatureforSyn25fuelw ithEGRof0.0,0.2and0.5 asafunctionofequivalenceratioatEGRT=1200K, res =5.0ms,P= 202.65kPa(2atm),T A =400K. 138

PAGE 139

EGR=0.5 EGR=0.2 EGR=0.0 T A =400K P=202.65kPa(2atm)Syn25,EGRT=1200K, res =5.0ms Equivalenceratio( )Combustioneciency( c )3.0 2.5 2.0 1.5 1.0 0.5 0.0 1.000.980.960.940.920.900.880.860.840.82 Figure4-21.ComparisonofcombustioneciencyforSyn25fu elwithEGRof0.0,0.2and 0.5asafunctionofequivalenceratioatEGRT=1200K, res =5.0ms,P= 202.65kPa(2atm),T A =400K. EGR=0.5 EGR=0.2 EGR=0.0 P=202.65kPa(2atm), T A =400K Syn25,EGRT=1200K, res =5.0ms Equivalenceratio( )EICO[g/kg]3.0 2.5 2.0 1.5 1.0 0.5 0.0 1e+031e+02 10 Figure4-22.ComparisonofCOemissionindexforSyn25fuelw ithEGRof0.0,0.2and 0.5asafunctionofequivalenceratioatEGRT=1200K, res =5.0ms,P= 202.65kPa(2atm),T A =400K. 139

PAGE 140

EGR=0.5 EGR=0.2 EGR=0.0 P=202.65kPa(2atm), T A =400K Syn25,EGRT=1200K, res =5.0ms Equivalenceratio( )EIUHC[g/kg]3.0 2.5 2.0 1.5 1.0 0.5 0.0 1e+02 10 1 Figure4-23.ComparisonofUHCemissionindexforSyn25fuel withEGRof0.0,0.2and 0.5asafunctionofequivalenceratioatEGRT=1200K, res =5.0ms,P= 202.65kPa(2atm),T A =400K. EGR=0.5 EGR=0.2 EGR=0.0 P=202.65kPa(2atm), T A =400K Syn25,EGRT=1200K, res =5.0ms Combustioneciency( c )EIUHC[g/kg]1.00 0.98 0.96 0.94 0.92 0.90 0.88 0.86 0.84 0.82 1e+02 10 1 Figure4-24.ComparisonofUHCemissionindexforSyn25fuel withEGRof0.0,0.2and 0.5asafunctionofcombustioneciencyatEGRT=1200K, res =5.0ms, P=202.65kPa(2atm),T A =400K, =0.4,0.6,0.8,1.0(fuel-leanside) and =1.2,1.6,2.0,3.0(fuel-richside). 140

PAGE 141

EGR=0.5 EGR=0.2 EGR=0.0 T A =400K EGRT=1200K, res =5.0ms Syn25,P=202.65kPa(2atm), Flametemperature[K]Combustioneciency( c )2400 2300 2200 2100 2000 1900 1800 1700 1600 1500 1400 1.000.980.960.940.920.900.880.860.840.82 Figure4-25.ComparisonofcombustioneciencyforSyn25fu elwithEGRof0.0,0.2and 0.5asafunctionoframetemperatureatEGRT=1200K, res =5.0ms,P =202.65kPa(2atm),T A =400K, =0.4,0.6,0.8,1.0(fuel-leanside)and =1.2,1.6,2.0,3.0(fuel-richside). EGR=0.5 EGR=0.2 EGR=0.0 T A =400K EGRT=1200K, res =5.0ms Syn25,P=202.65kPa(2atm), Combustioneciency( c )EICO[g/kg]1.00 0.98 0.96 0.94 0.92 0.90 0.88 0.86 0.84 0.82 1000 100 10 Figure4-26.VariationofCOemissionindexforSyn25fuelwi thEGRof0.0,0.2and0.5 asafunctionofcombustioneciencyatEGRT=1200K, res =5.0ms,P= 202.65kPa(2atm),T A =400K, =0.4,0.6,0.8,1.0(fuel-leanside)and =1.2,1.6,2.0,3.0(fuel-richside). 141

PAGE 142

4.1.4EectofFuelCompositionandEGRonExtinctionLimits Thelean-sideextinctionlimitsofsyngasmixturesSyn50,S yn25andMethanewere alsoanalyzedatadiabaticEGRof0.0and2.0.Figures 4-27 and 4-28 showtheplotsof blowoutequivalenceratioswithloadingparameterandresi dencetime,respectively.The combustionstabilitywasfoundtoincreasedramaticallyat anadiabaticEGRof2.Syn50 showedthebestblowoutperformance,followedbySyn25andt henMethane.Thisisnot unexpectedandmaybeattributedtohigherH 2 contentinSyn50,whichexhibitsmuch widerrammabilitylimitsthanotherhydrocarbonfuels. Syn25(EGR=0.0) Syn50(EGR=2.0) Syn50(EGR=0.0) Methane(EGR=2.0) Methane(EGR=0.0) T A =400K adiabaticEGR P=202.65kPa(2atm) BlowoutLP[ g mol s L atm n ]Blowoutequivalenceratio( )10000 1000 100 10 1 1.00.90.80.70.60.50.4 Figure4-27.VariationofblowoutequivalenceratioforMet hanefuelwithEGR=0.0, MethanefuelwithEGR=2.0(adiabatic),Syn50fuelwithEGR= 0.0,Syn50 fuelwithEGR=2.0(adiabatic)andSyn25fuelwithEGR=0.0as afunction ofblowoutloadingparameter(LP)atP=202.65kPa(2atm),T A =400K. InFigure 4-29 ,theeectsofadiabaticandnon-adiabaticEGRoncombustio n stabilityhavebeencompared.Itwasfoundthat,whiletheex tinctionlimitswereimproved withadiabaticEGR,atanEGRstreamtemperatureof1200K,ad ecreaseincombustion stabilitywasobservedwithrisinglevelsofrecirculation .Since,theincreaseinadiabatic EGRwasaccompaniedbyacorrespondingdecreaseininletmas srow,foranygiven 142

PAGE 143

residencetime,theneteectofincreasingadiabaticEGRwa sthatofequivalently increasingthevolumeofthereactor,allowingmoretimefor thefreshreactantstoreact.A nearlylinearrelationshipwasobservedbetweentheblowou tequivalenceratioandreactor rametemperaturefordierentconditionsofEGR,andisdepi ctedinFigure 4-30 .It canbeobservedthattheblowouttemperatureswerelower,wi thincreasingEGR.Hence, itisimportanttonotethatintermsoftemperature,thestab ilityofSCCengineswas improvedwithEGR.InFigure 4-31 ,themixtureinlettemperaturesforMethanefuel, havebeenplottedforvariousequivalenceratiosandEGR.It canbeseenclearly,thatthe inletmixturetemperatureincreasedwithincreasingEGRan dwithtemperatureofEGR stream.Thishasaneectofosettingtheimpactofreductio nincombustionstabilitydue todilutionoftheinletstream. Syn25(EGR0.0) Syn50(EGR2.0) Syn50(EGR=0.0) Methane(EGR=2.0) Methane(EGR=0.0) T A =400K P=202.65kPa(2atm)adiabaticEGR Blowoutresidencetime( res )[ms]Blowoutequivalenceratio( )10.00 1.00 0.10 0.01 0.00 1.00.90.80.70.60.50.4 Figure4-28.VariationofblowoutequivalenceratioforMet hanefuelwithEGR=0.0, MethanefuelwithEGR=2.0(adiabatic),Syn50fuelwithEGR= 0.0,Syn50 fuelwithEGR=2.0(adiabatic)andSyn25fuelwithEGR=0.0as a functionofblowoutresidencetimeatP=202.65kPa(2atm),T A =400K. 143

PAGE 144

EGR=2.0,Adiabatic EGR=0.5,EGRT=1200K EGR=0.2,EGRT=1200K EGR=0.0 T A =400K P=202.65kPa(2atm)MethaneFuel BlowoutLP[ g mol s L atm n ]Blowoutequivalenceratio( )10000 1000 100 10 1 1.00.90.80.70.60.50.4 Figure4-29.VariationofblowoutequivalenceratioforMet hanefuelforEGR=0.0,EGR =0.2(EGRT=1200K),EGR=0.5(EGRT=1200K)andEGR=2.0(adiabatic)asafunctionofblowoutLPatP=202.65kPa(2atm ),T A = 400K. EGR=2.0,Adiabatic EGR=0.5,EGRT=1200K EGR=0.2,EGRT=1200K EGR0.0 T A =400K P=202.65kPa(2atm)Methanefuel Blowoutequivalenceratio( )Blowoutrametemperature[K]1.0 0.9 0.8 0.7 0.6 0.5 0.4 1900180017001600150014001300120011001000 Figure4-30.VariationofblowoutrametemperatureforMeth anefuelforEGR=0.0, EGR=0.2(EGRT=1200K),EGR=0.5(EGRT=1200K)andEGR=2.0(adiabatic)asafunctionofblowoutequivalenceratioatP= 202.65kPa(2 atm),T A =400K. 144

PAGE 145

EGR=2.0,Adiabatic EGR=0.5,EGRT=1200K EGR=0.2,EGRT=1200K EGR=0.0 T A =400K P=202.65kPa(2atm)Methanefuel Blowoutequivalenceratio( )Reactantmixtureinlettemperature[K]1.0 0.9 0.8 0.7 0.6 0.5 0.4 1000 900800700600500400300 Figure4-31.Variationofreactantmixtureinlettemperatu reforMethanefuelforEGR= 0.0,EGR=0.2(EGRT=1200K),EGR=0.5(EGRT=1200K)andEGR=2.0(adiabatic),T A =400K,P=202.65kPa(2atm)asafunctionof blowoutequivalenceratio. 145

PAGE 146

4.1.5EectofFuelCompositionandEGRonSootPrecursorFor mation Thesootconcentrationswerequantiedbytheconcentratio nofthesootprecursors, thepyrenemolecules(A 4 species).AsdepictedinFigure 4-32 ,theEIA 4 valuesobserved werezeroforallconditionstestedforsyngasfuels.WithMe thanefuel,EIA 4 values wereobservedonlyatfuel-richequivalenceratiosgreater than1.6.Itwasobservedthat theA 4 emissionsincreasedwithincreaseinadiabaticEGRathighe quivalenceratios. Furtherinvestigationswithfuelswithhighersootingtend ency,isrequired,tobeable todrawrmconclusionsfromtheanalysis.Webelieve,thatt helowsootinglimitsmay purelybetheresultofhomogeneityofcombustion,assumedf oraCSTRmodel,andmay notberepresentativeoftheenvironmentinsidethecombust orofapracticalgasturbine combustor. EGR=2.0,Adiabatic EGR=1.0,Adiabatic EGR=0.0 T A =400K P=202.65kPa(2atm)Methanefuel Equivalenceratio( )EI A 4 [g/kg]3.0 2.8 2.6 2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 1 1e-051e-101e-151e-201e-251e-301e-351e-40 Figure4-32.VariationofpyreneemissionindexforMethane fuelwithEGR=0.0,EGR= 1.0(adiabatic)andEGR=2.0(adiabatic)asafunctionofblo wout equivalenceratioatP=202.65kPa(2atm). 146

PAGE 147

4.2CaseStudy2:Biofuelsandn-HeptaneFuelSimulations Acomparisonofthecombustionblowoutpredictionsfortheb iofuelsispresented inFig. 4-33 throughFig. 4-38 .Forthesakeofclarity,onlycasesforEGRof0.0(solid markers)andEGRof2.0(hollowmarkers)havebeendepicted. Theblowouttemperatures werefoundtodropwithdecreasingequivalenceratiosforal lfuels.Additionally,since thesimulationsclosetoblowoutlimitwereinherentlyunst able,somedispersionabout thetrendlinesisnotunexpected.Wefoundthatnumericalin stabilitiesweresignicant closetorameblowout,especiallyathighrecirculationrat ios.Insomecases,solution multiplicitieswerenoticedsuchthat,forthesameinitial conditions,bothrameout andsteadyramesolutionswereobtained.Combustioncouldb esustainedatlower temperatureswithincreasingEGR,hencelowerblowouttemp eratureswereobservedfor higherEGR,asseeninFig. 4-33 .Interestingly,atanEGRof2.0,thegapbetweenthe blowout(BO)temperaturesofallfuelsgrewnarrower,suppo rtingthefactthatthePower, WaterExtractionandRefrigeration(PoWER)systemwasless sensitivetovariabilitiesin fuelcomposition. AsshowninFig. 4-34 andFig. 4-35 ,highCOandUHCemissionswereobservedfor allfuelsjustbeforeblowout.Sincewewouldexpectthecomb ustioneciencytobevery lowaroundblowout,theseplotsshowtheupperlimitsofCOan dUHCemissionsand depicttherelativepollutantgenerationtendenciesforea chofthefuels.MaximumCO emissionsatblowoutwereproducedbyn-heptanefollowedby MethylButanoate(MB), ethanolandlastlyDimethylEther(DME)atanEGRof0.0.Atth eEGRof2.0,much lowerEICOvalueswereobserved,andthedierenceinEICOof dierentfuelswasmuch less.MinimumEICOandEIUHCvalueswereobservedatequival enceratiosofabout0.7 formostfuels.AtanEGRofzero,n-heptanehadthehighestEI UHCwhileallbiofuels producedroughlysimilarEIUHCvalues.ItwasfoundthatEGR signicantlyreducedthe EIUHCandEICO,inpartduetoreductionoftheexitmassrowra te(_ m T m R )asa resultofEGR. 147

PAGE 148

T A =EGRT=700K P=202.65kPa(2atm) Blowoutequivalenceratio( )Blowoutrametemperature[K]1.0 0.9 0.8 0.7 0.6 0.5 0.4 21002000190018001700160015001400130012001100 Legend n H e p t a n e ( E G R = 2 0 ) M B ( E G R 2 0 ) D M E ( E G R = 2 0 ) E t h a n o l ( E G R = 2 0 ) n H e p t a n e ( E G R = 0 0 ) M B ( E G R = 0 0 ) D M E ( E G R = 0 0 ) E t h a n o l ( E G R = 0 0 ) Figure4-33.Comparisonofblowoutrametemperatureasafun ctionofblowout equivalenceratioforEthanol,DME,MB,n-HeptanefuelsatE GR=0.0and 2.0, T A =EGRT=700K,P=202.65kPa(2atm). Figure 4-36 ,showstheplotofblowoutequivalenceratioversusloading parameter. Themaximumcombustionblowoutstabilitydidnotalwaysocc uratBOequivalence ratiosof1.Thisresultwasunexpectedsincewithequivalen ceratiosapproaching1.0, therametemperatureswouldbehigher,whichshouldenhance combustionstability.At EGRof0.0,thecombustionstabilityofMBwashighest,follo wedbyn-heptane,DME andethanolbeingtheleaststable.However,atEGRof2.0,et hanolhadmaximumBO stabilityfollowedbyMBandthenn-heptaneandDME.Hence,t heapplicationofMBand ethanolblendsmaybeadvantageousinPoWERsysteminimprov ingstability,inaddition toenhancingtheignitionperformanceasaresultofincreas edvolatility,assuggested byBolszoetal.[ 126 ].However,therewasasignicantdropinstabilitywithinc reasing EGRforallfuels.Clearly,lowerBOequivalenceratioswere possibleevenatmuchhigher 148

PAGE 149

T A =EGRT=700K P=202.65kPa(2atm) Blowoutequivalenceratio( )BlowoutEICO[g/kg]1.0 0.9 0.8 0.7 0.6 0.5 0.4 900800700600500400300200100 0 Figure4-34.ComparisonofblowoutCOemissionindexasafun ctionofblowout equivalenceratioforEthanol,DME,MB,n-HeptanefuelsatE GR=0.0and 2.0, T A =EGRT=700K,P=202.65kPa(2atm). loadingparametersforEGRof0.0incomparisontothecasewi thEGRof2.0.This highlightsaninherentdrawbackofreducedcombustionstab ilityinthePoWERsystem. Thisdrawbackmaybeovercomebytheuseoffuelswithwiderbl owoutstabilitylimits likehydrogenorsyngas,useofhigherrecirculationtemper atureswhereverapplicable,or possiblytechniqueslikeH 2 enrichment. Fig 4-37 depictstheplotofBOequivalenceratioasafunctionofresi dencetime,while Fig. 4-38 depictsBOequivalenceratioversustotalmassrowrate.The limitingresidence timesforEGRof0.0werealmostanorderof2.0lowerthanthos eforEGRof2.0. TheFigures 4-39 through 4-44 wereplottedataresidencetimeof10ms.Figure 4-39 showsplotofequivalenceratioversustemperature.Therei samarkedreduction inrametemperaturewithincreaseinEGR,forallfuels.With increasingEGR,the dierenceinrametemperaturesfordierentfuelsdecrease s.DMEproducedmaximum rametemperatureforallequivalenceratios,followedbynheptane,MBandthelowest 149

PAGE 150

T A =EGRT=700K P=202.65kPa(2atm) Blowoutequivalenceratio( )BlowoutEIUHC[g/kg]1.0 0.9 0.8 0.7 0.6 0.5 0.4 1000 100 10 1 Figure4-35.ComparisonofblowoutUHCemissionindexasafu nctionofblowout equivalenceratioforEthanol,DME,MB,n-HeptanefuelsatE GR=0.0and 2.0, T A =EGRT=700K,P=202.65kPa(2atm). beingthatofethanol.Combustioneciencyversusequivale nceratiohasbeenplottedin Fig. 4-40 .MBandDMEhadhigherorcomparableeciencies,followedby ethanoland thenn-heptaneovermostequivalenceratios.Forthecaseof zeroEGR,thecombustion eciencyincreasedwithdecreasingequivalenceratio,for theequivalenceratiosbetween 0.4to1.0.WithanincreaseinEGR,thecombustioneciencyi ncreasedandanoptimum equivalenceratioemergedformaximumeciency.Moreover, theoptimumequivalence ratioshiftedrighttowardsstoichiometric,withincrease inEGR. Figures 4-41 and 4-42 showplotsofequivalenceratioversusEICOandEIUHC respectively.TheCOemissionswerehighestforn-heptanew hiletheotherfuelsshowed nearlythesametendencytoCOformation,forEGRof0.0aswel las2.0.ForanEGR of0.0and2.0thelowestemissionswereobservedaroundaneq uivalenceratiosofabout 0.5and0.9,respectively.Thatis,therewasashiftintheop timumequivalenceratiofor lowestCOemisions,towardsstoichiometric,withanincrea seinEGR.TheUHCemissions 150

PAGE 151

T A =EGRT=700K P=202.65kPa(2atm) BlowoutLP[ g mol s L atm n ]Blowoutequivalenceratio( )1e+03 1e+02 10 1 0.1 0.01 1.00.90.80.70.60.50.4 Figure4-36.Comparisonofblowoutequivalenceratioasafu nctionofblowoutloading parameterforEthanol,DME,MB,n-HeptanefuelsatEGR=0.0a nd2.0, T A =EGRT=700K,P=202.65kPa(2atm). werehighestforn-heptane,followedbyethanol,DMEandthe lowestbeingthatofMB overmostequivalenceratios.Therangeofequivalencerati os,forastablerame,also decreasedwithincreasingEGR.Ithasbeendemonstratedtha tequivalenceratiosclose tostoichiometriccanbeusedwithincreasingEGRforagiven turbineinlettemperature [ 34 39 42 ].Inordertoachievehighgasturbinethermodynamiccyclee ciency,the designoutlettemperatureofthecombustorismaintainedcl osetothemaximumturbine inlettemperature,whichislimitedbythecurrentstateoft heartingasturbineblade materials.Higheroverallair-fuelratiosareemployedtoc ooltherowtothedesignvalue. Since,theuseofEGR,inherentlylowerstherametemperatur es,higherequivalenceratios canbeemployedforagiventurbineinlettemperature. InFig. 4-43 ,thevariationofEIUHCwithcombustioneciencyhasbeende picted. TheUHCemissionsincreasesignicantlyforlowercombusti oneciencies.Theplotis 151

PAGE 152

self-explanatory,sincetheUHCandCOemissionsintheexha ustareadirectmeasureof combustionineciency. T A =EGRT=700K P=202.65kPa(2atm) Blowoutresidencetime[ms]Blowoutequivalenceratio( )1e+03 1e+02 10 1 0.1 0.01 1.00.90.80.70.60.50.4 Figure4-37.Comparisonofblowoutresidencetimeasafunct ionofblowoutequivalence ratioforEthanol,DME,MB,n-HeptanefuelsatEGR=0.0and2. 0, T A = EGRT=700K,P=202.65kPa(2atm). Figure 4-44 showstheplotofcombustioneciencyversustemperaturefo rallfuels. Clearly,allbiofuels,namely,MB,DMEandethanolhaveasup eriororcomparable emissionsperformanceascomparedton-heptaneatEGRof0.0 and2.0.AtanEGR of2.0,asisevidentfromtheplot,theoverallimpactoncomb ustioneciencyduetoa changeinfuelissubstantiallydampenedout.Forexample,a tanEGRof0.0,theoverall ecienciesvarybetween0.80to0.98whileforanEGRof2.0,t hecombustioneciency variationsarelimitedtotherangeof0.96to1.00.Thisfurt herconrmstheimproved fuel-rexibilityofthePoWERsystem.Atthesametime,theli mitsofoperabilityin temperatureareconsiderablynarrower,makingthesystemv ulnerabletoramequenching incoldspots. 152

PAGE 153

T A =EGRT=700K P=202.65kPa(2atm) Blowoutmassrowrate(_ m T )[kg/s]Blowoutequivalenceratio( )1e+02 10 1 0.1 0.01 1.00.90.80.70.60.50.4 Figure4-38.Comparisonofblowoutequivalenceratioasafu nctionofblowoutmassrow rateforEthanol,DME,MB,n-HeptanefuelsatEGR=0.0and2.0 ,EGRT =700K,P=202.65kPa(2atm). T A =EGRT=700K, res =10.0ms P=202.65kPa(2atm) Equivalenceratio( )Flametemperature[K]1.0 0.9 0.8 0.7 0.6 0.5 0.4 26002400220020001800160014001200 Figure4-39.Comparisonoframetemperatureasafunctionof equivalenceratiofor Ethanol,DME,MB,n-HeptanefuelsatEGR=0.0and2.0, T A =EGRT= 700K, res =10.0ms,P=202.65kPa(2atm). 153

PAGE 154

T A =EGRT=700K, res =10.0ms P=202.65kPa(2atm) Equivalenceratio( )Combustioneciency( c )1.0 0.9 0.8 0.7 0.6 0.5 0.4 1.000.980.960.940.920.900.880.860.840.82 Figure4-40.Comparisonofcombustioneciencyasafunctio nofequivalenceratiofor Ethanol,DME,MB,n-HeptanefuelsatEGR=0.0and2.0, T A =EGRT= 700K, res =10.0ms,P=202.65kPa(2atm). T A =EGRT=700K, res =10.0ms P=202.65kPa(2atm) Equivalenceratio( )EICO[g/kg]1.0 0.9 0.8 0.7 0.6 0.5 0.4 400350300250200150100 50 0 Figure4-41.ComparisonofCOemissionindexasafunctionof equivalenceratiofor Ethanol,DME,MB,n-HeptanefuelsatEGR=0.0and2.0, T A =EGRT= 700K, res =10.0ms,P=202.65kPa(2atm). 154

PAGE 155

T A =EGRT=700K, res =10.0ms P=202.65kPa(2atm) Equivalenceratio( )EIUHC[g/kg]1.0 0.9 0.8 0.7 0.6 0.5 0.4 100 10 1 Figure4-42.ComparisonofUHCemissionindexasafunctiono fequivalenceratiofor Ethanol,DME,MB,n-HeptanefuelsatEGR=0.0and2.0, T A =EGRT= 700K, res =10.0ms,P=202.65kPa(2atm). T A =EGRT=700K, res =10.0ms P=202.65kPa(2atm), =0.4-1.0 Combustioneciency( c )EIUHC[g/kg]1.00 0.98 0.96 0.94 0.92 0.90 0.88 0.86 0.84 0.82 100 10 1 Figure4-43.ComparisonofUHCemissionindexasafunctiono fcombustioneciencyfor Ethanol,DME,MB,n-HeptanefuelsatEGR=0.0and2.0, T A =EGRT= 700K, res =10.0ms,P=202.65kPa(2atm), =0.4-1.0. 155

PAGE 156

T A =EGRT=700K, res =10.0ms P=202.65kPa(2atm), =0.4-1.0 Flametemperature[K]Combustioneciency( c )2600 2400 2200 2000 1800 1600 1400 1200 1.000.980.960.940.920.900.880.860.840.82 Figure4-44.Comparisonofcombustioneciencyasafunctio noframetemperaturefor Ethanol,DME,MB,n-HeptanefuelsatEGR=0.0and2.0, T A =EGRT= 700K, res =10.0ms,P=202.65kPa(2atm), =0.4-1.0. 156

PAGE 157

4.3CaseStudy3:StudyofEGRTemperature/EGRlevelonEthan ol BlowoutLimits Thetrendsinethanolfuelblowoutlimitswereinvestigated foroperatingconditions P=202650Pa(2atm)andV=0.00258m 3 (15.71in 3 )and =1 : 0,varyingEGRvalues to0.0,0.5,1.0,1.5,2.0,andEGRTto400,600,800,1000,120 0K.Figures 4-45 through 4-48 showthetrendsinrametemperature,LP,residencetimeandt otalmassrowrateat theblowoutlimit,respectively,asafunctionofvaryingEG RandEGRtemperature.The blowouttemperature,asshowningure 4-45 variedlinearlywithEGRtemperature,and non-linearlywithEGRlevel,asisapparentfromthespacing betweenpredictedtrends. AnearlylogarithmicvariationcanbeobservedforblowoutL P,residencetimeandtotal massrowrate,withvaryingEGRtemperature.Thedegreeofno n-linearity(onlogscale) wasfoundtoincreasewithincreasingEGR.TheLPandtotalma ssrowratedecreased, whileresidencetimeincreasedatblowout,withincreasing EGR.Thetrendsindicatea sharpdropinblowoutstabilitywithincreasingEGR,whilea nincreaseinstabilitywith temperature.Additionally,theeectoftemperaturewasor dersofmagnitudestrongerat higherEGR. Ingures 4-49 through 4-51 theblowoutresidencetimewasplottedagainstOH,H andCOmassfractions,respectively.Thetrendsshowedastr ongdependenceonthe speciesmassfractions,andthepredictionswerefoundtofa llnearlyonasinglecurve (withsomescatter),irrespectiveoftheEGRrowandtempera ture.Theseresultswere encouraging,andwesubsequentlyplottednormalizedpredi ctionsforvariousparameters, inordertocollapsethepredictionstoasinglecurve.Thepr edictionswerenormalizedto thecorrespondingvaluesforthezeroEGRcase(subscripted with0). Theplotofnormalizedblowoutresidencetimevs.normalize dblowouttemperatureas showningure 4-52 collapsedthemodelpredictionstoanearlysingleline,sug gesting astrongtemperatureeectonblowoutlimits.Theplotofnor malizedblowoutLPvs. normalizedblowouttemperatureisshowningure 4-53 .Similarly,normalizedblowoutLP 157

PAGE 158

EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 EGR=0.0 =1.0 Ethanolfuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]Blowoutrametemperature[K]1200 1100 1000 900 800 700 600 500 400 200019001800170016001500140013001200 Figure4-45.Comparisonofblowoutrametemperatureforeth anolfuelasafunctionof EGRtemperatureatEGR=0.0,0.5,1.0,1.5,2.0, =1.0,P=202.65kPa (2atm). vs.normalizedOHandCOmassfractionswereplottedingure 4-54 andgure 4-55 ,and thepredictionscollapsedtoasingleline.Aremarkabledep endenceonOHconcentrations wasfound.Thisisofgreatimportancetoexperimentalists, sinceOHemissionshavebeen usedtostudyramestructureinramelesscombustion.Itiswe llknownthatOHemissions correlatewellwithluminosityofthesystemandtheheatrel easerate.Moreover,the normalizedloadingparameterversusOHmassfractionpredi ctionsforstableoperating pointsmayalsoexplainthesuddendropinluminositywithin creasingEGR(fora constantEGRtemperature),astheoperatingpointjumpstoa lowerpointonthe normalizedLPvs.OHmassfractiontrendline.Hence,thiscu rvemightbeofsignicance todemarcatethetransitioningintotheramelessregime.Th eOHemissionsaregenerally measuredusingmoreexpensiveinstrumentationsuchaschem iluminescenceorLIF.Hence, weexaminedthecorrelationofnormalizedLPpredictionswi thnormalizedCO,whichmay bemeasuredusingrelativelyinexpensivehardware.Anothe rbenetofsuchacorrelation wouldbethat,COconcentrationisastrongfunctionofcombu stioneciency.Hence, 158

PAGE 159

combustioneciencycouldbeusedtoindicatea`blowoutmar gin.'ThenormalizedLP valueswerefoundtocorrelatewellwiththenormalizedCOva lues. EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 EGR=0.0 =1.0 Ethanolfuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]Blowoutloadingparameter(LP)[ g mol s L atm n ]1200 1100 1000 900 800 700 600 500 400 1000 100 10 10 Figure4-46.Comparisonofblowoutloadingparameterforet hanolfuelasafunctionof EGRtemperatureatEGR=0.0,0.5,1.0,1.5,2.0, =1.0,P=202.65kPa (2atm). Figure 4-56 showsthevariationofignitiondelaywithequilibriumrame temperature atvariousEGRlevels.Theignitiondelaywascalculatedfro mconstantpressurebatch reactorcalculations,basedonthecharacteristictimedel ayforattainmentofmaximum OHconcentration.Theresultshowsthattheequilibriumtem peratureforthezeroEGR casewouldneedtobeincreasedtoattainalowerignitiondel aycomparabletothatfor anon-zeroEGRcase(byincreasinginlettemperature).Alte rnately,comparingcasesat xedequilibriumtemperature,thecaseswithhigherEGRwou ldalsorequirehigherinlet temperature,andthuswouldhavealowerignitiondelay.Ont hecontrary,thetrends plottedingure 4-57 hadaverystrongcorrelationoftheignitiondelaywithresp ectto theinlettemperature,forallEGR.Surprisingly,theignit iondelaywasfoundtodrop withEGR,andmaybeattributedtothepresenceofreactivein termediatesintheEGR stream,assuggestedbyKalbandSattelmayer[ 16 ].Figure 4-58 showsthevariationin 159

PAGE 160

ignitiondelaywithEGRataxedreactantmixturetemperatu reof2600K.Ingure 4-59 blowoutresidencetimehasbeenplottedagainstignitionde lay.Theignitiondelays plottedcorrespondedtotheequilibriumtemperatureforth esteadystatecompositions ofcombustorCSTR,atvariousblowoutLP.Theseequilibrium temperatureswerefar lowerthantheequilibriumtemperaturescalculatedintheb atchreactorsimulations,to estimateignitiondelays.Asaresult,theignitiondelaysa ttheCSTRconditionshadto beextrapolated,andthereforemaybefarfromaccurate.How ever,theinterestbehind theignitiondelaypredictionswastocorrelatetheignitio nnumberswithLP,andobtain atleastqualitativetrendsfromtheanalysis.ThetrendofL Pasafunctionoftheratio ofresidencetimetoignitiondelay(ignitionnumber)hasbe endepictedingure 4-60 TheblowoutLPvaluesvariednearlylinearlywiththeigniti onnumberonalog-log plot,atEGR=0.0.However,strongnon-linearitiessurface dwithincreasingEGRfor predictionsatthelowerLPvalues(forlowerEGRtemperatur es).Finally,normalized LPversusnormalizedignitionnumberhasbeenshowningure 4-61 ,forvariousEGR values.Thisgureshowsasignicantincreaseinignitionn umberandacorresponding decreaseinloadingparameter,athigherEGRandlowerEGRT, whencomparedwiththe EGR=0.0case.However,itwasobservedthatatreasonablylo wEGRthenormalized ignitionnumberatblowoutmaydecreasewithdecreasingEGR Tbeforeincreasing,while normalizedblowoutloadingparameterdecreasesmonotonic ally.Asaresult,depending ontheEGRT,thenormalizedignitionnumbermaycorrespondt odierentnormalized blowoutloadingparameters(non-unique),especiallyatlo wEGR. Theeectofpressureonrameblowouthasbeendepictedingu res 4-62 through 4-65 .Theblowoutlimitswerereducedasthepressurewasreduced ,predominantlydueto weakeningoftheheatreleaserate.Figure 4-62 showsareductioninrametemperature withdecreasingpressure.Theeectofpressureseemedtobe morepronouncedwhen operatingwithEGR,ascomparedtotheopencycle(OC)mode,a sseeningures 4-63 through 4-65 forblowoutLP,residencetimeandblowoutmassrowrate,res pectively. 160

PAGE 161

EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 EGR=0.0 Ethanolfuel, =1.0,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]Blowoutresidencetime[ms]1200 1100 1000 900 800 700 600 500 400 10 1 0.1 0.01 Figure4-47.Comparisonofblowoutresidencetimeforethan olfuelasafunctionofEGR temperatureatEGR=0.0,0.5,1.0,1.5,2.0, =1.0,P=202.65kPa(2 atm). EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 EGR=0.0 =1.0 Ethanolfuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]Blowoutmassrowrate[kg/s]1200 1100 1000 900 800 700 600 500 400 100 10 10 Figure4-48.Comparisonofblowoutmassrowrateforethanol fuelasafunctionofEGR temperatureatEGR=0.0,0.5,1.0,1.5,2.0, =1.0,P=202.65kPa(2 atm). 161

PAGE 162

EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 EGR=0.0 =1.0 Ethanolfuel,P=202.65kPa(2atm) OHspeciesmassfraction( y OH )Blowoutresidencetime( res )[ms]0.004 0.0035 0.003 0.0025 0.002 0.0015 0.001 0.0005 0 10 100 Figure4-49.Comparisonofblowoutresidencetimeforethan olfuelasafunctionofOH speciesmassfractionatEGR=0.0,0.5,1.0,1.5,2.0,EGRT=4 00,600,800, 1000,1200K, =1.0,P=202.65kPa(2atm). EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 EGR=0.0 =1.0 Ethanolfuel,P=202.65kPa(2atm) Hspeciesmassfraction( y H )Blowoutresidencetime[ms]1.8e-04 1.2e-04 8.0e-05 4.0e-05 0.0e+00 10 1 0.1 0.01 Figure4-50.Comparisonofblowoutresidencetimeforethan olfuelasafunctionofH speciesmassfractionatEGR=0.0,0.5,1.0,1.5,2.0,EGRT=4 00,600,800, 1000,1200K, =1.0,P=202.65kPa(2atm). 162

PAGE 163

EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 EGR=0.0 =1.0 Ethanolfuel,P=202.65kPa(2atm) COspeciesmassfraction( y CO )Blowoutresidencetime[ms]0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0 10 1 0.1 0.01 Figure4-51.Comparisonofblowoutresidencetimeforethan olfuelasafunctionofCO speciesmassfractionatEGR=0.0,0.5,1.0,1.5,2.0,EGRT=4 00,600,800, 1000,1200K, =1.0,P=202.65kPa(2atm). EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 =1.0 Ethanolfuel,P=202.65kPa(2atm) =1.0 Ethanolfuel,P=202.65kPa(2atm) Normalizedrametemperature( T T 0 )Normalizedresidencetime( res res; 0 )0.95 0.9 0.85 0.8 0.75 0.7 1e+031e+02 10 1 Figure4-52.Comparisonofnormalizedresidencetimeforet hanolfuelasafunctionof normalizedrametemperatureatEGR=0.0,0.5,1.0,1.5,2.0, EGRT=400, 600,800,1000,1200K, =1.0,P=202.65kPa(2atm). 163

PAGE 164

EGR2.0 EGR1.5 EGR1.0 EGR0.5 =1.0 Ethanolfuel,P=202.65kPa(2atm) Normalizedrametemperature( T T 0 )Normalizedblowoutloadingparameter( LP LP 0 )0.95 0.9 0.85 0.8 0.75 0.7 1 0.1 0.01 0.001 Figure4-53.Comparisonofnormalizedblowoutloadingpara meterforethanolfuelasa functionofnormalizedrametemperatureatEGR=0.0,0.5,1. 0,1.5,2.0, EGRT=400,600,800,1000,1200K, =1.0,P=202.65kPa(2atm). EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 =1.0 Ethanolfuel,P=202.65kPa(2atm) NormalizedOHmassfraction( y OH y OH; 0 )Normalizedblowoutloadingparameter( LP LP 0 )0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 1 0.1 0.01 0.001 Figure4-54.Comparisonofnormalizedblowoutloadingpara meterforethanolfuelasa functionofnormalizedOHspeciesmassfractionatEGR=0.0, 0.5,1.0,1.5, 2.0,EGRT=400,600,800,1000,1200K, =1.0,P=202.65kPa(2atm). 164

PAGE 165

EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 =1.0 Ethanolfuel,P=202.65kPa(2atm) NormalizedCOmassfraction( y CO y CO; 0 )Normalizedblowoutloadingparameter( LP LP 0 )1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 1 0.1 0.01 0.001 Figure4-55.Comparisonofnormalizedblowoutloadingpara meterforethanolfuelasa functionofnormalizedCOspeciesmassfractionatEGR=0.0, 0.5,1.0,1.5, 2.0,EGRT=400,600,800,1000,1200K, =1.0,P=202.65kPa(2atm). EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 EGR=0.0 =1.0 Ethanolfuel,P=202.65kPa(2atm) Equilibriumtemperature[K]Ignitiondelay( I )[ s]3000 2900 2800 2700 2600 2500 2400 2300 2200 2100 2000 1e+041e+031e+02 10 Figure4-56.Comparisonofignitiondelayforethanolfuela safunctionofequilibrium rametemperatureatEGR=0.0,0.5,1.0,1.5,2.0,reactantmi xtureinlet temperature1200to2500K, =1.0,P=202.65kPa(2atm). 165

PAGE 166

EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 EGR=0.0 =1.0 Ethanolfuel,P=202.65kPa(2atm) Reactantmixtureinlettemperature[K]Ignitiondelay( I )[ s]2600 2400 2200 2000 1800 1600 1400 1200 1e+041e+031e+02 10 Figure4-57.Comparisonofignitiondelayforethanolfuela safunctionofreactant mixtureinlettemperatureatEGR=0.0,0.5,1.0,1.5,2.0, =1.0,P= 202.65kPa(2atm). Ignitiondelay =1.0 Ethanolfuel,P=202.65kPa(2atm) EGRIgnitiondelay( I )[ s]2.0 1.5 1.0 0.5 0.0 1e+041e+031e+02 10 Figure4-58.Variationofignitiondelayforethanolfuelas afunctionofEGRatreactant mixtureinlettemperature=2600K, =1.0,P=202.65kPa(2atm). 166

PAGE 167

EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 EGR=0.0 =1.0 Ethanolfuel,P=202.65kPa(2atm) Ignitiondelay[ms]Blowoutresidencetime[ms]1000 100 10 1 0 10 1 0.1 0.01 Figure4-59.Comparisonofblowoutresidencetimeforethan olfuelasafunctionof ignitiondelayatEGR=0.0,0.5,1.0,1.5,2.0,EGRT=400,600 ,800,1000, 1200K, =1.0,P=202.65kPa(2atm). EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 EGR=0.0 Ethanolfuel, =1.0,P=202.65kPa(2atm) Blowoutignitionnumber( N I )Blowoutloadingparameter(LP)[ g mol s L atm n ]1 0.1 0.01 0.001 0.0001 1000 100 10 10 Figure4-60.Comparisonofblowoutloadingparameterasafu nctionofignitionnumberat EGR=0.0,0.5,1.0,1.5,2.0,EGRT=400,600,800,1000,1200K =1.0, P=202.65kPa(2atm). 167

PAGE 168

EGR=2.0 EGR=1.5 EGR=1.0 EGR=0.5 =1.0 Ethanolfuel,P=202.65kPa(2atm) NormalizedblowoutignitionNumber( N I N I; 0 )Normalizedblowoutloadingparameter( LP LP 0 )1000 100 10 1 0.1 0.01 0.001 Figure4-61.Comparisonofnormalizedblowoutloadingpara meterasafunctionof normalizedignitionnumberatEGR=0.0,0.5,1.0,1.5,2.0,E GRT=400, 600,800,1000,1200K, =1.0,P=202.65kPa(2atm). 168

PAGE 169

P=0.5atm,EGR=2.0 P=1.0atm,EGR=2.0 P=1.5atm,EGR=2.0 P=2.0atm,EGR=2.0 P=0.5atm,EGR=0.0 P=1.0atm,EGR=0.0 P=1.5atm,EGR=0.0 P=2.0atm,EGR=0.0 =1.0 Ethanolfuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]Blowoutrametemperature[K]1200 1100 1000 900 800 700 600 500 400 200019001800170016001500140013001200 Figure4-62.Comparisonofblowoutrametemperatureasafun ctionofairstream/EGR temperatureatEGR=0.0and2.0,P=2.0,1.5,1.0,0.5atmand =1.0. P=0.5atm,EGR=2.0 P=1.0atm,EGR=2.0 P=1.5atm,EGR=2.0 P=2.0atm,EGR=2.0 P=0.5atm,EGR=0.0 P=1.0atm,EGR=0.0 P=1.5atm,EGR=0.0 P=2.0atm,EGR=0.0 =1.0 Ethanolfuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]BlowoutLP[ g mol s L atm n ]1200 1100 1000 900 800 700 600 500 400 1000 100 10 1 Figure4-63.Comparisonofblowoutloadingparameterasafu nctionofairstream/EGR temperatureatEGR=0.0and2.0,P=2.0,1.5,1.0,0.5atmand =1.0. 169

PAGE 170

P=0.5atm,EGR=2.0 P=1.0atm,EGR=2.0 P=1.5atm,EGR=2.0 P=2.0atm,EGR=2.0 P=0.5atm,EGR=0.0 P=1.0atm,EGR=0.0 P=1.5atm,EGR=0.0 P=2.0atm,EGR=0.0 =1.0 Ethanolfuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]Blowoutresidencetime[ms]1200 1100 1000 900 800 700 600 500 400 1e+01 1 0.1 0.01 Figure4-64.Comparisonofblowoutresidencetimeasafunct ionofairstream/EGR temperatureatEGR=0.0and2.0,P=2.0,1.5,1.0,0.5atmand =1.0. P=0.5atm,EGR=2.0 P=1.0atm,EGR=2.0 P=1.5atm,EGR=2.0 P=2.0atm,EGR=2.0 P=0.5atm,EGR=0.0 P=1.0atm,EGR=0.0 P=1.5atm,EGR=0.0 P=2.0atm,EGR=0.0 =1.0 Ethanolfuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]Blowoutmassrowrate[kg/s]1200 1100 1000 900 800 700 600 500 400 1e+021e+01 1 0.1 Figure4-65.Comparisonofblowoutmassrowrateasafunctio nofairstream/EGR temperatureatEGR=0.0and2.0,P=2.0,1.5,1.0,0.5atmand =1.0. 170

PAGE 171

4.4CaseStudy4:AStudyoftheEectofEGRTemperature/EGRl evel/ ResidenceTime/EquivalenceRatioonSootFormation(Acety leneFuel Combustion) Severalcases,asdescribedinsection 3.3 ,wereruninordertoisolatetheeectsof systemparameterssuchasEGR,EGRtemperature,equivalenc eratioandresidencetime. Theresultsofthesimulationshavebeenpresentedbelow.4.4.1EectofEGR TheeectofEGRonemissionscorrespondstocase1(EGR=0.05.0, =2 : 0, res =1 : 0,EGRT=1000K)insection 3.3 ,andtheresultshavebeenshowningures 4-66 through 4-71 .Therametemperaturedropsmonotonicallywithincreasing EGR,asshown ingure 4-66 .Figure 4-67 showsaplotofUHCemissionindexasafunctionofEGR.The UHCemissionindexwasfoundtoincreasewithEGR,untilanEG Rof3.0,afterwhich itwasfoundtobenearlyconstant.TheCOemissionindexshow ningure 4-68 ,onthe otherhand,decreasedinitiallywithEGR,andeventuallyst abilizedbeyondanEGRof about2.5.C 2 H 2 hasbeenshowntobeanimportantgrowthspeciesinthesootfo rmation mechanism,andhasbeenplottedingure 4-69 .Thecombustioneciencywasfoundto increaseslightlyandthendecreasewithincreasingEGR,as showningure 4-70 .Note thatthecombustionecienciesareverylow,becausetheres idencetimechosenforthis studywaslow(1ms). Thevariationofemissionindicesoffourpolyaromatichydr ocarbons(PAH)moleculesbenzene(A 1 ),naphthalene(A 2 ),anthracene(A 3 )andpyrene(A 4 )hasbeenpresentedin gure 4-71 .TheemissionindicesforallPAHmoleculesmodeledincreas edtoamaximum forEGRofabout2.0andthendecreased.Thisisnotsurprisin g,sincevitiationof thereactantswithEGRhasastrongthermaleect.Asaresult ,thecharacteristic \soot-bell"liketemperaturedependencewasrerectedinth etrendsofthePAHmolecules (sootprecursors)withEGR,asisevidentingure 4-71 ,correspondingtoadropin temperaturefromaprrox.2200K(atEGR=0.5)toapprox.1300 K(atEGR=5). Athightemperatures,theoxidativereactionsbecomepredo minant,reducingthesoot 171

PAGE 172

precursorconcentrations.Ontheotherhand,atverylowtem peratures,sootformation ratesarelowduetoinsucientradicalconcentrations,con tributingtothelowratesof hydrogenabstractionreactions. Flametemperature EGRT=1000K, =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) EGRFlametemperature[K]5 4 3 2 1 0 280026002400220020001800160014001200 Figure4-66.Variationoframetemperatureforacetylenefu elasafunctionofEGRat EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). EIUHC EGRT=1000K, =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) EGREIUHC[g/kg]5 4 3 2 1 0 350300250200150100 50 Figure4-67.VariationofUHCemissionindexforacetylenef uelasafunctionofEGRat EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). 172

PAGE 173

EICO EGRT=1000K, =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) EGREICO[g/kg]5 4 3 2 1 0 20001900180017001600150014001300120011001000 Figure4-68.VariationofCOemissionindexforacetylenefu elasafunctionofEGRat EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). EIC 2 H 2 EGRT=1000K, =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) EGREIC 2 H 2 [g/kg]5 4 3 2 1 0 250200150100 50 0 Figure4-69.VariationofC 2 H 2 emissionindexforacetylenefuelasafunctionofEGRat EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). 173

PAGE 174

Combustioneciency EGRT=1000K, =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) EGRCombustioneciency( c )5 4 3 2 1 0 0.480.470.460.450.440.430.420.410.40 Figure4-70.Variationofcombustioneciencyforacetylen efuelasafunctionofEGRat EGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). A 4 A 3 A 2 A 1 EGRT=1000K, =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) EGREI A i (PAHs)[g/kg]5 4 3 2 1 0 25201510 50 Figure4-71.ComparisonofPAHemissionindicesforacetyle nefuelasafunctionofEGR atEGRT=1000K, =2.0, res =1.0ms,P=202.65kPa(2atm). 174

PAGE 175

4.4.2EectofEGRTemperature TheeectofEGRtemperature(case2insection 3.3 withparametersEGRT= 400-1200K,EGR=0.0and2.0, =2 : 0, res =1 : 0)onsootprecursormoleculeswas modeled.Figures 4-72 through 4-79 depicttheeectofEGRtemperatureforEGR of0and2.Therametemperaturewasfoundtovarylinearlywit hincreasingEGR temperatureasplottedingure 4-72 ,withtheslopedependentonEGR.TheUHC andCOemissionindextrendswithrespecttoEGRtemperature havebeenpresented ingure 4-73 andgure 4-74 ,respectively.AtzeroEGR,theemissionindicesforboth COandUHCwerenearlyindependentofairstream/EGRtempera ture(theinletair temperaturewasvariedsimultaneouslywithEGRtemperatur e).Ontheotherhand, atEGRof2.0,theEIUHCdecreasedwhileEICOincreasedwithE GRtemperature. TheC 2 H 2 concentrations,asshowningure 4-75 ,decreasedwithEGRT(andinletair temperature),forbothEGRof0.0andEGR2.0. EGR=2.0 EGR=0.0 =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]Flametemperature[K]1400 1200 1000 800 600 400 3000280026002400220020001800160014001200 Figure4-72.Comparisonoframetemperatureforacetylenef uelasafunctionofair stream/EGRtemperatureatEGR=0.0andEGR=2.0, =2.0, res =1.0 ms,P=202.65kPa(2atm). 175

PAGE 176

Thecombustioneciencywasfoundtohaveoppositetrendsat EGR0.0andEGR 2.0,withvaryingEGRtemperatures.Itcanbeseenfromgure 4-76 thatcombustion eciencyimprovedremarkablywithEGRT,forEGR=2.0,while itdroppedslightly forEGR=0.0.ItmayalsobenotedthattheEIUHCandEIC 2 H 2 valuesaremuch higherforEGRof2.0,suggestiveofmorefavorablesootingc onditions.Inaddition,the EIC 2 H 2 concentrationsdroppedatalowerratewithincreasingEGRT forEGRof2.0,as comparedtoEGRof0.0.ThePAHemissionindices(EIA i ,fori=1,2,3and4)havebeen plottedingure 4-77 togure 4-79 .ThePAHemissionindiceswerehigheracrossallEGR temperaturesforEGRof2.0,comparedtotheEGR=0.0case.Th isisnotsurprising since,therametemperaturesforEGR=0.0casewerehigherth an2200K,indicating oxidizingregime,whiletheEGR=2.0casehadrametemperatu resinthe1400-2100K range,whichisthefavorablesootformationregime. EGR=2.0 EGR=0.0 =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]EIUHC[g/kg]1400 1200 1000 800 600 400 400350300250200150100 50 Figure4-73.ComparisonofUHCemissionindexforacetylene fuelasafunctionofair stream/EGRtemperatureatEGR=0.0andEGR=2.0, =2.0, res =1.0 ms,P=202.65kPa(2atm). 176

PAGE 177

EGR=2.0 EGR=0.0 =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]EICO[g/kg]1400 1200 1000 800 600 400 20001900180017001600150014001300120011001000 Figure4-74.ComparisonofCOemissionindexforacetylenef uelasafunctionofair stream/EGRtemperatureatEGR=0.0andEGR=2.0, =2.0, res =1.0 ms,P=202.65kPa(2atm). EGR=2.0 EGR=0.0 =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]EIC 2 H 2 [g/kg]1400 1200 1000 800 600 400 300250200150100 50 0 Figure4-75.ComparisonofC 2 H 2 emissionindexforacetylenefuelasafunctionofair stream/EGRtemperatureatEGR=0.0andEGR=2.0, =2.0, res =1.0 ms,P=202.65kPa(2atm). 177

PAGE 178

EGR=2.0 EGR=0.0 =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]Combustioneciency( c )1400 1200 1000 800 600 400 0.480.470.460.450.440.430.420.410.400.39 Figure4-76.Comparisonofcombustioneciencyforacetyle nefuelasafunctionofair stream/EGRtemperatureatEGR=0.0andEGR=2.0, =2.0, res =1.0 ms,P=202.65kPa(2atm). A 4 A 3 A 2 A 1 EGR=0.0, =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]EIA i (PAHs)[g/kg]1400 1200 1000 800 600 400 1 1e-051e-101e-151e-201e-251e-30 Figure4-77.ComparisonofPAHemissionindicesforacetyle nefuelasafunctionofair stream/EGRtemperatureatEGR=0.0, =2.0, res =1.0ms,P=202.65 kPa(2atm). 178

PAGE 179

A 4 A 3 A 2 A 1 EGR=2.0, =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]EIA i (PAH)[g/kg]1400 1300 1200 1100 1000 900 800 700 600 1e+02 10 1 0.1 0.01 0.001 0.0001 Figure4-78.ComparisonofPAHemissionindicesforacetyle nefuelasafunctionofEGR temperatureatEGR=0.0, =2.0, res =5.0ms. A 4 (EGR=0.0) A 3 (EGR=0.0) A 2 (EGR=0.0) A 4 (EGR=0.0) A 4 (EGR=0.0) A 3 (EGR=0.0) A 2 (EGR=0.0) A 1 (EGR=0.0) =2.0, res =5.0ms Acetylenefuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]EIA i (PAHs)[g/kg]1400 1200 1000 800 600 400 1e+05 1 1e-051e-101e-151e-201e-251e-30 Figure4-79.ComparisonofPAHemissionindicesforacetyle nefuelasafunctionofEGR temperatureatEGR=0.0, =2.0, res =5.0ms. 179

PAGE 180

4.4.3EectofEquivalenceRatio Thevariationofcombustioncharacteristicsandsootprecu rsorswithequivalenceratio (case3insection 3.3 withparameters =0.5-5.0,EGR=0.0and2.0, res =1 : 0,EGRT =1000K)hasbeenmodeledandtheresultspresentedingures 4-80 through 4-85 TherametemperatureforEGR=0.0andEGR=2.0havebeenplott edwithvarying equivalenceratiosingure 4-80 .Surprisingly,theequivalenceratioformaximumrame temperaturewasonfuel-richside( =1 : 5)forEGRof0.0,althoughpointsbetween =1 : 0and =1 : 5werenotmodeled.TheUHCemissions,plottedingure 4-81 increasednon-linearlywithequivalenceratioforEGR=2.0 ,whileithadaminimumat aboutequivalenceratioof2.0forEGR=0.0.TheCOemissions showningure 4-82 reachedamaximumbetweenequivalenceratioof1.5and2.5an dthendecreasedlinearly withincreasingequivalenceratioonthefuel-richside.Th eEIC 2 H 2 valueswereplotted ingure 4-83 .ItcanbeobservedthatthereisasharpriseinEIC 2 H 2 atequivalence ratiosbetween1.0and2.0.Consequently,dependingonathr esholdminimumC 2 H 2 concentrationnecessaryforsootformation,acriticalequ ivalenceratiomaybeidentied belowwhichnosootformationwouldoccur.Moreover,thecri ticalequivalenceratiofor EGR=2.0islikelytobelowerthanthatforEGR=0.0,sincethe EIC 2 H 2 increases rapidlyatlowerequivalenceratiosforEGR=2.0case.Thism aybeexplainedbythe factthatrecirculationoffuel-richexhaustgaseshasaten dencyofincreasingtheeective equivalenceratio,increasingtheoverallsooting.Thecom bustioneciencywasfound todecreasenon-linearlywithequivalenceratio,asshowni ngure 4-84 .Thetrendsalso indicatethecrossoverofcombustioneciencyforEGR=0.0a nd2.0atanequivalence ratioofabout2.0.TheEGR=2.0hadmuchhighercombustione ciencyatequivalence ratioslessthan2.0,whilebeyondthelowercombustioneci encyforhigherequivalence ratiosresultedfromhigherUHCemissions,comparedtoEGR= 0.0case.Figures 4-85 and 4-86 indicatethePAHemissionindicesforEGRof0.0and2.0cases (shownontwo 180

PAGE 181

dierenty-scalesforclarity).TheEIPAHtrendsseemtoclo selyfollowtheEIC 2 H 2 trends,sinceC 2 H 2 isakeyintermediateinthesootformationchain. EGR=2.0 EGR=0.0 EGRT=1000K, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Equivalenceratio( )Flametemperature[K]5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 28002600240022002000180016001400 Figure4-80.Comparisonoframetemperatureforacetylenef uelasafunctionof equivalenceratioatEGR=0.0andEGR=2.0,EGRT=1000K, = 2.0, res =1.0ms,P=202.65kPa(2atm). 181

PAGE 182

EGR=2.0 EGR=0.0 EGRT=1000K, res =5.0ms Acetylenefuel,P=202.65kPa(2atm) Equivalenceratio( )EIUHC[g/kg]5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 1000 100 10 Figure4-81.ComparisonofUHCemissionindexforacetylene fuelasafunctionof equivalenceratioatEGR=0.0andEGR=2.0,EGRT=1000K, = 2.0, res =1.0ms,P=202.65kPa(2atm). EGR=2.0 EGR=0.0 EGRT=1000K, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Equivalenceratio( )EICO[g/kg]5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 10000 1000 100 Figure4-82.ComparisonofCOemissionindexforacetylenef uelasafunctionof equivalenceratioatEGR=0.0andEGR=2.0,EGRT=1000K, = 2.0, res =1.0ms,P=202.65kPa(2atm). 182

PAGE 183

EGR=2.0 EGR=0.0 EGRT=1000K, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Equivalenceratio( )EIC 2 H 2 [g/kg]5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 1e+031e+02 10 1 0.1 Figure4-83.ComparisonofC 2 H 2 emissionindexforacetylenefuelasafunctionof equivalenceratioatEGR=0.0andEGR=2.0,EGRT=1000K, = 2.0, res =1.0ms,P=202.65kPa(2atm). EGR=2.0 EGR=0.0 EGRT=1000K, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Equivalenceratio( )Combustioneciency( c )5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 1.000.900.800.700.600.500.400.300.200.10 Figure4-84.Comparisonofcombustioneciencyforacetyle nefuelasafunctionof equivalenceratioatEGR=0.0andEGR=2.0,EGRT=1000K, = 2.0, res =1.0ms,P=202.65kPa(2atm). 183

PAGE 184

A 4 (EGR2.0) A 3 (EGR2.0) A 2 (EGR2.0) A 1 (EGR2.0) A 4 (EGR0.0) A 3 (EGR0.0) A 2 (EGR0.0) A 1 (EGR0.0) EGRT=1000K, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Equivalenceratio( )EIA i (PAHs)[g/kg]5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 353025201510 50 -5 Figure4-85.ComparisonofPAHemissionindicesforacetyle nefuelasafunctionof equivalenceratioatEGR=0.0andEGR=2.0,EGRT=1000K, = 2.0, res =1.0ms,P=202.65kPa(2atm). 184

PAGE 185

A 4 (EGR2.0) A 3 (EGR2.0) A 2 (EGR2.0) A 1 (EGR2.0) A 4 (EGR0.0) A 3 (EGR0.0) A 2 (EGR0.0) A 1 (EGR0.0) EGRT=1000K, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Equivalenceratio( )EIA i (PAHs)[g/kg]5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 1e+05 1 1e-051e-101e-151e-201e-251e-301e-351e-40 Figure4-86.ComparisonofPAHemissionindicesforacetyle nefuelasafunctionof equivalenceratioatEGR=0.0andEGR=2.0,EGRT=1000K, = 2.0, res =1.0ms,P=202.65kPa(2atm). 185

PAGE 186

4.4.4EectofResidenceTime Theresidencetimedependence(case4insection 3.3 withparameters =0.1-10.0, =2 : 0,EGR=0.0and2.0,EGRT=1000K)forvariouscombustionpara meters hasbeenplottedingures 4-87 through 4-94 .ForthecaseofEGR=0.0,therame temperature,EIUHC,EIC 2 H 2 ,EICOandcombustioneciencyvaluesnearlystabilize inlessthan2ms,whilethoseforthecaseofEGR=2.0donotsta bilizeuntil10ms. FromtheplotsofemissionindicesofPAHprecursors,itcanb eseenthattheEGR= 0.0case(gure 4-92 )wasinoxidizingregime(PAHconcentrationsdecreasewith time) whileEGR=2.0case(gure 4-93 )wasinthesootformationregime(PAHconcentrations increasewithtime).ThedecreaseinPAHconcentrationsinE GR=0.0case,appears toberesultingfromdecreaseinC 2 H 2 concentrations.Ontheotherhand,itseems likelythatthedecreaseinC 2 H 2 concentrationinEGR=2.0casemayberesultingfrom polymerizationreactionsthateventuallyresultinformat ionofaromaticstructures. EGR=2.0 EGR=0.0 EGRT=1000K, =2.0 Acetylenefuel,P=202.65kPa(2atm) Residencetime( res )[ms]Flametemperature[K]10 9 8 7 6 5 4 3 2 1 0 2800260024002200200018001600 Figure4-87.Comparisonoframetemperatureforacetylenef uelasafunctionofresidence timeatEGR=0.0andEGR=2.0, =2.0,EGRT=1000K,P=202.65 kPa(2atm). 186

PAGE 187

EGR=2.0 EGR=0.0 EGRT=1000K, =2.0 Acetylenefuel,P=202.65kPa(2atm) Residencetime( res )[ms]EIUHC[g/kg]10 9 8 7 6 5 4 3 2 1 0 350300250200150100 50 Figure4-88.ComparisonofUHCemissionindexforacetylene fuelasafunctionof residencetimeatEGR=0.0andEGR=2.0, =2.0,EGRT=1000K,P= 202.65kPa(2atm). EGR=2.0 EGR=0.0 EGRT=1000K, =2.0 Acetylenefuel,P=202.65kPa(2atm) Residencetime( res )[ms]EICO[g/kg]10 9 8 7 6 5 4 3 2 1 0 2000190018001700160015001400130012001100 Figure4-89.ComparisonofCOemissionindexforacetylenef uelasafunctionofresidence timeatEGR=0.0andEGR=2.0, =2.0,EGRT=1000K,P=202.65 kPa(2atm). 187

PAGE 188

EGR=2.0 EGR=0.0 EGRT=1000K, =2.0 Acetylenefuel,P=202.65kPa(2atm) Residencetime( res )[ms]EIC 2 H 2 [g/kg]10 9 8 7 6 5 4 3 2 1 0 250200150100 50 0 Figure4-90.ComparisonofC 2 H 2 emissionindexforacetylenefuelasafunctionof residencetimeatEGR=0.0andEGR=2.0, =2.0,EGRT=1000K,P= 202.65kPa(2atm). EGR=2.0 EGR=0.0 EGRT=1000K, =2.0 Acetylenefuel,P=202.65kPa(2atm) Residencetime( res )[ms]Combustioneciency( c )10 9 8 7 6 5 4 3 2 1 0 0.480.470.460.450.440.430.420.410.40 Figure4-91.Comparisonofcombustioneciencyforacetyle nefuelasafunctionof residencetimeatEGR=0.0andEGR=2.0, =2.0,EGRT=1000K,P= 202.65kPa(2atm). 188

PAGE 189

A 4 A 3 A 2 A 1 EGR=0.0,EGRT=1000K, =2.0 Acetylenefuel,P=202.65kPa(2atm) Residencetime( res )[ms]EIA i (PAHs)[g/kg]10 9 8 7 6 5 4 3 2 1 0 1 1e-051e-101e-151e-201e-251e-301e-35 Figure4-92.ComparisonofPAHemissionindicesforacetyle nefuelasafunctionof residencetimeatEGR=0.0, =2.0,EGRT=1000K,P=202.65kPa(2 atm). A 4 A 3 A 2 A 1 EGR=2.0,EGRT=1000K, =2.0 Acetylenefuel,P=202.65kPa(2atm) Residencetime( res )[ms]EIA i (PAHs)[g/kg]10 9 8 7 6 5 4 3 2 1 0 1e+02 10 1 0.1 0.01 Figure4-93.ComparisonofPAHemissionindicesforacetyle nefuelasafunctionof residencetimeatEGR=2.0, =2.0,EGRT=1000K,P=202.65kPa(2 atm). 189

PAGE 190

A 4 (EGR=2.0) A 3 (EGR=2.0) A 2 (EGR=2.0) A 1 (EGR=2.0) A 4 (EGR=0.0) A 3 (EGR=0.0) A 2 (EGR=0.0) A 1 (EGR=0.0) EGRT=1000K, =2.0 Acetylenefuel,P=202.65kPa(2atm) Residencetime( res )[ms]EIA i (PAHs)[g/kg]10 9 8 7 6 5 4 3 2 1 0 1e+05 1 1e-051e-101e-151e-201e-251e-301e-35 Figure4-94.ComparisonofPAHemissionindicesforacetyle nefuelasafunctionof residencetimeatEGR=0.0andEGR=2.0, =2.0,EGRT=1000K,P= 202.65kPa(2atm). 190

PAGE 191

4.5CaseStudy5:SootParticleGrowthandOxidationinVitia ted Combustion TheformationofparticulatesootinpresenceofEGRwasinve stigatedandthe resultshavebeenpresentedinthissection.Figure 4-95 depictstheeectofEGRonsoot formation.Itwasfoundthatthesootvolumefractionincrea sedwithEGR,fortherame temperaturesbetween1500to2000K.Fromtheplotofsootpar ticlediameterversus therametemperatureshowningure 4-96 ,itmaybeinferredthattheparticlediameter increasedwithEGR,andthattherewasanincreasingdominan ceofthesootgrowth and/orcoagulativeprocessesathighertemperatures(betw een1500-2000K).Figure 4-97 showstheplotofsootsurfaceareawithrespecttorametempe rature.Anincreaseof thesootsurfaceareaatlowertemperaturemaybeattributed tothenucleationofsoot particles,andanincreaseinthesootparticlesizesduetot hegrowthprocesses.Itis well-knownthatoverabout95%oftheobservedsootvolumein creasemaybeattributed tothesootsurfacechemistry[ 161 ].Thedecreaseinthesootsurfaceareaathigher temperaturesmaybeattributedtothecoagulationofsootpa rticles. Thevariationoftheabsolutevaluesofzero th momentgenerationratesfortheparticle distributionwithrametemperatureareplottedingure 4-98 andgure 4-99 ,forEGR 0.0and2.0respectively.Itrepresentstherateofchangeof thenumberdensityofthesoot particles.Itmaybeobservedthatthecontributionofbotht henucleation(positiveeect) andcoagulationprocesses(negativeeect)tothechangein numberdensitiesishigh, whilesurfacegrowthhasnoeectonparticlenumberdensiti es.Theparticlesnucleateat amuchlargerrate,however,theycoagulatetogrowinsize,w ithaconsequentdecrease innumberdensities.Asaresult,theneteectisamuchlower rateofincreaseinnumber densities.Acomparisonoftheabsolutevaluesofthezero th momentgenerationrates forEGRof0.0and2.0hasbeenpresentedingure 4-99 .Surprisingly,asuppression ofnucleationandcoagulationrateswithEGRwasobserved,d espitetheincreasein sootvolumefraction.Thismayberationalizedbythefactth atmostoftheincreasein 191

PAGE 192

EGR2.0 EGR0.0 =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]Sootvolumefraction( f v )2000 1900 1800 1700 1600 1500 6e-075e-074e-073e-072e-071e-07 0 Figure4-95.Comparisonofsootvolumefractionforacetyle nefuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0, =2.0, res =1.0ms,P= 202.65kPa(2atm). EGR2.0 EGR0.0 =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]Averagesootparticlediameter( D soot )[nm]2000 1900 1800 1700 1600 1500 1210 86420 Figure4-96.Comparisonofaveragesootparticlediameterf oracetylenefuelasafunction oframetemperatureatEGR=0.0andEGR=2.0, =2.0, res =1.0ms, P=202.65kPa(2atm). 192

PAGE 193

sootvolumefractioncomesfromthesurfacechemistry.Onth eotherhand,thesurface chemistry,hasnoeectonparticlepopulation,whileitdoe saectthesize. EGR2.0 EGR0.0 =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]Sootparticlesurfacearea( A soot )[ cm 3 m 3 ]2000 1900 1800 1700 1600 1500 1.41.2 1 0.80.60.40.2 0 Figure4-97.Comparisonofaveragesootparticlesurfacear eaforacetylenefuelasa functionoframetemperatureatEGR=0.0andEGR=2.0, =2.0, res = 1.0ms,P=202.65kPa(2atm). Therateofchangeoftotalmassofsootparticleshasbeencha racterizedbytheabsolute valuesoftherstmomentgenerationrates,plottedwithres pecttorametemperatures ingure 4-101 andgure 4-102 forEGR=0.0andEGR=2.0,respectively.Itmaybe observedthatmajorityofthesurfacegrowthmaybeattribut edtosurfacechemistry,and coagulationprocessesplaynorole.Oncomparingtheserate sforEGR=0.0andEGR= 2.0,showningure 4-103 ,wefoundthatsurfacechemistrywasreducedwithEGR.The naturalquestiontoaskatthispointthenwouldbethat,what wouldcauseanincreasein sootvolumefraction,iftheratesofnucleation,coagulati on,aswellassurfacechemistry reducedwithEGR?Onacloserinspection,itmaybeobservedt hatthesootformation processwasbeingcontrolledbytheresidencetimeofthefre shreactantstream.Inorder toholdthetotalresidencetimexed,allthemassrowsweres caledwithincreasing 193

PAGE 194

EGR.Thisnecessitatedreductioninfreshfuelandairstrea mmassrows,increasingthe residencetimeoffreshreactantsfromthisstream.Giventh eassumptionofhomogeneity ofthemodel,itmaybereasonedthattheEGRcompositionwasn earlythesameasthe productcomposition.Asaresult,onedominanteectofincr easingEGR,foragiven constantrametemperaturewouldbeincreasingtheresidenc etimeofthefreshreactants. FromresultspresentedinBrownetal.[ 159 ],itcanbeseenthatthesootvolumefraction increaseswithresidencetime,providedthatthetemperatu reswerewithinthecritical limitsofthesootformationregime.Therefore,despiteasu ppressionintheratesofthe sootgrowthprocesses,theincreasedresidencetimeofthef reshreactantstreamdueto increasedEGR,ultimatelyresultsinhighersootvolumefra ctions. Surfacegrowthrate( j W 0 j ) Coagulationrate( j G 0 j ) Nucleationrate( j R 0 j ) Totalmomentproductionrate( j dM 0 dt j ) =2.0,EGR=0.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]Zero th sootmomentproductionrate( dM 0 dt )[ cm 3 s 1 ]2000 1900 1800 1700 1600 1500 4e+16 3.5e+16 3e+16 2.5e+16 2e+16 1.5e+16 1e+165e+15 0 Figure4-98.Comparisonofzero th sootmomentproductionratecontributions(nucleation, coagulation,surfacegrowth)foracetylenefuelasafuncti onoframe temperatureat =2.0,EGR=0.0, res =1.0ms,P=202.65kPa(2atm). 194

PAGE 195

Surfacegrowthrate( j W 0 j ) Coagulationrate( j G 0 j ) Nucleationrate( j R 0 j ) Totalmomentproductionrate( j dM 0 dt j ) =2.0,EGR=2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) FlameTemperature[K]Zero th sootmomentproductionrate( dM 0 dt )[ cm 3 s 1 ]2000 1900 1800 1700 1600 1500 4e+16 3.5e+16 3e+16 2.5e+16 2e+16 1.5e+16 1e+165e+15 0 Figure4-99.Comparisonofzero th sootmomentproductionratecontributions(nucleation, coagulation,surfacegrowth)foracetylenefuelasafuncti onoframe temperatureat =2.0,EGR=2.0, res =1.0ms,P=202.65kPa(2atm). j W 0 j (EGR=2.0) j G 0 j (EGR=2.0) j R 0 j (EGR=2.0) j dM 0 dt j (EGR=2.0) j W 0 j (EGR=0.0) j G 0 j (EGR=0.0) j R 0 j (EGR=0.0) j dM 0 dt j (EGR=0.0) =2.0,EGR=0.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]Zero th sootmomentproductionrate( dM 0 dt )[ cm 3 s 1 ]2000 1900 1800 1700 1600 1500 3e+16 2.5e+16 2e+16 1.5e+16 1e+165e+15 0 Figure4-100.Comparisonofzero th sootmomentproductionratecontributions (nucleation,coagulation,surfacegrowth)foracetylenef uelasafunctionof rametemperatureatEGR=0.0andEGR=2.0, =2.0, res =1.0ms,P =202.65kPa(2atm). 195

PAGE 196

j W 1 j j G 1 j j R 1 j j dM 1 dt j =2.0,EGR=0.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]Firstsootmomentproductionrate( dM 1 dt )[ cm 3 s 1 ]2000 1900 1800 1700 1600 1500 1.2e+19 1e+198e+186e+184e+182e+18 0 Figure4-101.Comparisonofrstsootmomentproductionrat econtributions(nucleation, coagulation,surfacegrowth)foracetylenefuelasafuncti onoframe temperatureat =2.0,EGR=0.0, res =1.0ms,P=202.65kPa(2atm). j W 1 j j G 1 j j R 1 j j dM 1 dt j =2.0,EGR=0.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) FlameTemperature[K]Firstsootmomentproductionrate( dM 1 dt )[ cm 3 s 1 ]2000 1900 1800 1700 1600 1500 1e+199e+188e+187e+186e+185e+184e+183e+182e+181e+18 0 Figure4-102.Comparisonofrstsootmomentproductionrat econtributions(nucleation, coagulation,surfacegrowth)foracetylenefuelasafuncti onoframe temperatureat =2.0,EGR=2.0, res =1.0ms,P=202.65kPa(2atm). 196

PAGE 197

j W 1 j (EGR=2.0) j G 1 j (EGR=2.0) j R 1 j (EGR=2.0) j dM 1 dt j (EGR=2.0) j W 1 j (EGR=0.0) j G 1 j (EGR=.0) j R 1 j (EGR=0.0) j dM 1 dt j (EGR=0.0) =2.0,EGR=0.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]Firstsootmomentproductionrate( dM 1 dt )[ cm 3 s 1 ]2000 1900 1800 1700 1600 1500 1.2e+19 1e+198e+186e+184e+182e+18 0 Figure4-103.Comparisonofrstsootmomentproductionrat econtributions(nucleation, coagulation,surfacegrowth)foracetylenefuelasafuncti onoframe temperatureatEGR=0.0andEGR=2.0, =2.0, res =1.0ms,P= 202.65kPa(2atm). ThePAHconcentrationswereplottedasfunctionsoframetem peratureandEGR, asshowningure 4-104 throughgure 4-106 .WhilethePAHconcentrationsdecreased towardshighertemperatures,nogeneraltrendscouldbeinf erredabouttheeectofEGR, forcomparisonsatconstantrametemperatures. TheabsolutevaluesoftheeectivecontributionsofC 2 H 2 ,PAH,OHandO 2 to surfacegrowthrate(negativeforOHandO 2 )wereplottedwithrespecttotherame temperatureandEGRingure 4-107 throughgure 4-108 .Itwasfoundthatmaximum contributiontothesootgrowthwasduetoC 2 H 2 additionreactions.Thereactionsdue toOHoxidationwerefoundtobemoreimportantathighertemp eratures,whilethe PAHcondensationreactionswerefoundtobeactiveatlowert emperatures.Theeect ofEGR,onsootsurfacechemistrycouldbesummarizedasthat ofloweringofgrowth 197

PAGE 198

A 4 A 3 A 2 A 1 =2.0,EGR=0.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]EI A i (PAHs)[g/kg]2000 1900 1800 1700 1600 1500 0.001 0.0001 1e-051e-061e-071e-08 Figure4-104.ComparisonofA i (PAHs)emissionindicesforacetylenefuelasafunctionof rametemperatureat =2.0,EGR=0.0, res =1.0ms,P=202.65kPa (2atm). A 4 A 3 A 2 A 1 =2.0,EGR=2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]EIA i (PAHs)[g/kg]2000 1900 1800 1700 1600 1500 0.001 0.0001 1e-051e-061e-071e-08 Figure4-105.ComparisonofA i (PAHs)emissionindicesforacetylenefuelasafunctionof rametemperatureat =2.0,EGR=2.0, res =1.0ms,P=202.65kPa (2atm). 198

PAGE 199

A 4 (EGR=2.0) A 3 (EGR=2.0) A 2 (EGR=2.0) A 1 (EGR=2.0) A 4 (EGR=0.0) A 3 (EGR=0.0) A 2 (EGR=0.0) A 1 (EGR=0.0) =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]EI A i (PAHs)[g/kg]2000 1900 1800 1700 1600 1500 0.001 0.0001 1e-051e-061e-071e-08 Figure4-106.ComparisonofA i (PAHs)emissionindicesforacetylenefuelasafunctionof rametemperatureat =2.0,EGR=0.0andEGR=2.0, res =1.0ms,P =202.65kPa(2atm). ratesduetoacetyleneinthetemperaturesbetween1500-190 0K,ashiftinthepeakof acetylenecontributions,towardstheright,andslightlyh igheroxidationratesonthehigh temperatureend.Therefore,thelowersootgrowthratesint helowertemperatureside, weremainlyduetosuppressionacetylenereactions,whilet heincreasedratesonthehigh temperaturesideweredueacompetitionbetweenincreasedo xidationandC 2 H 2 addition reactions. Theplotingure 4-110 showsthatthetrendsinsootparticlemassdensitywith rametemperatureandEGR.ThecorrelationsforSNwerebased ontherecommendations ofColketIIIetal.[ 141 ].However,unrealisticestimatesofSN(greaterthan100)w ere obtained,forthecalculatedsootparticlemassdensitiesa ndhence,arenotshowninthe plots. 199

PAGE 200

OHrate( j W OH j ) O 2 rate( j W O 2 j ) C 2 H 2 rate( j W C 2 H 2 j ) PAHcondensationrate( j W PAH j ) Surfacegrowthrate( j W 1 j ) EGR=0.0, =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]Surfacegrowthrate( j W 1 j )[ cm 3 s 1 ]2000 1900 1800 1700 1600 1500 1e+201e+191e+181e+171e+16 Figure4-107.Comparisonofsurfacegrowthrateforacetyle nefuelasafunctionoframe temperatureat =2.0,EGR=0.0, res =1.0ms,P=202.65kPa(2atm). OHrate( j W OH j ) O 2 rate( j W O 2 j ) C 2 H 2 rate( j W C 2 H 2 j ) PAHcondensationrate( j W PAH j ) Surfacegrowthrate( j W 1 j ) EGR=2.0, =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]Surfacegrowthrate( j W 1 j )[ cm 3 s 1 ]2000 1900 1800 1700 1600 1500 1e+201e+191e+181e+171e+16 Figure4-108.Comparisonofsurfacegrowthratecontributi ons(PAHcondensation,C 2 H 2 addition,O 2 oxidation,OHoxidation)foracetylenefuelasafunctionof rametemperatureat =2.0,EGR=2.0, res =1.0ms,P=202.65kPa(2 atm). 200

PAGE 201

j W OH j (EGR=2.0) j W O 2 j (EGR=2.0) j W C 2 H 2 j (EGR=2.0) j W PAH j (EGR=2.0) j W 1 j (EGR=2.0) j W OH j (EGR=0.0) j W O 2 j (EGR=0.0) j W C 2 H 2 j (EGR=0.0)' j W PAH j (EGR=0.0) j W 1 j (EGR=0.0) EGR=2.0, =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]Surfacegrowthrate( j W 1 j )[ cm 3 s 1 ]2000 1900 1800 1700 1600 1500 1e+191e+181e+17 Figure4-109.Comparisonofsurfacegrowthratecontributi ons(PAHcondensation,C 2 H 2 addition,O 2 oxidation,OHoxidation)andOHoxidationreactionsfor acetylenefuelasafunctionoframetemperatureat =2.0,EGR=0.0and EGR=2.0, res =1.0ms,P=202.65kPa(2atm). EGR=2.0 EGR=0.0 =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]Sootmassdensity( soot )[ mg m 3 ]2000 1900 1800 1700 1600 1500 1e+041e+031e+02 10 Figure4-110.Comparisonofsootmassdensityforacetylene fuelasafunctionoframe temperatureat =2.0,EGR=2.0, res =1.0ms,P=202.65kPa(2atm). 201

PAGE 202

4.6ComparisonofFrenklachandRichterModelsofSootForma tion AcomparisonoftheresultsbasedonthemodelsfromFrenklac handcoworkersand thatfromtheRichteretal.havebeenpresentedingures 4-111 through 4-124 .Itmay beseenthattherewasgenerallyaverygoodagreementinEIUH C,EICO,combustion ecienciesobtainedfromthemodels,plottedingure 4-111 ,gure 4-112 andgure 4-114 respectively.Theresultsfromthefrenklachmodel,seemed tooverpredictthevaluesforEI C 2 H 2 ,asseeningure 4-113 .ThecomparisonofthePAHspecies,namelybenzene(A 1 ), naphthalene(A 2 ),anthracene(A 3 )andpyrene(A 4 ),predictedfrombothmodelshasbeen presentedingures 4-115 through 4-118 .Theseseemtoagreeprettywell. TheparticlebindistributionsforEGR=0.0andEGR=2.0were plottedfor temperatures1500-2000K,andareshowningures 4-119 through 4-124 .Thegrowthof particlediameterspredictedbytheFrenklachmodelingur e 4-96 ,maybeobservedas theincreaseintheconcentrationofhighermassspecies,pr edictedbytheRichtermodelin gures 4-119 through 4-124 .Wefoundgoodagreementfromthegeneraltrendsfromboth themodels. Frenklachmech.(EGR=2.0) Richtermech.(EGR=2.0) Frenklachmech.(EGR=0.0) Richtermech.(EGR=0.0) =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]EIUHC[g/kg]2000 1900 1800 1700 1600 1500 360340320300280260240220200180160 Figure4-111.ComparisonofUHCemissionindexforacetylen efuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandFrenklachmechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). 202

PAGE 203

Frenklachmech.(EGR=2.0) Richtermech.(EGR=2.0) Frenklachmech.(EGR=0.0) Richtermech.(EGR=0.0) =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]EICO[g/kg]2000 1900 1800 1700 1600 1500 1600150014001300120011001000 Figure4-112.ComparisonofCOemissionindexforacetylene fuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandFrenklachmechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). 203

PAGE 204

Frenklachmech.(EGR=2.0) Richtermech.(EGR=2.0) Frenklachmech.(EGR=0.0) Richtermech.(EGR=0.0) =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) FlameTemperature[K]EI C 2 H 2 [g/kg]2000 1900 1800 1700 1600 1500 300280260240220200180160140120100 Figure4-113.ComparisonofC 2 H 2 emissionindexforacetylenefuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandFrenklachmechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). Frenklachmech.(EGR=2.0) Richtermech.(EGR=2.0) Frenklachmech.(EGR=0.0) Richtermech.(EGR=0.0) =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]Combustioneciency( c )2000 1900 1800 1700 1600 1500 0.490.480.470.460.450.440.430.420.410.400.39 Figure4-114.Comparisonofcombustioneciencyforacetyl enefuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandFrenklachmechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). 204

PAGE 205

Frenklachmech.(EGR=2.0) Richtermech.(EGR=2.0) Frenklachmech.(EGR=0.0) Richtermech.(EGR=0.0) =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]EI A 1 [g/kg]2000 1900 1800 1700 1600 1500 1e+02 10 1 0.1 0.01 Figure4-115.ComparisonofA 1 emissionindexforacetylenefuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandFrenklachmechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). Frenklachmech.(EGR=2.0) Richtermech.(EGR=2.0) Frenklachmech.(EGR=0.0) Richtermech.(EGR=0.0) =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]EI A 2 [g/kg]2000 1900 1800 1700 1600 1500 10 1 0.1 0.01 0.001 0.0001 1e-05 Figure4-116.ComparisonofA 2 emissionindexforacetylenefuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandFrenklachmechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). 205

PAGE 206

Frenklachmech.(EGR=2.0) Richtermech.(EGR=2.0) Frenklachmech.(EGR=0.0) Richtermech.(EGR=0.0) =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]EI A 3 [g/kg]2000 1900 1800 1700 1600 1500 1 0.1 0.01 0.001 0.0001 1e-051e-061e-07 Figure4-117.ComparisonofA 3 emissionindexforacetylenefuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandFrenklachmechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). Frenklachmech.(EGR=2.0) Richtermech.(EGR=2.0) Frenklachmech.(EGR=0.0) Richtermech.(EGR=0.0) =2.0, res =1.0ms Acetylenefuel,P=202.65kPa(2atm) Flametemperature[K]EI A 4 [g/kg]2000 1900 1800 1700 1600 1500 0.1 0.01 0.001 0.0001 1e-051e-061e-071e-081e-091e-101e-11 Figure4-118.ComparisonofA 4 emissionindexforacetylenefuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0(RichtermechanismandFrenklachmechanism), =2.0, res =1.0ms,P=202.65kPa(2atm). 206

PAGE 207

EGR=2.0 EGR=0.0 Acetylenefuel,P=202.65kPa(2atm),EGRT=1500K, =2.0, res =1.0ms BinsMassfraction BIN20J BIN20 BIN19J BIN19 BIN18J BIN18 BIN17J BIN17 BIN16J BIN16 BIN15J BIN15 BIN14J BIN14 BIN13J BIN13 BIN12J BIN12 BIN11J BIN11 BIN10J BIN10 BIN9J BIN9 BIN8J BIN8 BIN7J BIN7 BIN6J BIN6 BIN5J BIN5 BIN4J BIN4 BIN3J BIN3 BIN2J BIN2 BIN1J BIN10.0001 1e-061e-081e-101e-121e-141e-161e-181e-20 Figure4-119.Comparisonofbinmassfractionsforacetylen efuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0,EGRT=1500K, =2.0, res =1.0ms,P=202.65kPa(2atm). EGR=2.0 EGR=0.0 Acetylenefuel,P=202.65kPa(2atm),EGRT=1600K, =2.0, res =1.0ms BinsMassfraction BIN20J BIN20 BIN19J BIN19 BIN18J BIN18 BIN17J BIN17 BIN16J BIN16 BIN15J BIN15 BIN14J BIN14 BIN13J BIN13 BIN12J BIN12 BIN11J BIN11 BIN10J BIN10 BIN9J BIN9 BIN8J BIN8 BIN7J BIN7 BIN6J BIN6 BIN5J BIN5 BIN4J BIN4 BIN3J BIN3 BIN2J BIN2 BIN1J BIN10.0001 1e-061e-081e-101e-121e-141e-16 Figure4-120.Comparisonofbinmassfractionsforacetylen efuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0,EGRT=1600K, =2.0, res =1.0ms,P=202.65kPa(2atm). 207

PAGE 208

EGR=2.0 EGR=0.0 Acetylenefuel,P=202.65kPa(2atm),EGRT=1700K, =2.0, res =1.0ms BinsMassfraction BIN20J BIN20 BIN19J BIN19 BIN18J BIN18 BIN17J BIN17 BIN16J BIN16 BIN15J BIN15 BIN14J BIN14 BIN13J BIN13 BIN12J BIN12 BIN11J BIN11 BIN10J BIN10 BIN9J BIN9 BIN8J BIN8 BIN7J BIN7 BIN6J BIN6 BIN5J BIN5 BIN4J BIN4 BIN3J BIN3 BIN2J BIN2 BIN1J BIN10.0001 1e-051e-061e-071e-081e-091e-101e-111e-121e-13 Figure4-121.Comparisonofbinmassfractionsforacetylen efuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0,EGRT=1700K, =2.0, res =1.0ms,P=202.65kPa(2atm). EGR2.0 EGR0.0 Acetylenefuel,P=202.65kPa(2atm),EGRT=1800K, =2.0, res =1.0ms BinsMassfraction BIN20J BIN20 BIN19J BIN19 BIN18J BIN18 BIN17J BIN17 BIN16J BIN16 BIN15J BIN15 BIN14J BIN14 BIN13J BIN13 BIN12J BIN12 BIN11J BIN11 BIN10J BIN10 BIN9J BIN9 BIN8J BIN8 BIN7J BIN7 BIN6J BIN6 BIN5J BIN5 BIN4J BIN4 BIN3J BIN3 BIN2J BIN2 BIN1J BIN10.0001 1e-051e-061e-071e-081e-091e-101e-11 Figure4-122.Comparisonofbinmassfractionsforacetylen efuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0,EGRT=1800K, =2.0, res =1.0ms,P=202.65kPa(2atm). 208

PAGE 209

EGR=2.0 EGR=0.0 Acetylenefuel,P=202.65kPa(2atm),EGRT=1900K, =2.0, res =1.0ms BinsMassfraction BIN20J BIN20 BIN19J BIN19 BIN18J BIN18 BIN17J BIN17 BIN16J BIN16 BIN15J BIN15 BIN14J BIN14 BIN13J BIN13 BIN12J BIN12 BIN11J BIN11 BIN10J BIN10 BIN9J BIN9 BIN8J BIN8 BIN7J BIN7 BIN6J BIN6 BIN5J BIN5 BIN4J BIN4 BIN3J BIN3 BIN2J BIN2 BIN1J BIN11e-051e-061e-071e-081e-091e-10 Figure4-123.Comparisonofbinmassfractionsforacetylen efuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0,EGRT=1900K, =2.0, res =1.0ms,P=202.65kPa(2atm). EGR2.0 EGR0.0 Acetylenefuel,P=202.65kPa(2atm),EGRT=2000K, =2.0, res =1.0ms BinsMassfraction BIN20J BIN20 BIN19J BIN19 BIN18J BIN18 BIN17J BIN17 BIN16J BIN16 BIN15J BIN15 BIN14J BIN14 BIN13J BIN13 BIN12J BIN12 BIN11J BIN11 BIN10J BIN10 BIN9J BIN9 BIN8J BIN8 BIN7J BIN7 BIN6J BIN6 BIN5J BIN5 BIN4J BIN4 BIN3J BIN3 BIN2J BIN2 BIN1J BIN11e-061e-071e-081e-091e-101e-111e-121e-13 Figure4-124.Comparisonofbinmassfractionsforacetylen efuelasafunctionoframe temperatureatEGR=0.0andEGR=2.0,EGRT=2000K, =2.0, res =1.0ms,P=202.65kPa(2atm). 209

PAGE 210

4.7CaseStudy6:ModelingofSemi-ClosedCycleEngineChara cteristics usingaKineticallyEquivalentUnvitiatedOpen-CycleCSTR System TheanalysisfordeterminingakineticallyequivalentOCsy stemforagivenSCC system,hasbeendiscussedinthissection.Themainassumpt ionsfortheanalysisare: 1.Idealgasmixtureisassumed2.volumeandpressureareconstant3.spatiallyhomogeneoussystem4.thereactingmixtureisinsteadystate4.7.1SemiclosedCase ForaCSTRinsteadystateasshowningure 3-7A ,thespeciesconservationacross controlvolumeCV1maybewrittengivenbyequation 4{1 dm j dt CV =(_ m F Y F;j +_ m A Y A 2 ;j +_ m R Y R;j m ex 1 Y ex 1 ;j )+ M j V R j =0(4{1) dm dt CV =(_ m F +_ m A 2 +_ m R ) m ex 1 =0(4{2) V R h S ( T ex ) dt CV =(_ m F h F;S +_ m A h A 2 ;S +_ m R h R;S ) V R Q gen (_ m ex 1 h ex 1 ;S )=0(4{3) where V R Q gen = P n S j =0 V R h F;j ( T 0 )_ j ( T ex ) Thetotalenthalpy h maybecalculatedasasumofsensibleenthalpy h S andenthalpy offormation h F ( T 0 ),givenbyequation 4{4 h = h ( T )= h S ( T )+ h F ( T 0 )= n S X j =0 Y j h f;j + Z T T 0 c P;j ( T ) dT (4{4) 4.7.2OpenCycleCase TheconservationequationsforanOCsystemaregivenbyequa tionsequation 4{5 throughequation 4{7 210

PAGE 211

dm j dt CV =(_ m F Y F;j +_ m A Y A 1 ;j m ex Y ex 2 ;j )+ M j V R j =0(4{5) dm dt CV =(_ m F +_ m A ) m ex 2 =0(4{6) V R h S ( T ex ) dt CV =(_ m F h F;S +_ m A h A;S ) V R Q gen Q L (_ m ex 2 h ex 2 ;S )=0(4{7) 4.7.3MatchingConditions ThematchingconditionsforakineticallyequivalentOCare basedonamatchinthe exitenthalpiesandspeciesproductionrates.Fourequival entsystemswereanalyzed,as discussedinsection 3.3.6 4.7.3.1CaseA:AnadiabaticOCsystemequivalenttoanSCCsy stem,with lowerreactantinlettemperaturesandlargerresidencetim ewith matchedfreshreactantrows AnadiabaticOCsystemequivalenttoanSCCsystemasshownin gure 3-7B ,with lowerinlettemperaturesandlargerresidencetimewithmat chfreshreactantrowsmay beusedtosimulatethesteadystatetemperaturesofanSCCsy stem.Thefollowing assumptionsweremade: [ P ] SCC =[ P ] OC ;[ V R ] SCC =[ V R ] OC [_ m F ] SCC =[_ m F ] OC ;[_ m A ] SCC =[_ m A ] OC [_ j ] SCC =[_ j ] OC ; h Q gen i SCC = h Q gen i OC [ h ex 1 ] SCC =[ h ex 2 ] OC ;[ h F;S ] SCC =[ h F;S ] OC [ Y j;F ] SCC =[ Y j;F ] OC ;[ Y j;A ] SCC =[ Y j;A ] OC ;[ Y j;ex 1 ] SCC =[ Y j;ex 2 ] OC ;[ Y j;R ] SCC = [ Y j;ex 1 ] SCC Thescalingfactorforthetotalmassrows,denedby r A ,isgivenbyequation 4{8 Matchingoftheexitenthalpies,obtainedfromtheenergyco nservationgivesequation 4{9 r A = [_ m ex 2 ] OC [_ m ex 1 ] SCC = [_ m F +_ m A ] OC [_ m F +_ m A +_ m R ] SCC (4{8) 211

PAGE 212

_ m A ( h A;S ) SCC ( h A;S ) OC =_ m R ( h ex 1 ;S ) SCC ( h R;S ) SCC (4{9) Thisallowscalculationoftheinletairenthalpyusingequa tion 4{10 .Thespecies balanceisautomaticallysatisedforthegivensimplifyin gassumptions. ( T A ) OC =( T A ) SCC EGR ( h ex 1 ;S ) SCC ( h R;S ) SCC C P;A (4{10) 4.7.3.2CaseB:AnadiabaticOCsystemequivalenttoanSCCsy stem,with lowerreactantinlettemperaturesandmatchedresidenceti me AnadiabaticOCsystemequivalenttoanSCCsystemasshownin gure 3-7B ,with lowerreactantinlettemperaturesandmatchedresidenceti memaybeusedtosimulatethe steadystatetemperaturesofanSCCsystem.Thefollowingas sumptionsweremade: [ P ] SCC =[ P ] OC ;[ V R ] SCC =[ V R ] OC [_ m ex 1 ] SCC =[_ m ex 2 ] OC [_ j ] SCC =[_ j ] OC ; h Q gen i SCC = h Q gen i OC [ h ex 1 ] SCC =[ h ex 2 ] OC ;[ h F;S ] SCC =[ h F;S ] OC [ Y j;F ] SCC =[ Y j;F ] OC ;[ Y j;A ] SCC =[ Y j;A ] OC ;[ Y j;ex 1 ] SCC =[ Y j;ex 2 ] OC ;[ Y j;R ] SCC = [ Y j;ex 1 ] SCC Thescalingfactorforthetotalmassrows,denedby r A ,isgivenbyequation 4{11 Hencethescalingoffreshairmassrowsmaybecalculatedfro mequation 4{12 r A = [_ m ex 2 ] OC [_ m ex 1 ] SCC =1(4{11) r B = [_ m A ] OC [_ m A ] SCC = h st: +1+ EGR i SCC h st: +1 i OC (4{12) Usingthedenitionsof r A and r B ,andmatchingthespeciesbalancesforOCand SCCsystems,givesequation 4{13 whichfurthersimpliestoequation 4{14 212

PAGE 213

_ m R Y R;j +[(_ m F ) SCC (_ m F ) OC ] Y F;j +[(_ m A ) SCC (_ m A ) OC ] Y A;j =0(4{13) EGR ( Y R;j )+ Y F;j st: + Y A;j (1 r B )=0(4{14) Thisimpliesthatthedecreaseinthespeciesmassfractions fromfreshairand fuelmassrowsintheSCCsystem,isequaltothemassfraction saddedfromthefresh stream.Thisequationisautomaticallysatisedfromtheto talmassbalance,forthegiven assumptions.Matchingoftheexitenthalpies,obtainedfro mtheenergyconservationgives equation 4{15 (_ m F h F;S ) SCC (_ m F h F;S ) OC + (_ m A h A;S ) SCC (_ m A h A;S ) OC +_ m R ( h R;S ) SCC =0(4{15) ThismeansthattheenergysuppliedbytheEGRstreammustbeb alancedo bytheinletethalpiesofthefuelandairstreamsintheOC.Fu rthersimplication allowscalculationoftheinletairenthalpyusingequation 4{16 .Thespeciesbalanceis automaticallysatisedforthegivensimplifyingassumpti ons. ( T A ) OC = h ( h A;S ) SCC + st: (1 r B )( h F;S ) SCC i + EGR ( h R;S ) SCC r B C P;A (4{16) 4.7.3.3CaseC:Anon-adiabaticOCsystemequivalenttoanSC Csystem, withapositiveheatlossandlargerresidencetimewithmatc hedfresh reactantrows. AnadiabaticOCsystemequivalenttoanSCCsystemasshownin gure 3-7C ,witha positiveheatlossandlargerresidencetimewithmatchedfr eshreactantrowsmaybeused tosimulatethesteadystatetemperaturesofanSCCsystem.T hefollowingassumptions weremade: [ P ] SCC =[ P ] OC ;[ V R ] SCC =[ V R ] OC 213

PAGE 214

[_ m F ] SCC =[_ m F ] OC ;[_ m A ] SCC =[_ m A ] OC [_ j ] SCC =[_ j ] OC ; h Q gen i SCC = h Q gen i OC [ h ex 1 ] SCC =[ h ex 2 ] OC ;[ h F;S ] SCC =[ h F;S ] OC [ Y j;F ] SCC =[ Y j;F ] OC ;[ Y j;A ] SCC =[ Y j;A ] OC ;[ Y j;ex 1 ] SCC =[ Y j;ex 2 ] OC ;[ Y j;R ] SCC = [ Y j;ex 1 ] SCC [ h F;S ] SCC =[ h F;S ] OC ;[ h A;S ] SCC =[ h A;S ] OC thesystemisnon-adiabatic Thescalingfactorforthetotalmassrows,denedby r A ,isgivenbyequation 4{17 Matchingoftheexitenthalpies,obtainedfromtheenergyco nservationgivesequation 4{18 whichonfurthersimplicationyieldsequation 4{19 r A = [_ m ex 2 ] OC [_ m ex 1 ] SCC = [_ m F +_ m A ] OC [_ m F +_ m A +_ m R ] SCC (4{17) Q L +_ m R h R;S =[(_ m ex 1 ) SCC (_ m ex 2 ) OC ] h ex 1 ;S =[_ m ex 1 ] SCC (1 r A ) h ex 1 ;S =_ m R h ex 1 ;S (4{18) Q L =_ m R [ h ex 1 ;S h R;S ](4{19) Thisshowsthattheheatlossthroughthereactorwallsofthe OCsystemmust equaltheheatlossintheEGRstream,asshowningure 3-7A .Thespeciesbalanceis automaticallysatisedforthegivensimplifyingassumpti ons. 4.7.3.4CaseD:Anon-adiabaticOCsystemequivalenttoanSC Csystem, withapositiveheatlossandmatchedresidencetime AnadiabaticOCsystemequivalenttoanSCCsystemasshownin gure 3-7C ,with apositiveheatlossandmatchedresidencetimemaybeusedto simulatethesteadystate temperaturesofanSCCsystem.Thefollowingassumptionswe remade: [ P ] SCC =[ P ] OC ;[ V R ] SCC =[ V R ] OC 214

PAGE 215

[_ m ex 1 ] SCC =[_ m ex 2 ] OC [_ j ] SCC =[_ j ] OC ; h Q gen i SCC = h Q gen i OC [ h ex 1 ] SCC =[ h ex 2 ] OC ;[ h F;S ] SCC =[ h F;S ] OC [ Y j;F ] SCC =[ Y j;F ] OC ;[ Y j;A ] SCC =[ Y j;A ] OC ;[ Y j;ex 1 ] SCC =[ Y j;ex 2 ] OC ;[ Y j;R ] SCC = [ Y j;ex 1 ] SCC [ h F;S ] SCC =[ h F;S ] OC ;[ h A;S ] SCC =[ h A;S ] OC thesystemisnon-adiabatic Thescalingfactorforthetotalmassrows,denedby r A ,isgivenbyequation 4{20 Hencethescalingoffreshairmassrowsmaybecalculatedfro mequation 4{21 r A = [_ m ex 2 ] OC [_ m ex 1 ] SCC =1(4{20) r B = [_ m A ] OC [_ m A ] SCC = h st: +1+ EGR i SCC h st: +1 i OC (4{21) Usingthedenitionsof r A and r B ,andmatchingthespeciesbalancesforOCand SCCsystems,giveequation 4{22 ,whichsimpliestoequation 4{23 m R Y R;j +[(_ m F ) SCC (_ m F ) OC ] Y F;j +[(_ m A ) SCC (_ m A ) OC ] Y A;j =0(4{22) EGR ( Y R;j )+ Y F;j st: + Y A;j (1 r B )=0(4{23) Thisimpliesthatthedecreaseinthespeciesmassfractions fromfreshairand fuelmassrowsintheSCCsystem,isequaltothemassfraction saddedfromtheEGR stream.Thisequationisautomaticallysatisedfromtheto talmassbalance,forthegiven assumptions.Matchingoftheexitenthalpies,obtainedfro mtheenergyconservationgives equation 4{24 215

PAGE 216

(_ m F h F;S ) SCC (_ m F h F;S ) OC + (_ m A h A;S ) SCC (_ m A h A;S ) OC +_ m R ( h R;S ) SCC + Q L =0 (4{24) Thismeansthattheheatlossmustequalthesumofenergysupp liedbytheEGR streamandthedierentialinletethalpiesbetweentheSCCa ndOCsystemsforthefuel andairstreams.Furthersimplicationallowscalculation oftheinletairenthalpyusing equation 4{25 Q L =(_ m A ) SCC ( r B 1) ( h A;S ) SCC + st: ( h F;S ) SCC EGR ( h R;S ) SCC (4{25) Thespeciesbalanceisautomaticallysatisedforthegiven simplifyingassumptions. Thevariationoframetemperature,EIUHC,EICOandEIOHwith respecttothe air/EGRstreamtemperatureasshowningures 4-125 through 4-128 ,allshowedexcellent agreementbetweentheSCCandOCsystems. equivalentOCforSCC(EGR=2.0) SCC(EGR=2.0) equivalentOCforSCC(EGR=1.0) SCC(EGR=1.0) =1.0, res =10.0ms Ethanolfuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]Flametemperature[K]1200 1100 1000 900 800 700 600 210020001900180017001600150014001300 Figure4-125.Comparisonoframetemperatureforethanolfu elasafunctionofair stream/EGRtemperatureinSCCsystemandanequivalentunvi tiatedOC systematEGR=0.0andEGR=2.0, =1.0, res =10.0ms. 216

PAGE 217

equivalentOCforSCC(EGR=2.0) SCC(EGR=2.0) equivalentOCforSCC(EGR=1.0) SCC(EGR=1.0) =1.0, res =10.0ms Ethanolfuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]EIUHC[g/kg]1200 1100 1000 900 800 700 600 1e+02 10 1 Figure4-126.ComparisonofUHCemissionindexforethanolf uelasafunctionofair stream/EGRtemperatureinSCCsystemandanequivalentunvi tiatedOC systematEGR=0.0andEGR=2.0, =1.0, res =10.0ms. equivalentOCforSCC(EGR=2.0) SCC(EGR=2.0) equivalentOCforSCC(EGR=1.0) SCC(EGR=1.0) =1.0, res =10.0ms Ethanolfuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]EICO[g/kg]1200 1100 1000 900 800 700 600 1e+031e+02 10 Figure4-127.ComparisonofCOemissionindexforethanolfu elasafunctionofair stream/EGRtemperatureinSCCsystemandanequivalentunvi tiatedOC systematEGR=0.0andEGR=2.0, =1.0, res =10.0ms. 217

PAGE 218

equivalentOCforSCC(EGR=2.0) SCC(EGR=2.0) equivalentOCforSCC(EGR=1.0) SCC(EGR=1.0) =1.0, res =10.0ms Ethanolfuel,P=202.65kPa(2atm) Airstreamtemperature(T A )/EGRtemperature(EGRT)[K]EIOH[g/kg]1200 1100 1000 900 800 700 600 1e+02 10 1 Figure4-128.ComparisonofOHemissionindexforethanolfu elasafunctionofair stream/EGRtemperatureinSCCsystemandanequivalentunvi tiatedOC systematEGR=0.0andEGR=2.0, =1.0, res =10.0ms. 218

PAGE 219

4.8CaseStudy7:ChemicalKineticSensitivityAnalysisatt heBlowout LimitforEthanolFuel Asensitivityanalysisstudywascarriedoutforethanolfue lchemistryattheblowout limit.TheresultsforcasesofEGR=0.0andEGR=2.0havebeen presentedin gures 4-130 and 4-131 ,respectively.ThenormalizedOHsensitivityindices(de ned byequation 4{26 [ 79 ])forthemostimportantreactions(indicesgreaterthan1. 0for EGRof0.0andgreaterthan0.1forEGRof2.0)havebeenindica ted.Itwasfoundthat thenormalizedOHsensitivitiesweresignicantlylowerat EGR=2.0thanatEGR= 0.0.Clearly,thethreereactions O + OH = H + O 2 CO + OH = H + CO 2 and HCO = H + CO hadthehighestsensitivityindicesatEGRof0.0and2.0.Fig ure 4-49 depictsthechangeinblowoutresidencetimeforethanolfue l,asafunctionofOHspecies massfraction(Seesection 4.3 fordetails).Forconvenience,thepredictionsofblowout residencetimeatEGR=0.0andEGR=2.0hasbeenshowningure 4-129 .Fromthe gure,thecurvetsforblowoutresidencetimeasafunction ofOHmassfractionwere obtained,withafunctionaldependencegivenby( res ) BO =202 : 86 exp ( 0 : 0075 y OH ) and( res ) BO =0 : 085 exp ( 0 : 0018 y OH )atEGR=0.0and2.0,respectively.These curvetequationsmayfurtherbedierentiatedtoderivean expressionfordierential changesinblowoutresidencetimes d res res BO = 0 : 0075 dy OH forEGR=0.0and d res res BO = 0 : 0018 dy OH forEGR=2.0.Anexpressionfor dy OH maythusbe derived,givenby dy OH = y OH s normi;OH @ ( lnk i ).Theestimatedpercentagechangein blowoutresidencetimeswerefoundfromequations 4{27 and 4{28 forEGRof0.0and2.0, respectively.Thevaluesof y OH atT A =EGRT=1000Kwere0.00338775(EGR=0.0) and0.00118884(EGR=2.0).Thesensitivityindicesofthebl owoutresidencetimes(given byequation 4{29 )forthemostdominantreactions,havebeensummarizedinta ble 4-1 .It wasfoundthatthesesensitivityindiceswerefairlylow,in dicatinggoodreliabilityofthe results. 219

PAGE 220

Table4-1.SensitivityAnalysisforblowoutresidencetime Reaction s normi;OH s normi;OH s norm;i s norm;i (EGR=0.0)(EGR=2.0)(EGR=0.0)(EGR=2.0) O + OH = H + O 2 53.17541.004070.001352.148e-6 CO + OH = H + CO 2 30.06050.3207190.000766.863e-7 HCO = H + CO 27.20740.6828720.000691.461e-6 s normij = dy j y j k i dk i = d ( lny j ) d ( lnk i ) (4{26) d res res BO = 0 : 0075 y OH s normi;OH d ( lnk i );(EGR=0.0)(4{27) d res res BO = 0 : 0018 y OH s normi;OH d ( lnk i );(EGR=2.0)(4{28) s norm;i = d ( ln res ) BO d ( lnk i ) (4{29) EGR=2.0 EGR=0.0 =1.0 Ethanolfuel,P=202.65kPa(2atm) OHspeciesmassfraction( y OH )Blowoutresidencetime( res )[ms]0.004 0.0035 0.003 0.0025 0.002 0.0015 0.001 0.0005 0 10.00 1.000.100.01 Figure4-129.Comparisonofblowoutresidencetimeforetha nolfuelasafunctionofOH speciesmassfractionatEGR=0.0and2.0,EGRT=400,600,800 ,1000, 1200K, =1.0,P=202.65kPa(2atm). 220

PAGE 221

EGR=0.0NormalizedOHSensitivity( d ( lnY OH ) d ( lnk i ) ) hco+o=co+oh ch2(s)=ch2 h+ch3hco=h2+ch2hco h+oh=h2o ch2o+oh=hco+h2o ch3ch2o=h+ch3hco c2h5oh=h2o+c2h4 ch2+o=2h+co ch3+o=h+ch2o h+ch2co=ch3+co ch2+o=h2+co h+ch2hco=h2+ch2co h+c2h2=c2h3 h+c2h3=h2+c2h2 o+c2h4=ch3+hco ch3+oh=h2o+ch2(s) oh+c2h5oh=h2o+ch3choh o+c2h5oh=oh+ch3ch2o h+hcco=co+ch2(s) ch3+oh=ch3oh h+ho2=h2+o2 o+c2h2=h+hcco oh+ch2hco=h2o+ch2co o2+ch2oh=ch2o+ho2 oh+ho2=o2+h2o h+ch3=ch4 h2+o=h+oh h+ch2o=ch2oh h+c2h5oh=h2+ch3choh ch2hco=h+ch2co h+ho2=2oh hco+o2=co+ho2 h2+oh=h+h2o c2h5oh=oh+c2h5 o+c2h5oh=oh+ch3choh o2+ch2(s)=h+co+oh c2h5oh=ch3+ch2oh h+hco=h2+co hco+oh=co+h2o hco=h+co co+oh=h+co2 o+oh=h+o2605040302010 0 -10-20 Figure4-130.NormalizedOHSensitivityIndicesbeforeram eblowoutatEGR=0.0, = 1.0, T A =1000K 221

PAGE 222

EGR=2.0NormalizedOHSensitivity( d ( lnY OH ) d ( lnk i ) ) oh+c2h5oh=h2o+ch3ch2o ch2hco=h+ch2co h+c2h5oh=h2+ch3choh o+c2h5oh=oh+ch3ch2o o+c2h5oh=oh+ch3choh hco+o2=co+ho2 h+oh=h2o h+hco=h2+co ch3+o=h+ch2o hco+oh=co+h2o h+ch3=ch4 co+oh=h+co2 hco=h+co o+oh=h+o21.2 1 0.80.60.40.2 0 -0.2-0.4 Figure4-131.NormalizedOHSensitivityIndicesbeforeram eblowoutatEGR=2.0, = 1.0, T A = EGRT =1000K 222

PAGE 223

CHAPTER5 SUMMARYANDCONCLUSIONS Therameblowout,combustioneciencyandpollutantsgener ation(includingCO, UHC,sootprecursorsandparticulatesoot)ofvariousconve ntionalandbio-derived fuelswasinvestigatedinvitiatedandnon-vitiatedenviro nment.Thesimulationswere performedusingaconstant-pressureCSTRmodeldevelopedu singtheCanteracombustion code,implementedinC++.Thefollowinggeneralconclusion smaybedrawnfromthis study: Theimpactoffuelcompositiononcombustioneciencyandra metemperature,is signicantlydampenedoutwithuseofEGR.Hence,PoWERsyst emshadsuperior fuel-rexibilitycharacteristicscomparedtoconventiona lopencyclesystems. Allbiofuelstestedhadsuperiororcomparablecombustione ciencieswhen comparedtoconventionalfuels. Thetransitionoframelesscombustioninhomogenoussystem smaybecharacterized basedoncriticaltemperatureboundariesofsootformation ,criticalequivalence ratio,thresholdvaluesforsootprecursors(PAHandacetyl ene)orOHemissions.In heterogenouscombustionsystems,ramelesscombustiontra nsitionmayfurtherbe classiedbasedondegreeofhomogeneityachieved. Thereisasignicantdropinblowoutlimitsofsemiclosedsy stemswithreduction inequivalenceratio,EGRTandpressureand,increaseinEGR .Theblowoutlimits werefoundtobeastrongfunctionofignitionnumberand,OHa ndCOemission indices.Inadditionitwasfoundthattheperformanceofsem iclosedsystemsmaybe derivedfromequivalentopen-cyclesystems. Thecorrelationsofblowoutloadingparameterswithigniti onnumber,basedon detailedchemicalkineticsmaybeusedtocharacterizetheb lowout,similarto loadingparameterversusDamkohlernumbertrendsfromglob almodelsforblowout. Theprimaryeectofvitiationwasfoundtobethatoftheincr easeintheresidence timeofthefreshreactantstreamsandtemperaturesuppress ion,correspondingto increasedlevelsofEGR(assumingthatrecirculationstrea mcompositionisthesame astheexhaustcomposition). Thespecicconclusionsdrawnfromindividualcasestudies havebeensummarizedin thesubsequentsections. 223

PAGE 224

5.1CaseStudy1:SyngasandMethaneFuelSimulations Thecombustionperformance(rameblowout,combustioneci encyandemissions) ofthreedierentmixturesofsyngasgivenby,Syn75(CO:0.2 5,H 2 :0.75),Syn50 (CO:0.5,H 2 :0.5)andSyn25(CO:0.75,H 2 :0.25),wereanalyzed.Theresultsof thesyngasmixtureswerecomparedtothatofMethanefuel,fo rvaryinglevelsof EGRandequivalenceratio.Theeectsofadiabaticandnon-a diabaticEGRwith temperatureof1200Kwerecompared.Thesimulationswerepe rformedatanair streamtemperatureof400K,equivalenceratioof0.4,0.6,0 .8,1.0(fuel-leanside) and1.2,1.6,2.0and3.0(fuel-richside),EGRof0.0,1.0and 2.0,pressureof202.65 kPa(2.0atm),andaresidencetimeof5ms. Themaximumtemperaturewasfoundtooccurunderrichcondit ionsforsyngas mixturesandnearequivalenceratioofunityforMethane.Me thanefuelproduced lowerrametemperaturescomparedtoallthesyngasmixtures tested.Syn75had highesttemperaturesatstoichiometricconditionsorslig htlyleanconditions.At equivalenceratiosawayfromstoichiometriconfuel-richs ide,Syn75exhibitedlower rametemperatures,despitethehigherH 2 contentofthefuel.Typically,Syn25had highesttemperaturesontherichsideaswellasatveryloweq uivalenceratios. Thecombustioneciencywasfoundtodropwithrisingequiva lenceratios,for syngasmixtures.IncreasingH 2 contentofsyngasfueldecreasedcombustion eciency.Syn75fuelhadUHCemissionindicescomparableto thatofmethane untilequivalenceratioof1.2,beyondwhichheUHCemission sincreasedconsiderably forMethanefuel.Syn75producedhighestUHCemissionindic esfollowedbySyn50 andthelowestwasthatfromSyn25. Thecombustioneciencyforallfuelmixtures,increasedwi thaccompanying decreasesinUHCandCOemissions,withincreasingadiabati cEGR.Atfuel-rich conditions,theincreaseineciencywasverylow.Theeect ofincreasingadiabatic EGRonUHCemissionswasofloweringUHCforsyngascombustio n.Thecombustor temperaturesapproachedequilibriumtemperatureswithin creasingadiabaticEGR, forallequivalenceratiostested.TheeectofEGRcooledto atemperatureof1200 Konthecombustorrametemperaturewasofloweringtheramet emperatureswith increasingEGR.TrendsinUHCandCOemissionsfornon-adiab aticEGRwerethe sameasthoseforadiabaticEGRexceptforreducedramestabi lity. AnincreaseintheblowoutlimitsforadiabaticEGR,whilead ecreaseinblowout limitsatnon-adiabaticEGRwithtemperatureof1200Kwasob served.The extinctioncharacteristicsimprovedconsiderablywithan increaseinH 2 content. Thesootconcentrationswerequantiedbytheconcentratio noftheprecursorsof sootnuclei,thepyrenemolecules(A 4 species).Allsyngasmixturestestedshowed verylowsootingtendency.WithMethanefuel,signicantly higherEIA 4 valueswere observedonlyatfuel-richequivalenceratiosgreaterthan 1.6.Itwasobservedthat 224

PAGE 225

theA 4 emissionsincreasedwithanincreaseinadiabaticEGRathig hequivalence ratios. 5.2CaseStudy2:Biofuelsandn-HeptaneFuelSimulations Thecombustionperformanceofthebiofuelsethanol,Dimeth ylEther(DME)and biodieselsurrogateMethylButanoate(MB)weremodeledina CSTRmodel,and comparedagainstthatofdieselfuelsurrogate,n-heptane. Acomparisonhasbeen presentedoftherameblowoutstability,combustionecien cyandemissionsfor equivalenceratiosvaryingbetween0.4to1.0andexhaustga srecirculation(EGR)of 0.0,1.0and2.0,withtherecirculationstreamcooledtotem peratureof700K.The emissionsandcombustioneciencywerecomparedatareside ncetimeof10.0ms. Theblowouttemperaturewasfoundtodropwithdecreasingeq uivalenceratioforall fuels.Theblowoutsimulationresultshadsomeinherentsca tterintheirtrendsdue totheinstabilitiesincomputation. AtanEGRof0.0,MBhadthehighestblowoutLPvaluesindicati nghigherblowout stabilityovermostblowoutequivalenceratios,whileetha nolandDMEhadlower blowoutLPvalues,incomparisonton-heptane.AtanEGRof2. 0,ethanolhad themaximumblowoutLPvalues,whilen-heptane,MBandDMEha dcomparable values. Aconsiderablereductionincombustionblowoutstabilityw asfound,withincrease inEGR(foranEGRtemperatureof700K).Thelimitsofoperabi lityintherange ofequivalenceratioandrametemperaturewereconsiderabl ynarrowed.Thisisan inherentdrawbackofthePoWERsystemthatmaybeovercomeby theutilizationof fuelswithwiderblowoutstabilitylimits,suchashydrogen andsyngas,usinghigher inletstreamtemperaturesorhigherEGRtemperatures,orpo ssiblymethodslikeH 2 enrichment. Forsimulationsperformedataresidencetimeof10.0ms,the combustioneciency increasedwithdecreasingequivalenceratiointherange0. 4to1.0,forEGRof0.0, whileitwasmaximumatanoptimumequivalenceratioofabout 0.9forEGRof 2.0.TheequivalenceratiosforminimumCOemissionswere0. 5and0.9,forEGR of0.0and2.0,respectively.Theoptimumequivalenceratio forminimumemissions andmaximumcombustioneciencywasfoundtoshifttowardss toichiometric,with increasingEGR.Hence,operatingatalowequivalenceratio athighEGR,could resultinreductionincombustionperformanceinsteadofim provingit. Allbiofuelstestedhadsuperiororcomparablecombustione cienciescomparedto n-heptane. MBhadgenerallyhighercombustioneciencyandsubstantia llylowerUHC, whileCOemissionswerereasonablylow.Hence,useofMBinaP oWERsystem, hadpotentialtosubstantiallylowertheoverallpollutant emissionsandimprove 225

PAGE 226

eciency,sinceNO x emissionsinPoWERsystemislowduetoloweroperatingrame temperatures. Theimpactoffuelcompositiononcombustioneciencyandra metemperature,is signicantlydampenedoutwithuseofEGR.Hence,PoWERsyst emshadsuperior fuel-rexibilitycharacteristicscomparedtoconventiona lopencyclesystems. 5.3CaseStudy3:StudyofEGRTemperature/EGRlevelonEthan ol BlowoutLimits Inthisstudy,theblowoutstabilityofethanolfuelwasinve stigated,intheregime ofvitiatedcombustion.ThetestconditionswereEGR=0.0,0 .5,1.0,1.5,2.0, T A =EGRT=400,600,800,1000,1200K, =1.0andP=202.65kPa(2atm). Itwasobservedthattheblowouttemperaturevariednearlyl inearlywithEGR temperature,butnon-linearlywithEGR.Amarkedreduction inrameblowout stabilitywasobservedwithincreasinglevelsofEGR. TheLP,residencetimeandtotalmassrowrateatblowoutvari edlinearlyona logarithmicscalewithEGRtemperature,forvariousEGRlev els. ThenormalizedLPpredictions(normalizedtothecorrespon dingresultsat EGR=0.0)werefoundtobeastrongfunctionofnormalizedtem peratures,and OHandCOemissionsatblowout. Theapproachmaybepotentiallyusefulinmakingprediction sofblowoutstability margin,basedonmeasurementsofcombustioneciency.More over,themethodology ofnormalizedpredictions,ifsuitablyapplied,canbeused topredicttheblowout limitsofSCCenginesfromtheblowoutpredictionsofconven tionalopencycle(OC) engines,potentiallyreducingtheexpenseforextensivete sting. Theloadingparameterwasalsofoundtocorrelatewellwitht heignitionnumber. 5.4CaseStudy4:AStudyoftheEectofEGRTemperature/EGRl evel/ ResidenceTime/EquivalenceRatioonSootFormation(Acety leneFuel Combustion) Modelingstudieswereconductedinordertoestablishacorr elationbetweenrameless combustionattainedthroughhighlevelsofexhaustgasreci rculation(EGR)ina Power,WaterExtractionandRefrigeration(PoWER)enginea ndtheprocessofsoot formation.Thegasphasechemistryforthesootformationpr ocessinthecombustor primaryzonehasbeenmodeledusingaCSTRmodel.TheCSTRpre ssurewas P=202.65kPa(2.0atm).Fourcaseswereinvestigatedinorde rtoisolatetheeects ofvariousparameters:1.case1:EGR=0.0-5.0, =2 : 0, =1 : 0,EGRT=1000K 226

PAGE 227

2.case2:EGRT=400-1200K,EGR=0.0and2.0, =2 : 0, =1 : 0 3.case3: =0.5-5.0,EGR=0.0and2.0, =1 : 0,EGRT=1000K 4.case4: =0.1-10.0, =2 : 0,EGR=0.0and2.0,EGRT=1000K TheemissionindicesforCO,UHCandsootprecursorsincreas ewithEGR,while decreasewithincreasingEGRtemperature. Forthecasestested,whiletheopencycle(OC)caseofzeroEG Rshowedpractically nosootprecursorformation,thesemi-closedcycle(SCC)ca sesdepictedrisinglevels ofsootformationwithincreasingEGR. EIUHCincreasedwhileEICOandEIC 2 H 2 decreasedwithEGR,beforestabilizing toaconstantvaluebeyondEGRofabout3.0 ThecombustioneciencywasfoundtodecreasewithEGR,afte raninitialslight increase. ThePAHemissionindicesincreasedtoamaximumataboutEGRo f2.0,andthen nearlylinearlydecreasedwithEGR. TheEGRtemperature(sameasinletairtemperature)didnota ectUHC,CO emissionsforzeroEGRcase,whileloweredEIUHCandEIC 2 H 2 ,andincreasedCO emissionindicesslightly.Thecombustioneciencywasnot aectedforEGR=0.0, whilecombustioneciencyimprovedforEGR=2.0. EIforPAHshadabell-shapeddependencewithrespecttoEGRt emperature, however,therewasalsoageneraldecreasingtrendatEGRof0 .0.TheEIforPAHs weremuchhigheratEGR=2.0ascomparedwithEGR=0.0. EIUHCandC 2 H 2 increasedwithincreasingequivalenceratio.However,a characteristicminimumexisteddependingonEGRandEGRtem perature.The EICOvaluesincreasedforrich-sideequivalenceratiostoa maximum,andthen nearlylinearlydecreased. Thecombustioneciencydecreasedwithequivalenceratio. Thecombustion eciencyforEGR=2.0wassignicantlyhigheruptoequivale nceratioofabout2.0. TherewasasharpincreaseinEIC 2 H 2 betweenequivalenceratios1.0and2.0.The EIPAHtrendsweresimilartotheEIC 2 H 2 trends. TheresidencetimeplotsindicatethatatEGR=0.0,highoxid ativeregimewas dominantduetohigherrametemperatures,whileEGR=2.0,th etemperatureswere between1400-2100K,andhenceEIPAHincreasedwithresiden cetime. 227

PAGE 228

5.5CaseStudy5:SootParticleGrowthandOxidationinVitia ted CombustionofAcetyleneFuel Inthisstudy,thevitiatedcombustionofacetyleneinaCSTR wasmodeled,in ordertoquantifytheeectofvitiationonparticulatesoot formation.Theconditions ofsimulationwere =2.0,P=202.65kPa(2.0atm),EGR=0.0and2.0andrame temperaturesbetween1500-2000K. Anincreaseinsootvolumefractionandparticlediametersw ithEGRwasobserved, althoughthesootgrowthprocessesofnucleation,coagulat ionandsurfacechemistry, wereallsuppressed. Theprimaryeectofvitiationcontributingtotheincrease insootvolumefraction wasfoundtobethatoftheincreaseintheresidencetimeofth efreshreactant streams,correspondingtoincreasedlevelsofEGR. 5.6CaseStudy6:ModelingofSemi-ClosedCycleEngineChara cteristics UsingaKineticallyEquivalentUnvitiatedOpen-CycleCSTR System AnapproachwasoutlinedontheuseofanunvitiatedOCsystem tomodelthe combustioncharacteristicsofanSCCsystem.Fourcasesofe quivalentsystemswere analyzed,todetermineanappropriatemodelingapproach.S imulationswererunfor = 1.0,and res =10ms(forSCCsystem),EGRT=600,800,1000and1200K,EGR=0 .0 and2.0.Averygoodagreementwasfoundbetweentheoretical modelingandsimulation results. ItmaybeconcludedthattheprimaryeectofEGRissimultane ousrame temperaturereductionandincreaseinresidencetimeofthe freshreactants,giventhe assumptionthatEGRstreamcompositionisthesameasthatof theexhaust. ForanygivenSCCsystem,akineticallyequivalentOCsystem maybedetermined thatwouldmatchtheexitstateandtherefore,blowoutand/o remissionscharacteristics oftheSCCsystem,ifappropriatemassrowscalingwasapplie dtotheOCsystem. 5.7CaseStudy7:ChemicalKineticSensitivityAnalysisatt heBlowout LimitforEthanolFuel Asensitivityanalysiswasperformedontheeectofvariati onofrateconstantsonthe speciesmassfractionsatblowout.TheOHmassfractionsatb lowouthavebeenshownto correlatewiththeloadingparameter.Therefore,aqualita tiveunderstandingofthesources 228

PAGE 229

ofthemodeluncertaintieswasfromthesensitivityindices .Theanalysiswasperformed fortheconditionsof =1 : 0,EGR=0.0and2.0,EGRT=1000K.Theuncertaintiesin theblowoutresidencetimesfromuncertaintiesinthereact ionrateswerefoundtobelow. However,forexactquantitativecorrelationsofthereside ncetime(orloadingparameter) variationswithOHemissions,thepublishedexperimentalu ncertainties,oneachofthe reactions,needtobetakenintoaccount. 5.8ProposedFutureStudies ToIntegrateFrenklach'ssootparticulategrowthmodel(in cludingfractalaggregation submodel)intothedetailedchemicalkineticsmodel. Toinvestigateramestabilityatconstantrametemperature s,usingstabilization throughheatruxacrossreactorwall. Tocompareeectsofindividualdiluents(CO 2 ,H 2 O,N 2 )oncombustionstability andsootformation,atxedrametemperature. Tostudysimultaneousoptimizationofgascyclethermodyna micsandemissions,and characterizationofblowoutlimitsforapracticalsystemu singreactornetworkfor givencombustorlinergeometry. 229

PAGE 230

APPENDIXA GOVERNINGEQUATIONSOFCSTRMODEL A.1TransientCSTR-SingleStreamModel ThesummaryofCSTRmodeldescribedinthissectionisbasedo n[ 210 211 ].The generalassumptionsforthemodelarelistedbelow: Thesystemconsistsofsingleinletandexitstreams. Thegasesareassumedtobethermallyideal. Instantaneousmixingofproductsandreactants, Da !1 Theeectsofspeciesdiusion,conductivityandviscocity areabsent. Thesystemischemicallycontrolled. Thereactorpressureisconstant. Thesystemisspatiallyhomogenous. Thesysteminitialconditionandinletstreamstatesarekno wn. Forthesimulationsincantera,furtherconstraintswereim posedaslistedbelow. Thesystemconsistsofmultipleinletandexitstreams. Thereactorwallvelocityiszero. Thereactorsurfacereactionsdonotoccur. Thechangesinkineticandpotentialenergiesthroughthesy stemarenegligible. A.1.1ReactionKineticsoftheGasMixture Ageneralreactionschemeforareactingspeciespool S j ,isgivenas: a ij S j n b ij S j Thespeciesmassbalancemaybewrittenas: n R X i =1 n S X j =1 a ij M j = n R X i =1 n S X j =1 b ij M j (A{1) 230

PAGE 231

n R X i =1 n S X j =1 ( b ij a ij ) M j = n R X i =1 n S X j =1 ( ij ) M j =0(A{2) Forelementalmassbalance,considerspecies S j =( E 1 ) e 1 ( E 2 ) e 2 ::: ( E k ) e k Thespeciesmolecularweightmaybecalculatedasthesumoft hemasscontributions fromeachoftheelements. M j = n k X k =1 M E k e kj (A{3) n R X i =1 n S X j =1 n k X k =1 ( ij e kj ) M E k =0(A{4) whichgives, n R X i =1 n S X j =1 ( ij e kj )=0(A{5) A.1.1.1Equilibriumconstants Theequilibriumconstant, k P;i = exp G R u T = n S Y j =1 p j;eq p ij = n S Y j =1 ( c j;eq ) ij R u T ij (A{6) k C;i = k f;i k b;i = n S Y j =1 ( c j;eq ) ij (A{7) k P;i = exp S R u H R u T = K C;i R u T ( P n S j =1 ij ) (A{8) A.1.1.2Rateofreaction Theforwardratecoecientisgivenby: k f;i = A i T b i e E i R u T (A{9) 231

PAGE 232

Therateofareaction i isgivenas: r i 1 V d dt = 1 ij V dn ij dt (A{10) r i = k f;i ( T ) n S Y j =1 ( c j ) b ij k b;i ( T ) n S Y j =1 ( c j ) a ij (A{11) r i = A i T b i e E i R u T ( T ) n S Y j =1 ( c j ) a ij 1 1 K C;i ( T ) n S Y j =1 ( c j ) ij # (A{12) A.1.1.3Speciesproductionrate Therateofareaction i isgivenas: j 1 V dn j dt = 1 V P n R i =1 dn ij dt = 1 V n R X i =1 ij r i (A{13) A.1.2SpeciesConservation ( Accumulation )=( Inflow ) ( Outflow )+( Generation )(A{14) A.1.2.1Massbasis dm j dt CV =(_ m j;in m j;ex )+ M j V R j +_ m j;int =(_ m in y j;in m ex y j;ex )+ M j V R j +_ m j;int (A{15) m ex = K V P (A{16) m j;int = M j A R s j (A{17) 232

PAGE 233

A.1.2.2Molarbasis dn j dt CV =(_ n j;in n j;ex )+( V R j )+_ n j;int =(_ n in x j;in n ex x j;ex )+( V R j )+_ n j;int (A{18) n j;int = A R s j (A{19) A.1.2.3Concentrationbasis dn j dt CV = Q in c j;in Q ex c j;ex +( V R j )+_ n j;int (A{20) A.1.3ConservationofTotalMassA.1.3.1Massbasis dm dt CV =(_ m in m ex )+(_ m gen )+_ m int (A{21) m gen = V R n S X j =1 M j j = V R n S X i =1 n S X j =1 n j ( ij r i )=0(A{22) V R n S X i =1 M j ij =0(A{23) dm dt CV = Q in g;in Q ex g;ex +(_ m gen )+ X j m j;int (A{24) A.1.3.2Molarbasis dn dt CV =(_ n in n ex )+(_ n gen )+_ n int (A{25) where, 233

PAGE 234

(_ n gen )= n S X j =0 n gen;j = n S X j =0 V R j = V R n S X j =0 n R X i =0 ij r i = n R X i =0 i;R r i (A{26) i;R = n S X j =0 ij (A{27) dn dt CV = Q in c g;in Q ex c g;ex +( V R c g;gen )+ V R c g;int (A{28) A.1.4ConservationofEnergy ( Accumulation )=( Inflow ) ( Outflow )+( Generation )(A{29) A.1.4.1Massbasis dE dt CV = E in E ex + A R Q int W (A{30) m R e dt CV =(_ m in e in m ex e ex )+ A R Q int W (A{31) W = P dV R dt + W s + P m ex P m in (A{32) e = u + e KE + e PE = u + 1 2 v 2 + gz (A{33) m R u + 1 2 v 2 + gz dt CV =_ m in u + 1 2 v 2 + gz in m ex u + 1 2 v 2 + gz ex + A R Q int P dV R dt + W s + P m ex P m in (A{34) Since,energycannotbegeneratedordestroyed,thereisnoe xplicitgenerationterm forenergy.However,thereiseneryconversion,insidether eactor,thatactslikeasource 234

PAGE 235

termevenatsteadystate.Thissourcetermispartofthecont rol-volumeaccumulation termintheaboveequation. m R u + 1 2 v 2 + gz dt CV =_ m in h + 1 2 v 2 + gz in m ex h + 1 2 v 2 + gz ex + A R Q int P dV R dt + W s (A{35) V R u + 1 2 v 2 + gz dt CV = V R ( u ) dt CV = 0@ V R h P dt 1A CV = V R ( h ) dt CV P dV R dt + V R dP dt CV (A{36) V R ( h ) dt CV P dV R dt + V R dP dt CV =_ m in h + 1 2 v 2 + gz in m ex h + 1 2 v 2 + gz ex + A R Q int P dV R dt + W s (A{37) Assumingathermallyidealgasmixture, PV R = n S X j =1 n j R u T = n g R u T (A{38) 1 P dP dt + 1 V R dV R dt = 1 n g dn g dt + 1 T dT dt (A{39) dP dt = P 1 n g dn g dt + 1 T dT dt 1 V R dV R dt (A{40) 235

PAGE 236

V R ( h ) dt CV V R P 1 n g dn g dt + 1 T dT dt 1 V R dV R dt CV = m in h + 1 2 v 2 + gz in m ex h + 1 2 v 2 + gz ex + V R Q int W s (A{41) Notethatforthespecialcaseofzerowall-velocity,theter m dV R dt iszero. Sincethegaseshavebeenassumedtobecaloricallynon-idea l, h = h ( T )= n S X j =0 Y j h f;j + Z T T ref c P;j ( T ) dT (A{42) where h ( T )aregivenbeNASAparametrizedpolynomials.SeeAppendix[ B ]. FromKinetics, dn j dt =(_ n j;in n j;ex )+ V R j (A{43) dm j dt CV =(_ m j;in m j;ex )+ V R M j j (A{44) Wecouldassumethatwereachthestateofproductsbysensibl yheatingthe reactantstotheexittemperatureandthenallowingforther eactiontotakeplaceat thattemperature.Toaccomplishthis,wecouldmultiplythe speciesconservationequation with h j ( T ex )toget, n S X j =0 h j ( T ex ) dm j dt = n S X j =0 h j ( T ex )(_ m j;in m j;ex )+ n S X j =0 V R h j ( T ex ) M j j (A{45) Notethattheterm h j ( T ) M j j canbewrittenas, 236

PAGE 237

h j ( T ex ) M j j = V R n S X j =0 n R X i =0 M j h j ( T ex )( ij r i )= V R n R X i =0 H rxn;i M j r i = V R n R X i =0 H rxn;i r i = V R Q gen;g (A{46) where H rxn;i ( T )= P n S j =0 h j ij Subtractingequation( A{46 )from( A{41 )gives, n S X j =0 m j dh j ( T ) dt CV V R P 1 n g dn g dt + 1 T dT dt 1 V R dV R dt CV = n S X j =0 m j;in 1 2 v 2 + gz in n S X j =0 m j;ex 1 2 v 2 + gz ex +( V R Q gen;g )+ A R Q int W s + n S X j =0 m j;in [ h j ( T in ) h j ( T ex )]+_ m j;ex [ h j ( T ex ) h j ( T ex )](A{47) A.1.4.2Molarbasis n g h g dt CV V R P 1 n g dn g dt + 1 T dT dt 1 V R dV R dt CV = n g;in h + 1 2 1 M g v 2 + 1 M g gz in n g;ex h + 1 2 1 M g v 2 + 1 M g gz ex + A R Q int W s (A{48) FromKinetics, dn j dt =(_ n j;in n j;ex )+ V R j (A{49) Multiplyingtheaboveequationwith h j ( T ex ),gives, n S X j =0 h j ( T ex ) dn j dt = n S X j =0 h j ( T ex )(_ n j;in n j;ex )+ n S X j =0 V R h j ( T ex )_ j (A{50) Notethattheterm h j ( T )_ j canbewrittenas, 237

PAGE 238

h j ( T ex )_ j = V R n S X j =0 n R X i =0 h j ( T ex )( ij r i )= V R n R X i =0 H rxn;i r i = V R Q gen;g (A{51) where H rxn;i ( T )= P n S j =0 h j ij Subtractingequation( A{51 )from( A{48 )gives, n S X j =0 n j dh j ( T ) dt CV V R P 1 n g dn g dt + 1 T dT dt 1 V R dV R dt CV = n S X j =0 n j;in 1 2 1 M g v 2 + 1 M g gz in n S X j =0 n j;ex 1 2 1 M g v 2 + 1 M g gz ex ( V R Q gen;g )+ A R Q int W s + n S X j =0 n j;in h j ( T in ) h j ( T ex ) +_ n j;ex h j ( T ex ) h j ( T ex ) (A{52) A.2TransientCSTR-Multi-streamModel A.2.1SpeciesConservationA.2.1.1Massbasis dm j dt CV = N in X s =1 m in;s;j N ex X s =1 m ex;s;j +(_ m gen;j )+_ m int;j (A{53) d ( V R Y j ) dt CV = N in X s =1 m s;in Y s;j N ex X s =1 m s;ex Y s;j +(_ m gen;j )+_ m int;j (A{54) d ( V R ) dt CV = N in X s =1 Q s;in j N s;ex X s =1 Q ex j +(_ m gen )+_ m int;j (A{55) A.2.1.2Molarbasis dn j dt CV = N in X s =1 n ex;s;j N ex X s =1 n ex;s;j +(_ n gen;j )+_ n int;j (A{56) 238

PAGE 239

dn j dt CV = N in X s =1 n in;s x s;j N ex X s =1 n ex;s x s;j +(_ n gen;j )+_ n int;j (A{57) d ( c j V R ) dt CV = N in X s =1 Q in;s c s;j N ex X s =1 Q ex;s c s;j +( V R j )+_ n int;j (A{58) A.2.2ConservationofTotalMassA.2.2.1Massbasis dm dt CV = N in X s =1 m s;in N ex X s =1 m s;ex +(_ m gen )+_ m int;j (A{59) m gen = V R n S X j =1 M j j = V R n S X i =1 n S X j =1 n j ( ij r i )=0(A{60) V R n S X i =1 M j ij =0(A{61) d ( V R ) dt CV = N in X s =1 Q s;in in;s;g N ex X s =1 Q s;ex ex;s;g +(_ m gen )+_ m int;j (A{62) A.2.2.2Molarbasis dn dt CV = N in X s =1 n in;s N ex X s =1 n ex;s +(_ n gen )+_ n int (A{63) where, (_ n gen;g )= n S X j =1 n gen;j = n S X j =1 V R j = V R n S X j =1 n R X i =1 ij r i = n R X i =1 i;R r i (A{64) i;R = n S X j =0 ij (A{65) 239

PAGE 240

d ( c g V R ) dt CV = N in X s =1 Q in;s c in;g;s N ex X s =1 Q ex;s c ex;g;s +( V R c gen;g )+ V R c int;g (A{66) A.2.3ConservationofEnergyA.2.3.1Massbasis n S X j =1 m j ( dh j ( T )) dt CV V R P 1 n g dn g dt + 1 T dT dt 1 V R dV R dt CV = N in X s =1 n S X j =1 m in;s;j 1 2 v 2 + gz in N ex X s =1 n S X j =1 m ex;s;j 1 2 v 2 + gz ex + V R Q gen;g + A R Q int W s + N in X s =1 n S X j =1 m in;s;j [ h j ( T in ) h j ( T ex )]+ N ex X s =1 n S X j =1 m ex;s;j [ h j ( T ex ) h j ( T ex )](A{67) A.2.3.2Molarbasis n S X j =0 n j dh j ( T ) dt CV V R P 1 n g dn g dt + 1 T dT dt 1 V R dV R dt CV = n in X j =1 n S X j =1 n in;s;j 1 2 1 M g v 2 + 1 M g gz in n ex X j =1 n S X j =1 n ex;s;j 1 2 1 M g v 2 + 1 M g gz ex +( V R Q gen;g )+ A R Q int W s + n in X j =1 n S X j =1 n in;s;j h j ( T in ) h j ( T ex ) + n ex X j =1 n S X j =0 n ex;s;j h j ( T ex ) h j ( T ex ) (A{68) 240

PAGE 241

APPENDIXB THERMODYNAMICPROPERTIESOFGASMIXTURES Thissectiondescribestheparametrizationandderivation ofthermodynamic properties[ 210 ]forvariousspeciesinachemicalkineticmechanism. B.1SpecicHeats B.1.1CaloricallyIdealGases Foracaloricallyidealgas, C o P ( T )= C o P ( T ref )(B{1) B.1.2CaloricallyNon-IdealGases Forcaloricallynon-idealGases, NASApolynomial C o p = a 0 + a 1 T + a 2 T 2 + a 3 T 3 + a 4 T 4 (B{2) Shomatepolynomial C o p = a + bT + cT 2 + dT 3 + e T 2 (B{3) B.2Enthalpy h oj ( T )= h oj ( T ref )+ Z T T ref c oP;j ( T ) dT (B{4) B.2.1ConstantC P Parametrization h oj ( T )= h oj ( T ref )+ C o P;j ( T T ref )(B{5) B.2.2NASAParametrization h oj ( T ) R u T = a 0 + a 1 2 T + a 2 3 T 2 + a 3 4 T 3 + a 4 5 T 4 + a 5 T (B{6) 241

PAGE 242

B.2.3ShomateParametrization h oj ( T ) R u T = a + b 2 T + c 3 T 2 + d 4 T 3 e T 2 + f (B{7) B.3Entropy Fromrstlawofthermodynamics, TdS = dh dP (B{8) Z s j ( T;P ) s oj ( T ref ;P o ) d s j = Z T T ref c oP;j ( T ) T dT Z P P o R u T P dP (B{9) s j ( T;P ) s oj ( T ref ;P o )= Z T T ref c oP;j ( T ) T dT R u Tln P P o (B{10) s oj ( T;P o )= s oj ( T ref ;P o )+ Z T T ref c oP;j ( T ) T dT (B{11) s j ( T;P )= s oj ( T;P o ) R u Tln ( P P o )(B{12) B.3.1ConstantC P Parametrization s oj ( T )= s oj ( T ref )+ c oP;j;ref ln T T ref (B{13) B.3.2NASAParametrization s oj ( T ) R u = a 0 ln ( T )+ a 1 T + a 2 2 T 2 + a 3 3 T 3 + a 4 4 T 4 + a 6 (B{14) B.3.3ShomateParametrization s oj ( T ) R u = aln ( T )+ bT + c 2 T 2 + d 3 T 3 e 2 T 2 + g (B{15) 242

PAGE 243

APPENDIXC METHODOFMOMENTS Thissectionpresentsasummaryofstatisticsappliedindet erminationofmomentsfor particlesizedistributionofaerosols[ 149 221 ]. C.1BasicStatistics Probabilitydistributionfunction Theprobabilitydistributionfunction( F P ( x )) representstheprobabilityofxlessthananumbera. F p ( x )= P ( x
PAGE 244

M C r = E [( x x ) r ]= Z 1 1 ( x x ) r p ( x ) dx (C{7) M C 0 = E [( x x ) 0 ]= Z 1 1 ( x x ) 0 p ( x ) dx =1(C{8) M C 1 = E [( x x ) 1 ]= Z 1 1 ( x x ) 1 p ( x ) dx = x x =0(C{9) Variance: x 2 = M C 2 = E [( x x ) 1 ]= Z 1 1 ( x x ) 2 p ( x ) dx = 2 x 2x (C{10) Skewness: Skewness = M C 3 3 (C{11) Kurtosis: Skewness = M C 4 4 (C{12) Super-Skewness: Super Skewness = M C 5 5 (C{13) C.2LagrangianInterpolation Thegeneral n th orderinterpolationatpoint x betweengridpoints x i forknown y i = y ( x i )isgivenby: P n ( x )= n X i =1 l i ( x ) y i (C{14) 244

PAGE 245

l i ( x )= n Y j =1( j 6 = i ) x x j x i x j (C{15) log ( p )= P n ( p )= n X i =1 l i ( p )[ log ( i )](C{16) p =10 P n ( p ) = n Y i =1 [ i ] l i ( p ) (C{17) 245

PAGE 246

REFERENCES [1]U.S.EnvironmentalProtectionAgency,AirPollutants, OntheWWW,URL http://www.epa.gov/air/airpollutants.html ,2008. [2]S.J.Horlock,AdvancedGasTurbineCycles,ElsevierPer gamon,ISBN0080442730, 2003. [3]J.A.Wunning,J.G.Wunning,Flamelessoxidationtore ducethermal NO-formation,ProgressinEnergyandCombustionScience23 (1)(1997)81{94. [4]K.Dobbeling,J.Hellat,H.Koch,25yearsofBBC/ABB/Als tomleanpremix combustiontechnologies,JournalofEngineeringforGasTu rbinesandPower129(1) (2007)2{12,URL http://dx.doi.org/10.1115/1.2181183 [5]A.S.Feitelberg,M.A.Lacey,GErich-quench-leangastu rbinecombustor,in: Proceedingsofthe1997InternationalGasTurbine&Aeroeng ineCongress& Exposition,June2,1997-June5,GECorporateResearchand, Development, Niskayuna,UnitedStates,ASME,Orlando,FL,USA,ISBN0402 1215,ASME,1997. [6]A.H.Lefebvre,Gasturbinecombustion,HemispherePub. Corp.,Washington,ISBN 007037029X,1983. [7]M.Xu,Y.Fan,J.Yuan,C.Sheng,H.Yao,ModelingofCombus tionCharacteristics andNO x EmissioninHighlyPreheatedandDilutedAirCombustion,In ternational JournalofEnergyResearch23(1999)683{693. [8]Y.D.Wang,Y.Huang,D.McIlveen-Wright,J.McMullan,N. Hewitt,P.Eames, S.Rezvani,Atechno-economicanalysisoftheapplicationo fcontinuous staged-combustionandramelessoxidationtothecombustor designingasturbines, FuelProcessingTechnology87(8)(2006)727{736,URL http://dx.doi.org/10. 1016/j.fuproc.2006.02.003 [9]Z.A.Mansurov,Sootformationincombustionprocesses( review),Combustion, ExplosionandShockWaves41(6)(2005)727{744,URL http://dx.doi.org/10. 1007/s10573-005-0083-2 [10]A.K.Gupta,Thermalcharacteristicsofgaseousfuelra mesusinghightemperature air,JournalofEngineeringforGasTurbinesandPower126(1 )(2004)9{19,URL http://dx.doi.org/10.1115/1.1610009 [11]B.Singh,W.E.Lear,Modelingoframeblowout,emission sandsootformation invitiatedcombustionofconventionalfuelsandsyngas:pa per#78,in:27th AnnualInternationalConferenceonThermalTreatmentTech nologies2008,vol.2, DepartmentofMechanicalEngineering,Bldg.MAE-C,Univer sityofFlorida, Gainesville,FL32611,AirandWasteManagementAssociatio n,Pittsburgh,PA 15222,UnitedStates,766{780,2008. 246

PAGE 247

[12]W.Ellis,W.E.Lear,B.Singh,A.Srinivasan,J.Critten den,S.A.Sheri,Flameless CombustionofBiofuelsinaSemi-ClosedCycleGasTurbine,i n:46 th AIAA AerospaceSciencesMeetingandExhibit,UniversityofFlor ida,Gainesville,FL 32611,UnitedStates,AmericanInstituteofAeronauticsan dAstronauticsInc,Reno, Nevada,UnitedStates,095843{095843,papernumberAIAA-2 008-1140,2008. [13]K.Maruta,K.Muso,K.Takeda,T.Niioka,Reactionzones tructureinrameless combustion,in:28thInternationalSymposiumonCombustio n,vol.28,Facultyof SystemsSci.andTechnol.,AkitaPrefecturalUniversity,H onjyo,Akita015-0055, Japan,CombustionInstitute,ISBN0082-0784,2117{2123,2 000. [14]M.D.Joannon,A.Cavaliere,T.Faravelli,E.Ranzi,P.S abia,A.Tregrossi,Analysis ofprocessparametersforsteadyoperationsinmethanemild combustiontechnology, ProceedingsoftheCombustionInstitute30(2)(2005)2605{ 2612. [15]T.L.Cong,P.Dagaut,Kineticsofnaturalgas,naturalg as/syngasmixtures oxidationandeectofburntgasrecirculation:Experiment alanddetailedmodeling, in:2007ASMETurboExpo,vol.1,CentreNationaldelaRecher cheScientique, 45071OrleansCedex2,France,AmericanSocietyofMechanic alEngineers,New York,NY10016-5990,UnitedStates,ISBN079184790X,387{3 95,2007. [16]J.R.Kalb,T.Sattelmayer,LeanblowoutlimitandNOxpr oductionofapremixed sub-ppmNOxburnerwithperiodicrecirculationofcombusti onproducts,Journalof EngineeringforGasTurbinesandPower128(2)(2006)247{25 4,URL http://dx. doi.org/10.1115/1.2061267 [17]B.Singh,W.E.Lear,ModelingofFlameBlowoutandEmiss ionsinVitiated CombustionofBiofuels,in:ProceedingsofASMETurboExpo2 009:Powerfor Land,SeaandAir,UniversityofFlorida,Gainesville,FL,U SA,AmericanSocietyof MechanicalEngineers,Orlando,USA,papernumberGT2009-6 0272,2009. [18]R.Schefer,Hydrogenenrichmentforimprovedleanrame stability,International JournalofHydrogenEnergy28(2003)1131{1141. [19]C.Duwig,D.Stankovic,L.Fuchs,G.Li,E.Gutmark,Expe rimentalandnumerical studyoframelesscombustioninamodelgasturbinecombusto r,Combustion ScienceandTechnology180(2)(2008)279{295,URL http://dx.doi.org/10. 1080/00102200701739164 [20]D.R.Noble,Q.Zhang,A.Shareef,J.Tootle,A.Meyers,T .Lieuwen,Syngas MixtureCompositionEectsUponFlashbackAndBlowout,in: ASMETurboExpo 2006:PowerforLand,SeaandAir,vol.49,GeorgiaInstitute ofTechnology,Atlanta, GA,ASME,May8-11,Barcelona,Spain,papernumberGT2006-9 0470,2006. [21]R.A.Strehlow,Combustionfundamentals,McGraw-Hill ,NewYork,ISBN 0070622213,1984. [22]D.B.Spalding,Somefundamentalsofcombustion,Acade micPress,NewYork,1955. 247

PAGE 248

[23]J.F.Crittenden,Dilute,kineticallycontrolledcomb ustioneciencypredictionfor recirculatingsemi-closedgasturbinecycles:anon-dimen sionalapproachusingthe rstDamkohlernumberasaparameter,1999. [24]S.K.Hoekman,BiofuelsintheU.S.-ChallengesandOppo rtunities,Renewable Energy34(1)(2008)14{22,URL http://dx.doi.org/10.1016/j.renene.2008. 04.030 [25]S.Sidhu,J.Graham,R.Striebich,Semi-volatileandpa rticulateemissionsfromthe combustionofalternativedieselfuels,Chemosphere42(57)(2001)681{690,URL http://dx.doi.org/10.1016/S0045-6535(00)00242-3 [26]C.K.Westbrook,W.J.Pitz,H.J.Curran,Chemicalkinet icmodelingstudyofthe eectsofoxygenatedhydrocarbonsonsootemissionsfromdi eselengines,Journal ofPhysicalChemistryA110(21)(2006)6912{6922,URL http://dx.doi.org/10. 1021/jp056362g [27]R.Anxionnaz,Installationturbinesgazcircuitsemiouvert,SocitRautear,SA, Franz(1945)999133. [28]R.Anxionnaz,Improvementsinorrelatingtogasturbin eplantwithsemi-open circuit,SocitRautear,SA,Brit.(1948)651166. [29]R.Davis,FinalTestReportofWolverineGasTurbinePla nt,Tech.Rep.,report numberNS-622-210,1956. [30]S.DeWitt,W.Boyum,InternallyFiredsemi-closedcycl egasturbineplantforNaval Propulsion,56-GTP-16,1956. [31]N.Gasparovic,Semi-closedgas-turbineinstallation s,BWK.Brennsto-WaermeKraft19(5)(1967)250{259. [32]N.Gasparovic,Advantageofsemi-closedcyclegasturb inesfornavalshippropulsion, NavalEngineersJournal80(2)(1968)275{281. [33]W.E.Lear,A.Laganelli,HighPressureRegenerativeTu rbineEngine:21stCentury Propulsion,FinalTestReportforContractNo.NAS3-27396, 1999. [34]N.Muley,W.Lear,EectofExhaustGasRecirculation(E GR)onThermalNOx FormationRate,in:41 st AerospaceSciencesMeetingandExhibit,Universityof Florida,Gainesville,FL,USA,AmericanInstituteofAeron auticsandAstronautics Inc,Reno,NV,USA,papernumberAIAA2003-0503,2003. [35]T.S.Nemec,W.E.Lear,ThermodynamicPerformanceofaS emi-ClosedGas TurbineCombinedwithaRankingBottomingCycle,in:41stAe rospaceSciences MeetingandExhibit,UniversityofFlorida,Gainesville,F L,USA,American InstituteofAeronauticsandAstronauticsInc,Reno,NV,US A,papernumber AIAA-98-0361,1998. 248

PAGE 249

[36]M.R.Harwood,W.E.Lear,RecirculatingPlugFlowReact or,in:SAE,AIAA, ASME,IEEE,AIChE,andANS,IECEC-IntersocietyEnergyConv ersion EngineeringConference,34 th ,UniversityofFlorida,Gainesville,FL,USA,Societyof AutomobileEngineers,Vancouver,Canada,papernumber199 9-01-2603,1999. [37]J.F.Crittenden,PredictionofDilute,KineticallyCo ntrolledGasTurbine CombustionEciency,in:1stInternationalEnergyConvers ionEngineering Conference,UniversityofFlorida,Gainesville,FL32611, UnitedStates,American InstituteofAeronauticsandAstronauticsInc,Portsmouth ,Virginia,papernumber AIAA2003-5947,2003. [38]R.MacFarlane,W.Lear,SystemimpactofH2Oproduction andinjectiononanovel semi-closedcyclegasturbine,in:35 th AerospaceSciencesMeetingandExhibit, UniversityofFlorida,Gainesville,FL,USA,AmericanInst ituteofAeronauticsand AstronauticsInc,Reno,NV,USA,papernumberAIAA97-0374, 1997. [39]P.L.Meitner,P.F.Senick,A.L.Laganelli,W.E.Lear,D emonstrationofa Semi-ClosedCycle,TurboshaftGasTurbineEngine,in:2000 ASMETurboExpo, NASAGlennResearchCenter,Cleveland,OH,USA;ScienticA pplications InternationalCorporation,KingofPrussia,PA,USA;Unive rsityofFlorida, Gainesville,FL,USA,AmericanSocietyofMechanicalEngin eers,Munich,Germany, 2000. [40]J.R.Khan,W.E.Lear,S.A.Sherif,Performanceofanove lcombinedcooling andpowergasturbinewithwaterharvesting,in:2004ASMETu rboExpo,vol.7, DepartmentofMechanicalEngineering,UniversityofFlori da,237MAEBldg. B,Gainesville,FL32611-6300,UnitedStates,AmericanSoc ietyofMechanical Engineers,NewYork,NY10016-5990,UnitedStates,211{223 ,2004. [41]C.Ryu,A.Srinivasan,D.R.Tiany,J.F.Crittenden,W. E.Lear,S.Sherif, Dynamicmodelingofanovelcooling,heat,power,andwaterm icroturbinecombined cycle,UniversityofFlorida,Gainesville,FL,USA,Jackso nville,Florida,USA,paper numberES2008-54095,2008. [42]A.L.Laganelli,C.Rodgers,W.E.Lear,P.Meitner,Semi -ClosedGasTurbine Cycles:anEcologicalSolution,in:2001ASMETurboExpo,SA IC,KingofPrussia, PA,USA;ITC,SanDiego,CA,USA;UniversityofFlorida,Gain esville,FL,USA; USArmy/ARL,Cleveland,OH,USA,AmericanSocietyofMechan icalEngineers, NewOrleans,Louisiana,USA,2001. [43]P.L.Meitner,A.L.Laganelli,P.F.Senick,W.E.Lear,D emonstrationof semi-closedcycle,turboshaftgasturbineengine,Tech.Re p.,1993. [44]M.R.Harwood,Are-examinationofcombustorperforman cefundamentalsthrough theuseofarecirculatingplugrowreactor,1999. 249

PAGE 250

[45]J.J.Boza,W.E.Lear,S.A.Sherif,PerformanceofaNove lSemi-ClosedGasTurbine RefrigerationCombinedCycle,in:Proceedingsofthe2003A SMETurboExpo: PowerforLand,SeaandAir,vol.3,UniversityofFlorida,Ga inesville,FL,USA, AmericanSocietyofMechanicalEngineers,Atlanta,Georgi a,USA,papernumber GT2003-38576,2003. [46]J.U.R.Khan,Designandoptimizationofadistributedg enerationsystemwiththe productionofwaterandrefrigeration[electronicresourc e],UniversityofFlorida, Gainesville,Fla.,URL Fulltext:http://purl.fcla.edu/fcla/etd/UFE0013767 2006. [47]J.R.Khan,W.E.Lear,S.A.Sherif,J.F.Crittenden,Sec ondlawanalysisofa novelcombinedcoolingandpowercyclewithwaterharvestin g,in:ASMETurbo Expo2005-GasTurbieTechnology:FocusfortheFuture,vol. 5,Department ofMechanicalandAerospaceEngineering,UniversityofFlo rida,210MAEBldg. B,Gainesville,FL32611-6300,UnitedStates,AmericanSoc ietyofMechanical Engineers,NewYork,NY10016-5990,UnitedStates,261{270 ,2005. [48]J.R.Khan,W.E.Lear,S.A.Sherif,J.F.Crittenden,Dem onstrationofanovel combinedcoolingandpowergasturbinewithwaterharvestin g,in:43rdAIAA AerospaceSciencesMeetingandExhibit,DepartmentofMech anicalandAerospace Engineering,UniversityofFlorida,237MAEBldg.B,Gaines ville,FL32611-6300, UnitedStates,AmericanInstituteofAeronauticsandAstro nauticsInc,Reston,VA 20191,UnitedStates,7537{7550,2005. [49]J.R.Khan,W.E.Lear,S.A.Sherif,E.B.Howell,J.F.Cri ttenden,Water extractionandperformanceofanovelpressurizedCHPsyste m,in:44thAIAA AerospaceSciencesMeeting2006,vol.23,UniversityofFlo rida,Gainesville, FL32611,AmericanInstituteofAeronauticsandAstronauti csInc,Reston,VA 20191-4344,UnitedStates,ISBN1563478072,17906{17915, 2006. [50]J.R.Khan,W.E.Lear,S.A.Sherif,E.B.Howell,J.F.Cri ttenden,P.L. Meitner,Testingandmodelingofasemi-closedgasturbinec ycleintegratedwith avaporabsorptionrefrigerationsystem,in:2006ASMEInte rnationalMechanical EngineeringCongressandExposition,IMECE2006,Departme ntofMechanicaland AerospaceEngineering,UniversityofFlorida,Gainesvill e,FL32611,UnitedStates, AmericanSocietyofMechanicalEngineers,NewYork,NY1001 6-5990,United States,ISBN1071-6947;0791837904,12,2006. [51]J.R.Khan,W.E.Lear,S.A.Sherif,E.B.Howell,J.F.Cri ttenden,P.L.Meitner, AnovelpressurizedCHPsystemwithwaterextraction,in:4t hInternationalEnergy ConversionEngineeringConference,vol.2,UniversityofF lorida,Gainesville,FL 32611,AmericanInstituteofAeronauticsandAstronautics Inc,Reston,VA20191, UnitedStates,1315{1332,2006. 250

PAGE 251

[52]W.E.Lear,C.Ryu,J.F.Crittenden,A.Srinivasan,W.El lis,D.R.Tiany, S.Sherif,SystemDesignofaNovelCombinedCooling,Heat,P owerandWater MicroturbineCombinedCycle,UniversityofFlorida,Gaine sville,FL,USA, AmericanSocietyofMechanicalEngineers,Berlin,Germany ,papernumber GT2008-51454,2008. [53]S.M.Camporeale,B.Fortunato,Thermodynamicanalysi sofsemi-closedgasturbine combinedcycleswithhightemperaturedilutedaircombusti on,in:2007ASME TurboExpo,vol.3,DipartimentodiIngegneriaMeccanicaeG estionale,Politecnico diBari,70100Bari,Italy,AmericanSocietyofMechanicalE ngineers,NewYork,NY 10016-5990,UnitedStates,ISBN079184790X,459{464,2007 [54]D.Fiaschi,G.Manfrida,Newsemi-closedgasturbinecy clewithCO2separation, EnergyConversionandManagement40(15)(1999)1669{1678, URL http://dx. doi.org/10.1016/S0196-8904(99)00060-6 [55]K.Jordal,O.Bollard,A.Klang,Aspectsofcooledgastu rbinemodelingforthe semi-closedO2/CO2cyclewithCO2capture,JournalofEngin eeringforGas TurbinesandPower126(3)(2004)507{515,URL http://dx.doi.org/10.1115/1. 1762908 [56]O.Bolland,P.Mathieu,ComparisonoftwoCO2removalop tionsincombinedcycle powerplants,EnergyConversionandManagement39(16-18)( 1998)1653{1663, URL http://dx.doi.org/10.1016/S0196-8904(98)00078-8 [57]T.Faravelli,E.Ranzi,P.Sabia,A.Tregrossi,M.Joann on,A.Cavaliere,Analysisof processparametersforsteadyoperationsinmethanemildco mbustiontechnology, ProceedingsofCombustionInstitute30(2005)2605{2612. [58]M.D.Joannon,A.Saponaro,A.Cavaliere,Zero-dimensi onalanalysisofdiluted oxidationofmethaneinrichconditions,in:28thInternati onalSymposiumon Combustion,vol.28,DipartimentodiIngegneriaChimica,U niversitaFedericoII, Napoli,Italy,CombustionInstitute,ISBN0082-0784,1639 {1645,2000. [59]A.Gupta,DevelopmentsinHighTemperatureAirCombust ion(Flameless Oxidation)andFuelReforming,in:44thAIAAAerospaceScie ncesMeeting andExhibit,UniversityofMaryland,CollegePark,MD,Amer icanInstituteof AeronauticsandAstronauticsInc,Reno,NV,papernumberAI AA2006-1490,2006. [60]M.Katsuki,T.Hasegawa,Thescienceandtechnologyofc ombustioninhighly preheatedair,Symposium(International)onCombustion27 (2)(1998)3135{3146. [61]I.O.Awosope,N.H.Kandamby,F.C.Lockwood,Flameless oxidationmodelling: Onapplicationtogasturbinecombustors,JournaloftheEne rgyInstitute79(2) (2006)75{83. 251

PAGE 252

[62]H.Schutz,R.Luckerath,T.Kretschmer,B.Noll,M.Aign er,Analysisofthe pollutantformationintheFLOX[registeredtrademark]com bustion,Journalof EngineeringforGasTurbinesandPower130(1)(2008)011503 ,URL http://dx. doi.org/10.1115/1.2747266 [63]B.Singh,W.E.Lear,FlameBlowoutStabilityinVitated Combustion,in:7th AnnualInternationalEnergyConversionEngineeringConfe rence,Universityof Florida,Gainesville,FL,USA,AmericanInstituteofAeron auticsandAstronautics Inc,Denver,Colorado,papernumberAIAA-2009-4527,2009. [64]B.Singh,W.E.Lear,SootFormationinVitiatedCombust ion,in:45th AIAA/ASME/SAE/ASEEJointPropulsionConference&Exhibit ,Universityof Florida,Gainesville,FL,USA,AmericanInstituteofAeron auticsandAstronautics Inc,Denver,Colorado,papernumberAIAA-2009-4970,2009. [65]Y.Levy,V.Sherbaum,P.Ar,BasicthermodynamicsofFL OXCOM,thelow-NOx gasturbinesadiabaticcombustor,AppliedThermalEnginee ring24(11-12)(2004) 1593{1605,URL http://dx.doi.org/10.1016/j.applthermaleng.2003.11. 022 [66]M.Zheng,M.C.Mulenga,G.T.Reader,M.Wang,D.S.-K.Ti ng,J.Tjong, Biodieselengineperformanceandemissionsinlowtemperat urecombustion,Fuel 87(6)(2008)714{722,URL http://dx.doi.org/10.1016/j.fuel.2007.05.039 [67]S.Adachi,A.Iwamoto,S.Hayashi,H.Yamada,S.Kaneko, Emissionsincombustion ofleanmethane-airandbiomass-airmixturessupportedbyp rimaryhotburned gasinamulti-stagegasturbinecombustor,in:31stInterna tionalSymposium onCombustion,vol.31II,DepartmentofMechanicalEnginee ring,Universityof Tokyo,Japan,ElsevierLtd,Oxford,OX51GB,UnitedKingdom ,ISBN1540-7489, 3131{3138,URL http://dx.doi.org/10.1016/j.proci.2006.07.239 ,2007. [68]M.Flamme,LowNOxcombustiontechnologiesforhightem peratureapplications, EnergyConversionandManagement42(15-17)(2001)1919{19 35,URL http://dx. doi.org/10.1016/S0196-8904(01)00051-6 [69]N.Aida,T.Nishijima,S.Hayashi,H.Yamada,T.Kawakam i,A.Nicolle, G.Bourque,J.Hermanson,Combustionofleanprevaporizedf uel-airmixtures mixedwithhotburnedgasforlow-NOxemissionsoveranexten dedrangeoffuel-air ratios,in:30thInternationalSymposiumonCombustion,vo l.30II,Aeronautical EnvironmentTechnologyCenter,JapanAerospaceExplorati onAgency,Chofu-shi, Tokyo182-8522,Japan,ElsevierLtd,Oxford,OX51GB,Unite dKingdom,ISBN 1540-7489,2885{2892,URL http://dx.doi.org/10.1016/j.proci.2004.08.040 2005. 252

PAGE 253

[70]R.Weber,J.P.Smart,W.V.Kamp,A.Cavaliere,A.N.Sayr e,Onthe(MILD) combustionofgaseous,liquid,andsolidfuelsinhightempe raturepreheatedair, in:30thInternationalSymposiumonCombustion,vol.30II, InstituteofEnergy ProcessEngineeringandFuelTechnology,TUClausthal,Ger many,ElsevierLtd, Oxford,OX51GB,UnitedKingdom,ISBN1540-7489,2623{2629 ,URL http://dx. doi.org/10.1016/j.proci.2004.08.101 ,2005. [71]S.Kajita,S.Ohga,M.Ogata,S.Itaka,J.Kitajima,T.Ki mura,A.Okuto,Advanced developmentofasecond-generationdry,low-NOxcombustor for1.5MWgasturbine, in:Proceedingsofthe1996InternationalGasTurbineandAe roengineCongress& Exhibition,KawasakiHeavyIndustries.Ltd,Akashi,Jpn,A SME,NewYork,NY, USA,ISBN0402-1215,6,1996. [72]N.S.Muley,Eectofexhaustgasrecirculationontherm alNOxformationrateina gasturbineengine,2002. [73]T.S.Nemec,Thermodynamicdesignpointstudyofasemiclosedrecuperated intercooledgasturbinecombinedwithaRankinebottomingc ycle,1995. [74]A.Cavaliere,M.deJoannon,MildCombustion,Progress inEnergyandCombustion Science30. [75]M.Derudi,A.Villani,R.Rota,Sustainabilityofmildc ombustionof hydrogen-containinghybridfuels,in:31stInternational SymposiumonCombustion, vol.31II,DipartimentodiChimica,MaterialieIngegneria ChimicaGiulioNatta, CIIRCO,PolitecnicodiMilano,20131Milan,Italy,Elsevie rLtd,Oxford,OX51GB, UnitedKingdom,ISBN1540-7489,3393{3400,URL http://dx.doi.org/10.1016/ j.proci.2006.08.107 ,2007. [76]S.Kumar,P.J.Paul,H.S.Mukunda,J.Bellan,M.Delicha tsios,G.S.Nathan, Investigationsofthescalingcriteriaforamildcombustio nburner,in:30th InternationalSymposiumonCombustion,vol.30II,Combust ionGasication andPropulsionLaboratory,DepartmentofAerospaceEngine ering,IndianInstitute ofScience,Bangalore560012,India,ElsevierLtd,Oxford, OX51GB,United Kingdom,ISBN1540-7489,2613{2621,URL http://dx.doi.org/10.1016/j. proci.2004.07.045 ,2005. [77]S.Aceves,D.Flowers,ADetailedChemicalKineticAnal ysisofLowTemperature Non-SootingDieselCombustion,URL http://www.osti.gov/bridge// product.biblio.jsp?query_id=0&page=0&osti_id=150159 06 ,reportnumber UCRL-CONF-206985,2005. [78]I.Glassman,O.Nishida,G.Sidebotham,Criticaltempe raturesofsootformation, chap.59,Springer-VerlagGmbH&CompanyKG,Berlin33,Ger, ISBN0172-6218, 316{324,1994. [79]I.Glassman,R.A.Yetter,Combustion,Elsevier,Amste rdam;Boston,ISBN 0120885735(hbk.);9780120885732(hbk.),2008. 253

PAGE 254

[80]T.Plessing,N.Peters,J.G.Wnning,Laseropticalinve stigationofhighlypreheated combustionwithstrongexhaustgasrecirculation,Symposi um(International)on Combustion27(2)(1998)3197{3204. [81]W.Blasiak,W.H.Yang,K.Narayanan,J.vonScheele,Fla melessoxyfuel combustionforfuelconsumptionandnitrogenoxidesemissi onsreductionsand productivityincrease,JournaloftheEnergyInstitute80( 1)(2007)3{11,URL http://dx.doi.org/10.1179/174602207X174379 [82]M.Flamme,Newcombustionsystemsforgasturbines(NGT ),AppliedThermal Engineering24(11-12)(2004)1551{1559,URL http://dx.doi.org/10.1016/j. applthermaleng.2003.10.024 [83]C.R.Shaddix,T.C.Williams,R.W.Schefer,Eectofsyn gascompositionand CO2-dilutedoxygenonperformanceofapremixedswirl-stab ilizedcombustor,in: 2007ASMETurboExpo,vol.2,CombustionResearchFacility, SandiaNational Laboratories,Livermore,CA94550,UnitedStates,America nSocietyofMechanical Engineers,NewYork,NY10016-5990,UnitedStates,ISBN079 184790X,889{897, 2007. [84]A.K.Gupta,T.Hasegawa,TheEectofAirPreheatTemper atureandOxygen ConcentrationinAirontheStructureofPropaneAirDiusio nFlames,in: Proceedingsof37thAIAAAerospaceSciencesMeetingandExh ibit,University ofMaryland,CollegePark;NipponFurnaceKogyoKaisha,Ltd .,Yokohama,Japan, AmericanInstituteofAeronauticsandAstronauticsInc,Ja nuary11-14,Reno,NV, papernumberAIAA99-0725,1999. [85]C.Ellul,M.Pourkashanian,A.Williams,K.A.M.Rabea, Eectsoffuel compositioninhightemperatureaircombustion,Journalof theEnergyInstitute 81(3)(2008)143{148,URL http://dx.doi.org/10.1179/174602208X331643 [86]M.Mancini,R.Weber,U.Bollettini,PredictingNOxemi ssionsofaburneroperated inramelessoxidationmode,in:Twenty-NinthInternationa lSymposiumon CombustionHokkaidoUniversitySapporoJapan,July21,200 2-July25,vol.29, Intl.FlameResearchFoundation,ResearchStationB.V.,PO Box10000,1970CA Ijmuiden,Netherlands,CombustionInstitute,Sapporo,Ja pan,ISBN15407489, 1155{1162,2002. [87]A.Nicolle,P.Dagaut,OccurrenceofNO-reburninginMI LDcombustionevidenced viachemicalkineticmodeling,Fuel85(17-18)(2006)2469{ 2478,URL http://dx. doi.org/10.1016/j.fuel.2006.05.021 [88]M.Hamdi,H.Benticha,M.Sassi,Detailedchemicalmode llingofarameless combustionturbineforpollutionprevention,Internation alJournalofEnvironment andPollution32(3)(2008)410{421,URL http://dx.doi.org/10.1504/IJEP. 2008.017919 254

PAGE 255

[89]M.Fuchihata,M.Katsuki,Y.Mizutani,T.Ida,Observat ionoftheramestructures emergingatlowDamkhlernumberelds,ProceedingsoftheCo mbustionInstitute 31(1)(2007)1353{1359. [90]C.Galletti,A.Parente,L.Tognotti,Numericalandexp erimentalinvestigationof amildcombustionburner,CombustionandFlame151(4)(2007 )649{664,URL http://dx.doi.org/10.1016/j.combustflame.2007.07.01 6 [91]P.J.Coelho,N.Peters,NumericalSimulationofaMildC ombustionBurner, CombustionandFlame124(3)(2001)503{518,URL http://dx.doi.org/10.1016/ S0010-2180(00)00206-6 [92]J.Yuan,I.Naruse,Modelingofcombustioncharacteris ticsandNOxemission inhighlypreheatedanddilutedaircombustion,Internatio nalJournalofEnergy Research22(14)(1998)1217{1234,URL http://dx.doi.org/10.1002/ (SICI)1099-114X(199811)22:14<1217::AID-ER443>3.0.CO ;2-S [93]O.Piepers,P.P.Breithaupt,A.B.N.V.Beelen,Stabili tyoframescloseto auto-ignitiontemperaturesgeneratedbyextremeseparate dgas-airinlets,Journalof EnergyResourcesTechnology,TransactionsoftheASME123( 1)(2001)50{58,URL http://dx.doi.org/10.1115/1.1345731 [94]D.Tabacco,C.Innarella,C.Bruno,Theoreticalandnum ericalinvestigationon ramelesscombustion,CombustionScienceandTechnology17 4(7)(2002)1{35. [95]C.Marek,R.R.Tacina,EectofExhaustGasRecirculati ononEmissions fromaFlameTubeCombustorUsingJet-AFuel,OntheWWW,URL http://ntrs.nasa.gov/search.jsp?R=566767&id=2&as=fa lse&or=false& qs=Ne%3D20%26Ns%3DHarvestDate%257c1%26N%3D429491318 4%2B302%2B51 ,report numberNASA-TM-X-3464,1976. [96]V.Lisianski,H.Yang,Z.Qin,M.Mueller,K.Shin,W.Gar diner,HighTemperature MeasurementsoftheRateCoecientoftheH+CO 2 CO+OHReaction, ChemicalPhysicsLetters240(1{3)(1995)57{62. [97]Y.Levy,E.Olchanski,V.Sherbaum,V.Erenburg,A.Burc at,Modelingof CombustionCharacteristicsandNOxEmissioninHighlyPreh eatedandDiluted AirCombustion,JournalofPropulsionandPower22(3)(2006 )669{676. [98]S.Macadam,Sootsurfacegrowthmechanismsinstationa rycombustionsystems, Ph.D.thesis,MassachusettsInstituteofTechnology,1997 [99]X.Hui,Z.Zhang,K.Mu,Y.Wang,Y.Xiao,EectofFuelDil utionontheStructure andPollutantEmissionofSyngasDiusionFlames,in:Proce edingsofASMETubo Expo2007:PowerforLand,SeaandAir,ChineseAcademyofSci ences,Beijing, China,ASME,Montreal,Canada,2007. 255

PAGE 256

[100]D.A.Wilson,K.M.Lyons,Eectsofdilutionandco-row onthestabilityoflifted non-premixedbiogas-likerames,Fuel87(3)(2008)405{413 [101]R.M.Kowalik,L.Ruth,W.Blazowski,SootFormationin Jet-StirredReactors, chap.SessionIII:FormationDuringContinuousCombustion ,PlenumPress,New York,ISBN0306408813,285{297,1981. [102]M.Huth,W.Leuckel,SootPrecursorParticlesinFlame s,chap.23,Springer-Verlag GmbH&CompanyKG,Berlin33,Ger,ISBN0172-6218,371{381,1 994. [103]F.Pischinger,G.Lepperho,M.Houben,Sootformatio nandoxidationindiesel engines,chap.59,Springer-VerlagGmbH&CompanyKG,Berli n33,Germany, ISBN01726218,382{395,1994. [104]G.Sturgess,S.Heneghan,M.Vangsness,D.Ballal,A.L esmerises,LeanBlowoutin aResearchCombustoratSimulatedLowPressures,Journalof EngineeringforGas TurbinesandPower118(1996)773{781. [105]M.D.Joannon,A.Matarazzo,P.Sabia,A.Cavaliere,Mi ldcombustionin HomogeneousChargeDiusionIgnition(HCDI)regime,in:31 stInternational SymposiumonCombustion,vol.31II,DipartimentodiIngegn eriaChimica, UniversitaFedericoII,Naples,Italy,ElsevierLtd,Oxfor d,OX51GB,United Kingdom,ISBN1540-7489,3409{3416,URL http://dx.doi.org/10.1016/j. proci.2006.07.039 ,2007. [106]J.P.Longwell,R.J.Lang,HighTemperatureReactionR atesinHydrocarbon Combustion,IndustrialandEngineeringChemistry47(8)(1 955)1634{1643. [107]M.A.Weiss,R.J.Lang,J.P.Longwell,CombustionRate sinSphericalReactors, IndustrialandEngineeringChemistry50(2)(1958)257{264 [108]D.Ballal,M.Vangsness,S.Heneghan,G.Sturgess,Stu diesofLeanBlowoutina ResearchCombustor,in:AGARDConferenceProceedings,NAT OResearchand TechnologyOrganisation,536,1993. [109]J.Zelina,D.R.Ballal,Combustorstabilityandemiss ionsresearchusinga well-stirredreactor,JournalofEngineeringforGasTurbi nesandPower-Transactions oftheAsme119(1)(1997)70{75. [110]J.W.Blust,D.R.Ballal,G.J.Sturgess,Fueleectson leanblowoutandemissions fromawell-stirredreactor,JournalofPropulsionandPowe r15(2)(1999)216{223. [111]Y.Lei,Y.Cui,G.Xu,TheEectsofFuelDilutionwithSt eamonGasTurbine CombustorPerformance,in:ProceedingsofASMETurbineExp o2007:Powerfor Land,SeaandAir,InstituteofEngineeringThermophysics, ChineseAcademyof Sciences,Beijing,China,ASME,Montreal,Canada,papernu mberGT2007-27681, 2007. 256

PAGE 257

[112]I.T.Osgerby,Literaturereviewofturbinecombustor modelingandemissions,AIAA Journal12(6)(1974)743{754. [113]H.Mohamed,H.B.Ticha,S.Mohamed,Simulationofpoll utantemissionsfrom agas-turbinecombustor,CombustionScienceandTechnolog y176(5-6)(2004) 819{834. [114]N.K.Rizk,H.C.Mongia,Three-dimensionalcombustor performancevalidation withhigh-densityfuels,JournalofPropulsionandPower6( 5)(1990)660{667. [115]N.K.Rizk,H.C.Mongia,Semianalyticalcorrelations forNOx,CO,andUHC emissions,JournalofEngineeringforGasTurbinesandPowe r115(3)(1993) 612{619. [116]N.K.Rizk,H.C.Mongia,NOxmodelforleancombustionc oncept,Journalof PropulsionandPower11(1)(1995)161{169. [117]N.K.Rizk,H.C.Mongia,Semi-analyticalemissionmod elfordiusionrame, rich/leanandpremixedleancombustors,JournalofEnginee ringforGasTurbines andPower117(2)(1995)290{301. [118]I.V.Novosselov,Chemicalreactornetworkmodelingo fcombustionsystems,Ph.D. thesis,UniversityofWashington,Seattle,WA,2006. [119]I.V.Novosselov,P.C.Malte,S.Yuan,R.Srinivasan,J .C.Y.Lee,Chemicalreactor networkapplicationtoemissionspredictionforindustial DLEgasturbine,in:2006 ASME51stTurboExpo,May6,2006-May11,vol.1,Departmento fMechanical Engineering,UniversityofWashington,Seattle,WA981952600,AmericanSocietyof MechanicalEngineers,Barcelona,Spain,ISBN0791842363, 221{235,2006. [120]R.C.Steele,J.H.Tonouchi,D.G.Nicol,D.C.Horning, P.C.Malte,D.T. Pratt,CharacterizationofNOx,N2O,andCOforLean-Premix edCombustion inaHigh-PressureJet-StirredReactor,in:Proceedingsof ASMETurboExpo 1996:PowerforLand,SeaandAir,vol.128,UniversityofWas hington,Seattle, WA,AmericanSocietyofMechanicalEngineers,NewYork,NY, papernumber 96-GT-128,1996. [121]D.G.Nicol,T.Rutar,S.M.Martin,P.Malte,D.T.Pratt ,ChemicalReactor ModelingAppliedtotheProductionofPollutantEmissionin LPCombustors,in: 33rdAIAA/ASME/SAE/ASEEJointPropulsionConferenceandE xhibit,American InstituteofAeronauticsandAstronauticsInc,Seattle,WA ,UnitedStates,paper numberAIAA-1997-2711,1997. [122]S.D.Stouer,B.M.Mortimer,D.M.Ostdiek,M.J.DeWit t,R.Pawlik,R.Reich, C.W.Frayne,H.Mayeld,Sootreductionresearchusingawel l-stirredreactor,in: 43 rd AIAAAerospaceSciencesMeetingandExhibit,Universityof DaytonResearch Institute,Dayton,OH45469,UnitedStates,AmericanInsti tuteofAeronauticsand AstronauticsInc,Reston,VA20191,UnitedStates,10953{1 0966,2005. 257

PAGE 258

[123]R.Ashen,K.Cushman,ChemicalKineticSimulationoft heCombustionof Bio-BasedFuels,OntheWWW,URL http://www.osti.gov/bridge// product.biblio.jsp?query_id=1&page=0&osti_id=931051 ,reportnumber ORNL/PTS-8239,2007. [124]D.Agarwal,S.Sinha,A.K.Agarwal,Experimentalinve stigationofcontrolofNOx emissionsinbiodiesel-fueledcompressionignitionengin e,RenewableEnergy31(14) (2006)2356{2369,URL http://dx.doi.org/10.1016/j.renene.2005.12.003 [125]C.Hasimoglu,M.Ciniviz,I.Ozsert,Y.Icingur,A.Par lak,M.S.Salman, Performancecharacteristicsofalowheatrejectiondiesel engineoperatingwith biodiesel,RenewableEnergy33(7)(2008)1709{1715,URL http://dx.doi.org/ 10.1016/j.renene.2007.08.002 [126]C.Bolszo,V.McDonell,S.Samuelsen,Impactofbiodie selonfuelpreparationand emissionsforaliquidredgasturbineengine,in:2007ASME TurboExpo,vol.2, UCICombustionLaboratory,UniversityofCalifornia,Irvi ne,92697-3550,United States,AmericanSocietyofMechanicalEngineers,NewYork ,NY10016-5990, UnitedStates,ISBN079184790X,493{502,2007. [127]C.D.Bolszo,V.G.McDonell,Emissionsoptimizationo fabiodieselredgas turbine,ProceedingsoftheCombustionInstitute32(2)(20 09)2949{2956. [128]R.L.Muncrief,C.W.Rooks,M.Cruz,M.P.Harold,Combi ningbiodieseland exhaustgasrecirculationforreductioninNOxandparticul ateemissions,Energy andFuels22(2)(2008)1285{1296,URL http://dx.doi.org/10.1021/ef700465p [129]W.K.Metcalfe,S.Dooley,H.J.Curran,J.M.Simmie,A. M.El-Nahas,M.V. Navarro,ExperimentalandmodelingstudyofC5H10O2ethyla ndmethylesters, JournalofPhysicalChemistryA111(19)(2007)4001{4014,U RL http://dx.doi. org/10.1021/jp067582c [130]N.Hashimoto,Y.Ozawa,N.Mori,I.Yuri,T.Hisamatsu, Fundamentalcombustion characteristicsofpalmmethylester(PME)asalternativef uelforgasturbines,Fuel 87(15-16)(2008)3373{3378,URL http://dx.doi.org/10.1016/j.fuel.2008.06. 005 [131]A.Tsolakis,A.Megaritis,M.L.Wyszynski,K.Theinno i,Engineperformanceand emissionsofadieselengineoperatingondiesel-RME(rapes eedmethylester)blends withEGR(exhaustgasrecirculation),Energy32(11)(2007) 2072{2080,URL http://dx.doi.org/10.1016/j.energy.2007.05.016 [132]P.Dagaut,S.Gail,M.Sahasrabudhe,Rapeseedoilmeth ylesteroxidationover extendedrangesofpressure,temperature,andequivalence ratio:Experimentaland modelingkineticstudy,in:31stInternationalSymposiumo nCombustion,vol.31II, CNRS,LaboratoiredeCombustionetSystemesReactifs,1C,4 5071OrleansCedex 2,France,ElsevierLtd,Oxford,OX51GB,UnitedKingdom,IS BN1540-7489, 2955{2961,URL http://dx.doi.org/10.1016/j.proci.2006.07.142 ,2007. 258

PAGE 259

[133]M.Karabektas,Theeectsofturbochargerontheperfo rmanceandexhaust emissionsofadieselenginefuelledwithbiodiesel,Renewa bleEnergy34(4)(2009) 989{993,URL http://dx.doi.org/10.1016/j.renene.2008.08.010 [134]C.Arcoumanis,C.Bae,R.Crookes,E.Kinoshita,Thepo tentialofdi-methylether (DME)asanalternativefuelforcompression-ignitionengi nes:Areview,Fuel87(7) (2008)1014{1030,URL http://dx.doi.org/10.1016/j.fuel.2007.06.007 [135]M.C.Lee,S.B.Seo,J.H.Chung,Y.J.Joo,D.H.Ahn,Comb ustionperformance testofanewfuelDMEtoadapttoagasturbineforpowergenera tion,Fuel 87(10-11)(2008)2162{2167,URL http://dx.doi.org/10.1016/j.fuel.2007. 11.017 [136]N.M.Marinov,Detailedchemicalkineticmodelforhig htemperatureethanol oxidation,InternationalJournalofChemicalKinetics31( 3)(1999)183{220,URL http://dx.doi.org/10.1002/(SICI)1097-4601(1999)31:3 <183::AID-KIN3>3.0. CO;2-X [137]Y.S.H.Najjar,E.M.Goodger,SootFormationinGasTur binesusingHeavyFuels. 1,Fuel60(10)(1981)980{986. [138]H.F.Calcote,MECHANISMSOFSOOTNUCLEATIONINFLAMES -A CRITICALREVIEW,CombustionandFlame42(3)(1981)215{242 [139]J.Xi,B.-J.Zhong,Sootindieselcombustionsystems, ChemicalEngineeringand Technology29(6)(2006)665{673,URL http://dx.doi.org/10.1002/ceat. 200600016 [140]K.C.Smyth,J.H.Miller,Chemistryofmoleculargrowt hprocessesinrames, Science236(4808)(1987)1540{1546. [141]M.B.ColketIII,R.J.Hall,S.D.Stouer,Modelingsoo tformationinastirred reactor,in:2004ASMETurboExpo,vol.1,UnitedTechnologi esResearchCenter, EastHartford,CT,UnitedStates,AmericanSocietyofMecha nicalEngineers,New York,NY10016-5990,UnitedStates,673{678,2004. [142]Y.S.H.Najjar,R.M.Droubi,Semi-empiricalmethodfo rthepredictionofsoot formationingasturbinecombustors,Hightemperaturetech nology5(2)(1987) 95{99. [143]A.D.Champlain,D.Kretschmer,J.Tsogo,G.F.Pearce, Predictionofsoot emissionsingas-turbinecombustors,JournalofPropulsio nandPower13(1)(1997) 117{122. [144]J.S.Ganey,N.A.Marley,Theimpactsofcombustionem issionsonairqualityand climate-Fromcoaltobiofuelsandbeyond,AtmosphericEnvi ronment43(1)(2009) 23{36,URL http://dx.doi.org/10.1016/j.atmosenv.2008.09.016 259

PAGE 260

[145]M.Kampa,E.Castanas,Humanhealtheectsofairpollu tion,Environmental Pollution151(2)(2008)362{367. [146]B.Dellinger,A.D'Allessio,A.D'Anna,CombustionBy productsandTheirHealth Eects:Summaryofthe10thInternationalCongress,Enviro nmentalEngineering Science25(8)(2008)1107{1113. [147]P.K.Hopke,ContemporaryThreatsandAirPollution,A tmosphericEnvironment 43(3)(2009)87{93. [148]J.Appel,H.Bockhorn,M.Frenklach,Kineticmodeling ofsootformationwith detailedchemistryandphysics:laminarpremixedramesofC 2hydrocarbons, CombustionandFlame121(1)(2000)122{136,URL http://dx.doi.org/10.1016/ S0010-2180(99)00135-2 [149]M.Frenklach,H.Wang,Detailedmechanismandmodelin gofsootparticle formation,chap.59,Springer-VerlagGmbH&CompanyKG,Ber lin33,Ger, ISBN0172-6218,165{192,1994. [150]M.Frenklach,H.Wang,Detailedmodelingofsootparti clenucleationandgrowth, Symposium(International)onCombustion23(1)(1991)1559 {1566. [151]I.Glassman,Combustion,AcademicPress,SanDiego,C alif.,ISBN0122858522(alk. paper),1996. [152]A.Lefebvre,PredictionofSootEmissionsinGas-Turb ineCombustors,Journalof Aircraft21(11)(1984)887{898. [153]Y.S.H.Najjar,EngineeringPredictionofSootConcen trationinPrimaryZoneof theGasTurbinecombustor,Fuel64(1)(1985)93{98. [154]H.F.Calcote,D.M.Manos,Eectofmolecularstructur eonincipientsoot formation,CombustionandFlame49(1-3)(1983)289{304. [155]S.V.Shishaev,Evaluationofthetendencyoffuelstof ormsootandsmokewhen burned.Anewlaboratorymethod,ChemistryandTechnologyo fFuelsandOils 38(3)(2002)207{211. [156]E.Amin,G.Andrews,M.Pourkashnian,A.Williams,R.Y etter,AComputational StudyofPressureEectsonPollutantGenerationinGasTurb ineCombustors, TransactionsofASME:JournalofEngineeringforGasTurbin esandPower119(1) (1997)76{83. [157]M.Frenklach,S.Taki,R.A.Matula,Aconceptualmodel forsootformationin pyrolysisofaromatichydrocarbons,CombustionandFlame4 9(1-3)(1983)275{282. [158]A.Kazakov,H.Wang,M.Frenklach,Detailedmodelingo fsootformationinlaminar premixedethyleneramesatapressureof10bar,Combustiona ndFlame100(1-2) (1995)111{120,URL http://dx.doi.org/10.1016/0010-2180(94)00086-8 260

PAGE 261

[159]N.J.Brown,K.L.Revzan,M.Frenklach,Detailedkinet icmodelingofsoot formationinethylene/airmixturesreactinginaperfectly stirredreactor,Symposium (International)onCombustion1(1998)1573{1580. [160]V.R.Katta,W.M.Roquemore,ModelingofSootinaSwirl Combustor,in:42nd AIAAAerospaceSciencesMeetingandExhibit,AmericanInst ituteofAeronautics andAstronauticsInc,Reno,Nevada,UnitedStates,papernu mberAIAA2004-645, 2004. [161]J.Warnatz,U.Mass,R.Dibble,Combustion,ElsevierS cience,1998. [162]H.Wagner,PlenaryLecture:SootFormationinCombust ion,Seventeenth InternationalSymposiumonCombustion01(03)(1981)215{2 42. [163]T.R.Melton,F.Inal,S.M.Senkan,Theeectsofequiva lenceratioonthe formationofpolycyclicaromatichydrocarbonsandsootinp remixedethanerames, CombustionandFlame121(4)(2000)671{678. [164]S.L.Manzello,D.B.Lenhert,A.Yozgatligil,M.T.Don ovan,G.W.Mulholland, M.R.Zachariah,W.Tsang,Sootparticlesizedistributions inawell-stirred reactor/plugrowreactor,ProceedingsoftheCombustionIn stitute31(1)(2007) 675{683. [165]M.Balthasar,F.Mauss,A.Knobel,M.Kraft,Detailedm odelingofsootformation inapartiallystirredplugrowreactor,CombustionandFlam e128(4)(2002) 395{409,URL http://dx.doi.org/10.1016/S0010-2180(01)00344-3 [166]Y.-N.Song,B.-J.Zhong,Modelingofsootandpolycycl icaromatichydrocarbonsin dieseldiusioncombustion,ChemicalEngineeringandTech nology31(10)(2008) 1418{1423,URL http://dx.doi.org/10.1002/ceat.200800248 [167]K.D.Brundish,M.N.Miller,C.W.Wilson,M.Jeeries, M.Hilton,M.P.Johnson, Measurementofsmokeparticlesizeanddistributionwithin agasturbinecombustor, JournalofEngineeringforGasTurbinesandPower127(2)(20 05)286{294,URL http://dx.doi.org/10.1115/1.1839921 [168]D.F.Kronholm,J.B.Howard,Analysisofsootsurfaceg rowthpathwaysusing publishedplug-rowreactordatawithnewparticlesizedist ributionmeasurements andpublishedpremixedramedata,in:28thInternationalSy mposiumon Combustion,July30,2000-August4,vol.28,DepartmentofC hemicalEngineering, MassachusettsInst.ofTechnology,77MassachusettsAvenu e,Cambridge,MA02139, UnitedStates,CombustionInstitute,Edinburgh,Unitedki ngdom,ISBN00820784, 2555{2561,2000. [169]H.Tuovinen,COformationfromsootandCO2inthehosga slayer.Brandforsk project621-001,OntheWWW,sPReport2002:08,2002. 261

PAGE 262

[170]M.Colket,D.Liscinsky,B.True,AdvancedFuelDevelo pmentandFuelCombustion, OntheWWW,URL http://oai.dtic.mil/oai/oai?verb=getRecord& metadataPrefix=html&identifier=ADA457155 ,2006. [171]M.Frenklach,U.S.N.Aeronautics,S.Administration ,Finaltechnicalreporton kineticsandmechanismofsootformationinhydrocarboncom bustionmicroform, NationalAeronauticsandSpaceAdministration,Washingto n,DC,1990. [172]I.Glassman,Sootformationincombustionprocesses, Symposium(International)on Combustion22(1)(1989)295{311. [173]I.M.Kennedy,Modelsofsootformationandoxidation, ProgressinEnergyand CombustionScience23(2)(1997)95{132,URL http://dx.doi.org/10.1016/ S0360-1285(97)00007-5 [174]B.S.Haynes,H.G.Wagner,Sootformation,Progressin Energyand CombustionScience7(4)(1981)229{273,URL http://dx.doi.org/10.1016/ 0360-1285(81)90001-0 [175]H.Palmer,C.Cullis,Theformationofcarbonfromgase s,vol.1,MarcelDekker, NewYork,265,1965. [176]K.C.Smyth,J.H.Miller,Chemistryofmoleculargrowt hprocessesinrames, Science236(4808)(1987)1540{1546. [177]S.J.Harris,A.M.Weiner,Chemicalkineticsofsootpa rticlegrowth,vol.36, AnnualReviewsInc,PaloAlto,CA,USA,ISBN0066426X;08243 10365,31{52,URL http://dx.doi.org/10.1146/annurev.pc.36.100185.0003 35 ,1985. [178]H.J.Miller,K.C.Smyth,G.W.Mallard,Calculationso ftheDimerizationof AromaticHydrocarbons:ImplicationsforSootFormation,T wentiethSymposium (International)onCombustion31(1984)1139{1147. [179]A.V.Krestinin,Detailedmodelingofsootformationi nhydrocarbonpyrolysis, CombustionandFlame121(3)(2000)513{524. [180]H.Richter,A.J.Labrocca,W.J.Grieco,K.Taghizadeh ,A.L.Lareur,J.B. Howard,Generationofhigherfullerenesinrames,Journalo fPhysicalChemistryB 101(9)(1997)1556{1560,URL http://dx.doi.org/10.1021/jp962928c [181]H.Richter,T.G.Benish,O.A.Mazyar,W.H.Green,J.B. Howard,Formationof PolycyclicAromaticHydrocarbonsandTheirRadicalsinNea rlySootingPremixed BenzeneFlames,ProceedingsoftheCombustionInstitute28 (2000)2609{2618. [182]H.Richter,J.B.Howard,Formationandconsumptionof single-ringaromatic hydrocarbonsandtheirprecursorsinpremixedacetylene,e thyleneandbenzene rames,PhysicalChemistryChemicalPhysics4(11)(2002)20 38{2055. 262

PAGE 263

[183]H.Richter,S.Granata,W.H.Green,J.B.Howard,V.D.K nyazev,Detailed modelingofPAHandsootformationinalaminarpremixedbenz ene/oxygen/argon low-pressurerame,in:30thInternationalSymposiumonCom bustion,July25,2004 -July30,vol.30I,DepartmentofChemicalEngineering,Mas sachusettsInstituteof Technology,Cambridge,MA02139-4307,UnitedStates,Else vierLtd,Chicago,IL, Unitedstates,ISBN15407489,1397{1405,URL http://dx.doi.org/10.1016/j. proci.2004.08.088 ,2005. [184]A.Ergut,S.Granata,J.Jordan,J.Carlson,J.B.Howar d,H.Richter,Y.A. Levendis,PAHformationinone-dimensionalpremixedfuelrichatmosphericpressure ethylbenzeneandethylalcoholrames,CombustionandFlame 144(4)(2006) 757{772,URL http://dx.doi.org/10.1016/j.combustflame.2005.07.01 9 [185]C.J.Pope,J.B.Howard,Simultaneousparticleandmol eculemodeling(SPAMM): Anapproachforcombiningsectionalaerosolequationsande lementarygas-phase reactions,AerosolScienceandTechnology27(1)(1997)73{ 94. [186]J.Z.Wen,M.J.Thomson,M.F.Lightstone,S.N.Rogak,D etailedkineticmodeling ofcarbonaceousnanoparticleinceptionandsurfacegrowth duringthepyrolysisof C6H6behindshockwaves,EnergyandFuels20(2)(2006)547{5 59,URL http:// dx.doi.org/10.1021/ef050081q [187]M.Frenklach,Methodofmomentswithinterpolativecl osure,Chemical EngineeringScience57(12)(2002)2229{2239,URL http://dx.doi.org/10. 1016/S0009-2509(02)00113-6 [188]M.Frenklach,S.J.Harris,Aerosoldynamicsmodeling usingthemethodof moments,Journalofcolloidandinterfacescience118(1)(1 987)252{261. [189]M.Frenklach,DynamicsofdiscretedistributionforS moluchowskicoagulation model,Journalofcolloidandinterfacescience108(1)(198 5)237{242,URL http:// dx.doi.org/10.1016/0021-9797(85)90256-5 [190]M.Frenklach,Computermodelingofinnitereactions equences:achemical lumping,ChemicalEngineeringScience40(10)(1985)1843{ 1849,URL http:// dx.doi.org/10.1016/0009-2509(85)80119-6 [191]F.Mauss,T.Schafer,H.Bockhorn,Inceptionandgrowt hofsootparticlesin dependenceonthesurroundinggasphase,CombustionandFla me99(3-4)(1994) 697{705,URL http://dx.doi.org/10.1016/0010-2180(94)90064-7 [192]M.Dewar,W.G.Jr.,M.Frenklach,I.Oref,RateConstan tfor Cyclization/DecyclizationofthePhenylRadical,Journal ofAmericanChemical Society109(1987)4456{4457. [193]M.Frenklach,T.Yuan,M.K.Ramachandra,Sootformati oninbinaryhydrocarbon mixtures,Energy&Fuels2(4)(1988)462{480. 263

PAGE 264

[194]H.Wang,M.Frenklach,Enthalpiesofformationofbenz enoidaromaticmolecules andradicals,TheJournalofphysicalchemistry97(15)(199 3)3867{3874,URL http://dx.doi.org/10.1021/j100117a038 [195]H.Wang,M.Frenklach,Adetailedkineticmodelingstu dyofaromaticsformationin laminarpremixedacetyleneandethylenerames,Combustion andFlame110(1-2) (1997)173{221. [196]A.Kazakov,M.Frenklach,Dynamicmodelingofsootpar ticlecoagulationand aggregation:Implementationwiththemethodofmomentsand applicationto high-pressurelaminarpremixedrames,CombustionandFlam e114(3-4)(1998) 484{501,URL http://dx.doi.org/10.1016/S0010-2180(97)00322-2 [197]M.Frenklach,Reactionmechanismofsootformationin rames,PhysicalChemistry ChemicalPhysics4(11)(2002)2028{2037. [198]P.Mitchell,M.Frenklach,Particleaggregationwith simultaneoussurfacegrowth, PhysicalReviewE67(6)(2003)061407{1,URL http://dx.doi.org/10.1103/ PhysRevE.67.061407 [199]M.Balthasar,M.Frenklach,Detailedkineticmodelin gofsootaggregateformation inlaminarpremixedrames,CombustionandFlame140(1)(200 5)130{145,URL http://dx.doi.org/10.1016/j.combustflame.2004.11.00 4 [200]M.Balthasar,M.Frenklach,J.Boris,E.Oram,P.Roth, D.E.Rosner,R.Dobbins, Monte-Carlosimulationofsootparticlecoagulationandag gregation:Theeectof arealisticsizedistribution,ProceedingsoftheCombusti onInstitute30(1)(2005) 1467{1475,URL http://dx.doi.org/10.1016/j.proci.2004.07.035 [201]J.Singh,R.I.A.Patterson,M.Kraft,H.Wang,Numeric alsimulationand sensitivityanalysisofdetailedsootparticlesizedistri butioninlaminarpremixed ethylenerames,CombustionandFlame145(1-2)(2006)117{1 27,URL http://dx. doi.org/10.1016/j.combustflame.2005.11.003 [202]H.Richter,J.B.Howard,Formationofpolycyclicarom atichydrocarbonsandtheir growthtosoot-areviewofchemicalreactionpathways,Prog ressinEnergyand CombustionScience26(4)(2000)565{608,URL http://dx.doi.org/10.1016/ S0360-1285(00)00009-5 [203]C.A.Schuetz,M.Frenklach,Nucleationofsoot:Molec ulardynamicssimulations ofpyrenedimerization,in:Twenty-NinthInternationalSy mposiumonCombustion HokkaidoUniversitySapporoJapan,vol.29,CombustionIns titute,ISBN1540-7489, 2307{2313,2002. [204]R.A.Dobbins,H.Subramaniasivam,Sootprecursorpar ticlesinrames,chap.59, Springer-VerlagGmbH&CompanyKG,Berlin33,Germany,ISBN 01726218, 290{301,1994. 264

PAGE 265

[205]R.J.Hall,M.B.Colket,Predictionsofsootparticleg rowthbasedonaerosol dynamicsmodelling,JournalofAerosolScience23(1992)S1 29{S132. [206]K.Neoh,J.Howard,A.F.Sarom,SootOxidationinFlam es,chap.SessionII: BasicMechanismsandKinetics,PlenumPress,NewYork,ISBN 0306408813, 261{282,1981. [207]M.Frenklach,W.C.Gardiner,RepresentationofMulti stageMechanismsinDetailed ComputerModelingofPolymerizationKinetics,JournalofP hysicalChemistry 88(25)(1984)6263{6266. [208]K.L.Revzan,N.J.Brown,M.Frenklach,SootFormation Code,OntheWWW, URL http://http://www.me.berkeley.edu/soot/ ,2009. [209]M.Balthasar,Detailedsootmodellinginlaminarandt urbulentreactingrows,Ph.D. thesis,LundInstituteofTechnology,2000. [210]D.Goodwin,Cantera:object-orientedsoftwareforre actingrows,OntheWWW, URL http://www.cantera.org ,2006. [211]I.Tosun,Modellingintransportphenomena:aconcept ualapproach,Elsevier, Boston,ISBN0444510524(alk.paper),2002. [212]SocietyofAutomotiveEngineers,AerospaceRecommen dedPractice,1994. [213]LawrenceLivermoreNationalLaboratoryCombustionM echanisms: UCRL-WEB-227522Rev15,OntheWWW,URL http://www-cmels.llnl. gov ,2007. [214]H.Richter,S.Granata,W.H.Green,J.B.Howard,V.D.K nyazev,KineticsModels andDataUsedinCalculations(DetailedModelingofSootFor mation),Onthe WWW,URL http://web.mit.edu/anish/www/MITcomb.html ,2009. [215]S.L.Fischer,F.L.Dryer,H.J.Curran,Reactionkinet icsofdimethylether.I: high-temperaturepyrolysisandoxidationinrowreactors, InternationalJournalof ChemicalKinetics32(12)(2000)713{740,URL http://dx.doi.org/10.1002/ 1097-4601(2000)32:12<713::AID-KIN1>3.0.CO;2-9 [216]H.J.Curran,S.L.Fischer,F.L.Dryer,Reactionkinet icsofdimethylether. II:low-temperatureoxidationinrowreactors,Internatio nalJournalof ChemicalKinetics32(12)(2000)741{759,URL http://dx.doi.org/10.1002/ 1097-4601(2000)32:12<741::AID-KIN2>3.0.CO;2-9 [217]E.W.Kaiser,T.J.Wallington,M.D.Hurley,J.Platz,H .J.Curran,W.J. Pitz,C.K.Westbrook,Experimentalandmodelingstudyofpr emixed atmospheric-pressuredimethylether-airrames,Journalo fPhysicalChemistry A104(35)(2000)8194{8206,URL http://dx.doi.org/10.1021/jp994074c 265

PAGE 266

[218]E.M.Fisher,W.J.Pitz,H.J.Curran,C.K.Westbrook,D etailedchemicalkinetic mechanismsforcombustionofoxygenatedfuels,in:28thInt ernationalSymposium onCombustion,vol.28,SibleySchoolofMech./Aerosp.Eng. ,CornellUniversity, Ithaca,NY14853,UnitedStates,CombustionInstitute,ISB N0082-0784,1579{1586, 2000. [219]R.Seiser,H.Pitsch,K.Seshadri,W.J.Pitz,H.J.Curr an,Extinctionand autoignitionofn-heptaneincounterrowconguration,in: 28thInternational SymposiumonCombustion,vol.28,Dept.ofMech.andAerospa ceEng.,Univ.of CaliforniaatSanDiego,SanDiego,CA92093-0417,UnitedSt ates,Combustion Institute,ISBN0082-0784,2029{2036,2000. [220]J.Appel,H.Bockhorn,M.Frenklach,SootReactionMec hanisms,OntheWWW, URL http://www.me.berkeley.edu/soot/mechanisms/mechanis ms.html ,2009. [221]J.S.Bendat,A.G.Piersol,Randomdata:analysisandm easurementprocedures, Wiley,NewYork,ISBN0471040002,1985. 266

PAGE 267

BIOGRAPHICALSKETCH BhupinderSinghwasbornin1979,inthecityBaroda,India,w herehespentmost ofhislife.HegraduatedwithaBachelorofScienceinmechan icalengineeringin2001 fromSardarPatelUniversity,Gujarat,India.HejoinedAls tomIndiaasanengineering traineein2001andcontinuedtoworkasaleadengineeruntil 2004.In2004,hejoinedthe UniversityofFloridatopursueaPhDinmechanicalengineer ing.Hehasbeenpursuing hisresearchintheareaofcombustion,morespecically,on modelingofemissions,soot formationandrameblowoutduringvitiatedcombustion,int hePoWERengine.Broadly, hisresearchinterestsarebiofuels,rameblowout,rameles scombustion,combustion kineticsandsootformation. 267