T "-...;".:- V 325:?!“ 5:.- ,.. 39L.“ ' 5:.- 3;- —._o 3. 3’ 31,1 -‘i’F’:‘ ,,W _.. 3! ‘1' “1.43 3" T 3.H' '01; .u TI. it“; TTLT$3TT1TTTS ‘EQTTTA H}: ”Tr flSTj $35!“, 3" T'i ' :T T'“ TT': TIT/383% 4511. “TAT'T 'T, 3"! U*3 3"“,3-Pi3 “2’": 1'3"” .111: ,3} V k TT'W‘; it: .: 3:3}! Til?” T l‘lifiTT'élTTT'uTT‘STT'T .fi'T'm' +3"? &%:‘“ 7 t 9: ”)18‘1'. l:3n5 {2...- Sr 3 ,; ..' \ '13:? T‘TTTT‘TT'g T . ‘vJ é: . ,, 3 5;; 3;; fizz}; u..]'\ :,ti- !. lisfi ‘ ‘T‘ ' l 31‘: T4h‘ { MW r? TuiT' TT TS 32333;” $913 335:3: '. LIP 33:" U. 3;. ‘ ‘ , ”T'OT'T'T'T T'T'W N TUT w}! T??? ‘T‘l"!T"é">T 133T T'WT 'tTT ,- 'w: _,“':- o. 3- ..T‘.T‘1"T,'"T'T'-!‘333'TT‘: '33?" T -" 3" iL. :‘hfi'vhw V onk‘l :a‘z? beL¥T aq'“ LIT: 3"; ‘3‘ z‘L '2! 'Tgk'Tg'TKT 1&3??fo T‘"' T w ‘T 'T T‘ 'TT‘T'T‘TT m m; 'T TT'T- L" "T A. ”'T' 'T'3T. TTT 'T.2 31.! ~93 " ‘iJ-TTI 333 ~ ‘2',' 1.1.0'Uvfllg‘. w ,3. TM ""333 ‘TnTH ,t 1313‘ U 3!!!! Mai. T ' . . ..... . -. 3%,?!" .‘t‘. ' 1.3T? MT “cTzT 13;“? .,: T3591: “35"{331’15'5‘ 5f“ ,- TIT xi??? lights 393% “‘1 3 S;T"_ “Ff '. 33' fail,“ n,” :‘JJ'TF 3 z’ft'zf‘figt L. T-T} ' pufiaz, .: I , {TET i'T'l' iii“. TTLT'T ‘5‘: ‘3: Mr ‘EPTTTTE; “[0,; kt, a? gal TETT- gm}. TT'T £5332:pr ;i'TT'i'. 3‘03-FTTTT' 'Tfitna‘higp: w'uquT W* 33 T flTWW$ :W' ”%%w wan: 2:31» 32:; 3w :13; i'gazzafifig..3~ ' z ' ' ‘ ,‘TT-Tm “.131‘géxig'TLb13“ . "HT 3 . 7:. '\ . TTTT . , f3" . ‘Tz'h T'TT' 1 3 93‘? 'T’ TTT) 3i L! . ,(I'. ._ 3 " P. "' :31 £3!“ gt“. 5' “RTE l'TITi| 'T 3&8." '2‘: 3?. l.) T? ; . “‘11:”? 3.; . 2 '5": TTTJT'TT'E‘T 1.3"? 3' ' ‘1 M . {$35.53 2'! 3 "’3! ”A. '1 n. “9. T:L'T'TTT f"TITTTT: ' "NAT? ' 43E! 3 -. .: (Him: {33“}? ‘ .Tflif ; ' WET”; b g ,' ,T ~33T3d THE-5&3??? ' 7-”... r35 2'3“} 3 - "3 ‘hf-u TY ~13‘t23 {I‘TTTM‘TN 12.31“? T he 3% . , ‘ an?" , - 5‘35; ' :t $fifiMme Tmmw” ,3fiw,.' -$mwt_ % 3%,}; M13742“??? :TTtt-tgJ .6 "'TTTT'TI 3592.], ' ”TTT ' . ' . .-3‘ 39,3”. , 3| 'TT'TT'TT .TT-xTnbfila 331W 2“.” T .- ' .- 1 3 f. wag: _ I 3.51:3“; ; 2'3 'T:TTT': g, 3 . *éT TIT, 3' .-3 - 1 ' I :4' T ‘3 ‘I. . flWHwMTn THanwwflw ' w" .prfl ‘ 113T?! Tzfiizin' T 7333’ -§T-‘ '.. “ ' ' "" «3.14m». ' .“ 3 a: ,1 ~ ~ u ~ . . 3 {3,1 'TTTT’TLTT'TT." . l #me,w wme3. ' '. Ti'l'JTtE :TQ;_ ‘ '3) :v ". *- .__. w A-.._.—v-- .. « ._ -,- - “~77? _ . 1‘1. 4‘ 0" " >. . "33. —, -‘ - _ 3 . v" ‘_., . _. 2 .3 L“, a???” l..:.‘ I“ r T}. T, ET; l'TaT 1313, TINT 'iT’Tt N T 3 3 “II'IITT 'TT'” I'! 3T" Effie, , --. 3 1),. z 'T 3 '3‘” 3313 'H' T 313:?! ‘ 3 'T' , v I . yw'i1}315‘ T31. '1'“);an M. _ .3? . “Mus T?! “1.1!,” 'TT‘T'fi I b1 hfiTT'R‘J TTTT'T'T rr' 'TT":M 3.' ‘31-”; ‘. '3'.~'.. ‘ ' " w '5' "d1 '3: " "'T"'T -'v3.‘ 7' ' '3 ..- “AMT" TT:...' "~""'"‘4§'c.3.)31;".1~.§‘-._‘.~.;‘ “-. A“??? ”"9 H ' ' “ '- élET' .13.,rTJ .. 3'2“.“ TT 'TT' ‘ lTj'T t;T T T'TTTT M. 'T'TT . 3 ‘3' 'TT'M 'TTT' ' T', "5T3!" TNT}. T TTITHTT '3' all fnT'T TTTT' :3:;. £7 {@333 TH 31"“,3 .- TQT'hT f, .33 ;:.3 4|!an». z is”; !' .zs- 71‘3“! 3” I W T'TT Max 3 !"13T“TI'Asr-3-"'T" " ,.,, 2:: mm 'r'TTTT' 3. h“. 13:: ". M” ' . 3. 4' 1,;1. ,- T TT “34".: 1,1393%, ' . M3, ""T T! 3 T I T ‘TT TT " TFT "T"! ' .l '3 , , 3 , .:. I .‘x .1 ' . 'T":r::""25"];:" ”5;; :T 3 "11.5 ."[".’."" 1T. -:~T:"T's'TT,3.':,,"JT 5.; TTF'T'T" ' 3.. .3333 331? 3'1! 3:533??? 3,3: 1'37’33?‘51"T.‘3"31"T 331,, g, TZ‘TTi' 3 T'TT'T'I. TT TT'TTT 'TiT'T ... id, '3"; 3', T?!" '6" .3T'.‘ " iTlTT ‘T 'TTT ’11 "I‘ 3‘3 IT; TIT?” T'" 3fTTIAITITb "' T" T " T.'T ST . 'TTMWT TT'T‘TT T'! T T? T"? T'""'"' 3TT3T T" ""T?' " T'T' r‘T ‘1? 3T???” 31:171.... .3 3 nv TaThJ" E'E'J I“, QT.“ TWW“'flL,,¢WW flrTstT MT , I TTT'TTT Hf: LT'T $3: TTT TTII"?}3 '..|‘ t” a”: l . "" TT‘ '{Tngw TT :3" T‘Tf".".TE‘... . TTTTTTT m, .3T,,1,3T In??? an $31 LAM?" W1; “17...!” T? ”T'T'TT'T 'T'T" TTTMTM Ti, THESIS TATE UNIVERS Ilililll‘l’ill Ill Ililillllll Ill 3 1293 01694 3569 This is to certify that the dissertation entitled VISUALIZATION OF TWO-PHASE FLUID DISTRIBUTIONS USING LASER-INDUCED EXCIPLEX FLUORESCENCE presented by J ongUk Kim has been accepted towards fulfillment of the requirements for Ph.D . degree in Physics M ‘ 4 I‘M/QC: M\\ \ M a jor professor / a,/ Date QAJ‘S/{WT’zi iqq7 \ i (J MSU i: an Affirmariw Action/Equal Opportunity Institution 042771 LIBRARYI Michigan State University PLACE lN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE 1!” WM“ VISUALIZATION OF TWO-PHASE FLUID DISTRIBUTIONS USING LASER-INDUCED EXCIPLEX FLUORESCENCE By JouUlt Kill A DISSERTATION Submitted to W State University in partial fulfillment of the regain-at: forthcdqneof DOCTOR OF PHILOSOPHY Depart-at of Physics ' 1997 ABSTRACT VISUALIZA‘I'ION OF TWO-PHASE FLUID DISTRIBUTIONS USING LASER-INDUCED EXCIPLEX FLUORESCENCE By JongUk Kim Experimentaloptieal methodshavebeendeveloped fortheptn'poseofvimalizing two-phase fluid disu'ibutions. Laser-induced exciplex (excited state complex) flmrescencehasbeenusedmgenauetwodimensimdimagesofdispasedliqtndmd vaporphaseswithspectrallywellresolvedtwo-coloremissions. Inthismethodthevapor phaseistaggedbymemonomerflumescencewhfletheflquidphaseisnackedbythe exciplex fluorescence. For the purpose of observing the highly turbulent dynamic behaviorofinjected liquidthroughasmallorifice, exciplexviwalizationptovidesanon- inuusivediagnosdcwithgoodspecnalandspafialmohmomAnewexciplex visualization system consisting of DMA (N,N-dimethylaniline) and 1,4,6-1'MN (trimethylnaphthalene) in an isooctane (2,2.4-trimethylpentane) solvent was developed. Among the many formulations tested. a 5%DMAe5%l,4,6-TMN exciplex system, in 90%ismcnnesolvenhshowedomstmdingopfiealchmactedsfimThekineficsof exciplexfotmationanddeesyhavebeenexaminedasafimction oftempetatureand preeaIemdemflmdmlnedmmeMOphysiesofmesystem.Thedirectcalflnfimof theflumueenceintensityasafimcfionoftheflmreecmgdopmtconeennafimsthen permiuedthedetaminafionofqumfiunveconeennafionmapsofliqindandvapor phases in the flow-field. The exciplex visualization methods developed have been applied toexaminealiquid-vaporspraywhichevolvesrapidlyintimemdspace.Theremlts have a direct bearing on the fuel injection procees in the direct-injection spark-ignition engine and will lead to improved mderstanding of engine efficiency and the formation of pollutants. To my parents & my wife for theirencouragement. faithmdsupportthroughommydificultdays. ACKNOWLEDGMENTS FnsdyonuldliketothankmyadvisorstfeesoangeGoldingandefessOt mudmemmmmmmfimmmw Theirmggestionsandinsightshavegreettyenrichedthisreeearch. IwonidiiketothankDr.Nocuaforhisvalmblemggesfions.Hegreatlyheiped memmdasmdmebadsofChemimymmyMImddlikemalsothmkmmy gtddmcecommineemembasDr.Mahnfi,Dr.Birge,andDr.chhesfahmiformeir helpfidmggesfimDrsGfletheretondeamchulLeefortheirWefl‘onm fisteningmmyquesfimsmdprovidingmemmchedmalsmmismearch ThanksgoesmlonDarrowathishelpmdesigningandblfldingthetestengine assemblyandcotmtiesshomsinthenightfortakingdata.Aspecialthanksgoesto‘l‘om SWforhiseneomagem&h&isresearch.“Dr.Kim”or“measorKim”-his jokeheldagreatpowerwbenevu'lhadhardfimes;Myfiiend.SangilHym.forhis fi-iendshipespeeiallydin'inghardtimes;andmy labfriends. LowellMcCann. Amy EngebremomWenhaonMmkNovahEficmLmyDafimmmMikhail Ejakov,andHansHascher,fortheirefi‘ortsandfiiendshipinheipingmyresearchand bemingmymskifledEnglishAgreatthanksgoesmBobbieSliderforherdmein correctingEngiishmdhelpingmegetthingsdonequickly,easily,andwithasmile. Mygrmnappreciafimgoesmmypcenmespeciaflythelateth-Baem myfmther,whogavemeloveandreallywantedmysuccess.Mybrothersandsister, for menamponmdenmgememmroughommismchMylovemdspecialmanksgo mmywifeGymymanmhuendlmlovecomdeeshmdnmamdgreasauifice. Mylovelydmghtermdsomfortheirtmdermdmgmdfmgivingofme'huydad.” IthmktheMSUCenterforFmdamenthatefialsRmchtheNSFCenterfor SensorMaterialsatMSU,andtheChrysletcompanyforfinancialsupport iv TABLE OF CONTENTS LIST OF TABLE LIST OF FIGURE CHAPTER L INTRODUCTION 1.1. Background and Motivation 1.2. Application of Exciplex System 1.3. Overviewofl’resent Research CHAPTER II. BACKGROUND AND THEORY 2.1. Review of Flow Visualintion Methods 2.1.1. Fuel Spray Visualization Using Fluorescence and Other Optical Methods 2.1.2. Flow Velocity Field Visualintion (PI'V, PIV and MTV) 2.2. Photochemistry 2.2.1. Spectroscopy 2.2.2. Quenching Mechanism A. Steady-State Quenching B. Time-Resolved (Dynamic) Quenching 2.2.3. Excitation and Emission Dynamics 2.3. Laser Induced Fluorescence (LIF) 2.4. Thermodynamics 2.5. Droplet Vaporintion 2.5.1. General Characteristics 2.5.2. Shple Model of Droplet Vaporintion A. Simple Model of Draplet B. Droplet Vaporization a. Heat-Transfer Controlld Vaporintion b. Mass-Transfer Controlled Vaporintion 2.5.3. Kinetics of Droplet Vaporization A. Gas Phase Equations 8. Liquid Phase Equations 2.6. Exciplex (Excited State Complex) Fluorescence 11 ll l3 I3 18 19 21 27 3O 30 32 32 33 33 34 35 35 37 CHAPTER III. EXPERIMENTAL METHODS (I): DEVELOPMENT OF NEW EXCIPLEX SYSTEMS 3.1. Chapter Overview 3.2. Development&Clnr-acm'izationofNewExciplex Systems 3.2.1. Optical Characterintion A. Sample Preparation and Apparatus B. Spectroscopy of the Sample C. Quantum Yield of Sample 3.2.2. Optical Properties of Exciplex Farmers A. Room Temperance Meaanements A-l. Photophysics of DMAOTMN Exciplex System a. DMA'(2.3.6-TMN) Exciplex System b. DMA-(l,4,6-TMN) Exciplex System c. General Discussions of Exciplex Remits A-2. Exciplex Quenching Analysis a. DMA0(2,3,6-‘1MN) Exciplex System b. DMAo(1,4,6—TMN) Exciplex System A-3. Fluorescence Lifetime Measurement 8. High Temperature Measurements B-l. Sample Preparation and Apparatus B-2. TemperatureDependent Fluorescence Quenching 3.3. Application of Exciplex Fluorescence to Thermometry 3.3.1. Overview of Exciplex Thermometry 3.3.2. DMAe'I'MN Exciplex Thermometry CHAPTER IV. EXPERIMENTAL METHODS (II): PHYSICAL CHARACTERIZATION OF THE EXCIPLEX 4.1. Chapter Overview 4.1.1. Exciples Emission in Liquid and Vapor Phases 4.2. Optical SeurpforTwoPhaseFluorescenceIntemityCah‘bration 4.2.1. General Field-Optics 4.2.2.Field of View of Image 4.3. Intensity-Concentration Calibration 4.3.1. Vapor Phase A. Absorhance Measurements (Indirect-Pressure Measurement) a. Sample Preparation and Apparatus b. Calibration Procedure c. Image Acquisition and Analysis 8. Direct Pressure Measurement a. Pressure Sensor Calibration 45 45 45 49 5 1 5 1 5 1 55 59 6 1 65 67 69 69 71 73 74 78 79 81 87 87 92 93 93 98 98 b. Experimental Apparatus for Pressure Sensor Calibration c. Calibration Procedures 4.3.2. Liquid Phase 4.3.3. Result and Discussion A. Vapor Phase a. Absorbance Measurement b. Direct Pressure Measurement c. Analysis of Calibration data d. Comparisons and Discussions B. Liquid Phase 4.4. Experimental for Actual Measurements 4.4.1. Fuel Spray LIEF Imaging in a Test-Qumtz Cylinder A. Fuel LIEF Method B. Experimental Setup 4.4.2. ln-Cylinder Fuel LIEF Imaging in a Motored Engine A. Engine Assembly and Optical Chamber B. Intake System C. Direct Injection Fuel Control & Laser Synchronization CHAPTER V. RESULTS AND INTERPRETATIONS 5.1. 2-D Section Images ofFuel Spray in a TestQuartz Cylinder 5.1.1. Fuel Spray LIF Imaging 5.1.2. Fuel Spray LIEF Imaging A. Time-Dependent Analysis of the Spray B. Simple Model of Droplet Vaporization 5.2. Direct Engine Applications 5.2.1. In-Cylinder Fuel Vapor Calibration 5.2.2. Actual Engine Measurements and Interpretation A. Liquid Phase Fuel Distribution B. Vapor Phue Fuel Distribution CHAPTER VI. CONCLUSIONS AND FURTHER RECOMMENDATION 6.1. Summary and Conclmions 6.1.1. Development of New Exciplex System 6.1.2. Calibration of Exciplex Fluorescence A. Ambient Condition B. Motored Engine Condition vii 101 103 104 109 109 109 112 115 117 122 127 127 127 127 133 133 138 141 142 142 142 147 160 164 169 171 179 182 187 192 192 192 193 193 194 6.1.3. Resultsoftbe Fuel SprayinaStmicQuartszlinder 194 6.1.4. Results ina MotoredEngine 195 6.2. Further Recommendations 196 APPENDIX 198 A.1.EquipmentConnectionforLaaerlnducedF1uorescencelmaging 198 A.2. Operational Information forthe Laser Induced Fluorescence ImagingSystem 199 A.3. Cah’brationProceduresin ln-CylinderMeuurements 209 REFERENCES 212 viii Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Table 9 Table 10 LIST OF TABLES Relative fluorescencequantrnnyieldofinvestigatedmonomers ..... 51 Monomerandexciplex emissionwavelengthofme investigated exciplex systems 65 PicosecondfluorescarcelifetimesofSV. (wt) DMAand DMA01,4,6-TMN exciplex as the wt % of 1,4,6-TMN is varied. 72 Temperature-dependent equilibrirnn constant 1((1'). 77 Praemonomerflflandexciplexfluorescenceaaintensityand theirratiosasafimctionoftemperature.Bothofthe intensities wereobtainedbyusingtheGaussianbandslnpeanalysis. .............. 83 The boiling points of individual exciplex-forming dopants and thelatentheatofvaporintiom - 121 Trmsienttempaannepressrnemixnnedensityand fluorescenceabsorptionefi'ectintheemism'onfluorescence intensity - - .......... 178 Average intensity gain versus potentiometer setting. ...................... 203 Gate pulse width versus potentiometer setting. - - - ..... - 205 Delaytime versus potentiometer setting. - - - - .-207 LIST OF FIGURES Figure 1. Statediagramforamoleculeshowingthevarious intermolecularproceasesresultingfiomtheabsorption ofradiation.” - - 12 Figure 2. Abbreviated version of the preceding Fig. 1.” - -- -- _ 14 Figure 3. Two-level model for laser-induced fluorescence. W.2 and W2, representtherate(s")ofstimulatedabsorption andemission. AnistherateofspontmeorrsemissionandQnistherateof Figure4. Sketchofthedropvaporintionprocess.” WhereTisthe temperahnerfisthedropletradius, v,istheradia1velocity, andY.andY,aremassfiactionofambientgasandfireL Figure 5. The mechanism of exciplex liquid and vapor visualimtion systems”- - - -- - - _ _ _- - - ------42 Figure 6. Excipiexenergydiagram." - - - _- - - -- - - 43 Figure 7. Absorptionandemissionspectrafor (a) DMAand(b) DEA monomers in We at room temperatures. ....................... 47 Figure 8. Absorption and emission spectra for trimethyl-naphthalenes (a) 1,4 ,,6-TMN (b) 1 ,,4 5-TMN, and (c) 2.3,6-TMN in isooctane at room temperature..- _ _ _ _ 47 Figure 9. Absorption spectrum of the We (2,2,4-trimethylpentane) solvent at room temperature. - - - - 48 Figure 10. The photo-chemistry of DMA¢1,4,6-TMN exciplex system. .......... 53 Figure 11. The ground-state configuration of the DMA01,4,6-TMN exciplex system in isooctane solvent, at room temperature. ............ 54 Figure 12. The fluorescence spectrum of5%DMA and the fluorescence spectra upon the addition of2,3,6—TMN in isooctane at weight ratios of (a) 0.1%, (b) 0.5%, (c) 1.0%, (d) 3.0%. (e) 5.0%. (t) 7.0 % and (g) 10.0%. The fluorescence intensities of monomer and exciplex are relative to each other. ........................... 56 Figure 13. Figure 14. Figure 15. Figrn'e 16. Figure 17. Figure 18. Figure 19. Figure 20. Figure 21. Fluorescenceintensitychangesofthe5% DMA(a)and the resulting exciplex upon the addition of 2,3,6-TMN inisooctane(b). (a): monomerintensitychangeat343 nm. (b): exciplexintensityclnngeatatol nm,rsspectively - - _ - _ 56 The fluorescence spectrum of 5%2,3,6-TMN and the fluorescencespectraupontheadditionofDMAinisooctane at weight ratios of (a) 0.1%, (b) 0.5%, (c) 1.0%, (d) 3.0%, (e) 5.0% and (t) 7.0 %. The fluorescence intensities of monomerandexciplexarerelstivetoeschodrer. ........................... 58 Fluorescence intensity changes of the 5% 2,3,6-‘1'MN (a) and the resulting exciplex upon the addition of DMA inisooctane(b). (a): monomerintensitychangeat347nm, (b): exciplex intensity change at 400 nm, respectively. ................... 58 Absorption and emission spectra for (a) 1,4,6-TMN and (b) DMA monomersinisooctane(0.8uM)atroomtemperatine .................... 60 Thefluorescencespectrumof5% DMAandthefluorescence spectra upon the addition of 1,4,6-TMN in isooctane at weight ratios of (a) 0.1%. (b) 0.5%, (c) 1.0%, (d) 3.0%, (e) 5.0%, (t) 7.0% and (g) 10.0%. The fluorescence intensifies of monomerandexciplexarerelativetoeachother. - - - - 60 Fluorescence intensitycbangesofthe 5% DMA(a)andthe resulting exciplex upon the addition of 1,4,6-TMN in isooctane (b). (a): monomer intensity change at 343 nm, (b): exciplex intensity change at 405 nm, respectively. ................... 62 Fluorescence spectra of (a) 5% DMA and the resulting exciplex fluorescence spectra upon the addition of (b) 7% 1,4,6-TMN, (c) 7% 1,4,5-‘1'MN and (d) 7% 2,3,6-TW in isooctane. Relativespectralintensitiesarepresented. -- - - - - - ..... 62 Firmescence spectra of(a) 5% 1,4,6-TMN and the resulting exciplex fluorescence spectra upon the addition of(b) 5% DMA. (c) 5% DEA in isooctane. Relative spectral intensities are presented. ...................................... 64 Fluorescence spectra of (a) 5% 2,3,6-TMN and the resulting exciplex fluorescence spectra upon the addition of (b) 5% DMA, (c) 5% DEA in isooctane. Relative spectral intensities are presented. ...................................... 64 xi Figure 22. Figure 23. Figure 24. Figure 25. Figure 26. Figure 27. Figure 28. Figure 29. Figrn'e 30. Figure 31(a). Figure 31(b). A modified Stern-Volmer plot with remeet to the molar concentration of the quencher; (a) 2,3,6-TMN, (b) DMA quencher, respectively. The solid line is a fit oqu.(47)and(48)tothesedats.--- - -- -----68 Aplotofthe MdataofFigJ7versustheconeentration of 1,4,6-TMN in isooctane solvent. The solid line is a fit of the modified Stern-Velma Eq.(47) to these data- - - 68 Time-resolved emission fluorescence spectroscopy of (a) instrument rise-time response, (b) 5% DMA monomer and (e) 5% DMA-5% 1,,4 6-TMN exeiplex m isooctane solvent”... 72 The heating elnmber for high temperature exeiplex fluorescence measurements.-- - ..... - -- 75 Fluorescence spectra of isooctane solutions of DMA-1,4,6-TMN (5% w:w) at (a) 28°C, (b) 60°C, (c) 100°C and (d) 150°C. The fluorescence spectral intensities are relativeto eachother. - - - - - - - - 75 Van’t Hofl' plot of equilibrium constant for exeiplex 5%DMA05%1,4,6-TMN formation. The exeiplex concentrations weredeterminedfromthe integratedintensityofthe deconvoluted exeiplex fluorescence bands of Fig. 26. .................... 77 Thechangeoftotal fluoresceneeintensity(a)and corresponding wavelength change at peak intensity (b) inFig. 26withrespeettotemperature. Thetotalintensitiesare relative to eachother. - - - - - - - 82 Thefluorescenceintensityratioflwm] ofexcipiextothe monomerwithrespeettoincreasingtemperature. Thefluoreseenceintensitieswereealculatedbasedon Gaussianbandshapeanalysis'l'hesolidlineisadatafitequation..82 Vapor and liquid phase fluorescence emissions (A: vapor phase, B: liquid phase). 5% DMA-5%1,4,6-TMN exciplexsystemmW/oisooctanesolventwasused. ...................... 86 Sebematicdiagramofcah‘brationexperiment offuel vapor and liquid droplet. -- ..... _ 88 Schematicdisgramot'fieldoptiestomakeasheetoflight. (a) and(b)areinitialandfinalbeamwidths,6.6mmand xii Figure 32. Figure 33. Figure 34. Figure 35. Figure 36. Figure 37. Figure 38. Figure 39. Figure 40. Figure 41. Figure 42. Figure 43. Figure 44. Figure 45. 800nm, respectively. L, and14arefusedsi1icaconvexand concave lenswrthfocsl leughf; =300m. f, =-37.5mm, BauQassfilterSS- transmittance (Azformphueflz forliquidph-se) ------ - Imagingfieldofview.'” DMAssmpleabsorbsucemessIn'ementinstrlnueutusinga amperauue-conu'olledsampleheuingsystemwith speetrophotometa. Experimental setup fordetermining fluorescence intensity ofDMAvaporplnse (sample heating system with insulation removed). -- A sehematiediagramofMPXlelO-D pressrn'e sensor calibration experimentsetup. MPX-ZIOOD presstrn sensor calibration result. Solid line is dredatafitequarionwithy=-0.0524x+39.,6 max.devistiouof0.343. - ,- - -- Vaporphasccslibratiou instrumentation- - - -- - Experimental arrangement for droplet calibration'” Schematic diagram of the droplet formation module of the isolateddropletgenerator.” - - - _ - - AviewofflumescemimageofisolateddroplaThedroplet was made by using a 5%DMA05%1,4,6-TMN exeiplex” system in 90% isooctane solvent. -- ..... OOOOOOOOOOOOOOOOOOOOO AzoomofsomeiudividualisolatedrhopletinFig.41.CSMA sofiwarewasusedtomakeazoom- - - - - PlotsofDMAvaporphaseabmbaneeswithrespeettothe wavelengh (um). Absorbsnce was measured with increasing temperatrneatsZO°Cintervalfiom30°C(A)t0200°C(l). ....... Aplotot‘DMAvaporabsorbauceversmtemperamreat308nm excitationline.’1‘hisplotwasmadebyusingFig.43. ................... AplotofDMAabsorbanceversusconceutratiouto calculate molar absorption coeficieut. A is for liquid phase, 89 -95 97 -99 100 105 -106 107 107 .110 ..110 Figure 46. Figure 47. Figure 48. Figure 49. Figure 50. ‘Figrn'e 51. Figure 52. Figure 53. Figure 54. Figure 55. Figure 56. and(B)isforgaseousphase.Thefittingconstantare3700 and482(Lmol" cm“)forhqmdmdgasemuphase respectively.-- -- -111 Vaporpressureotemperstra'ephasediagramoftheexciplex forming dopants; (A): isooctane, (3): DMA, and - (C): l,4,6-TMN.Solid line is the Clausius—Clapeyrou Eq. ............... 111 AplotofDMAvaporphasemolarconcentration versustempersture. -- - - 113 AplotofDMAvaporphaseabsorbmcewimrespectto molarconcentration.” -- _ -------ll3 AplotofBeer-LambertlawofDMAvaporphase. Molar absorption coeficient (a) of DMA vapor phase 15 formd tobe482(Lmol"em'). - - ----114 A plot of fluorescence intensity versus concentration of DMA monomer. The data points were fitted to Eq. (17) (solid line). ........ 116 FluorescenceinteusityasafimctionofDMA vaporconcentration. - - - -- - -- 116 DMAvaporpressrre-temperatm'erelatiouship:(A)isfi'om absorbancemeasuremeut using Melton’s method. (3) is the referencedata,"°and(C)isfiomthedireetpressmemeasin-ement. Thesolidlineistheliuearregressioubasedon The Clausius'Clapeyron relation of the exeiplex forming dopants; (A) 1 ,4,,6-TMN (B) DMA, md(C) mooetane, respectively.- - -- -- - - - -- 121 Fluorescence intensityasafimctionofdropletmass. ...................... 123 Laserbesmintensity profile“-.- -- - - - -- 123 Saturation curve for the fluorescence of 5%DMA05%TMNO90% isooctane exeiplex system versus laser energy. Exciplex liquid stream (50m in dis. 20.6mm long) was excited with 308(nm) fiom exeimer laser. The DMA number of moles is approximately 8.8526 x 10" in that volume. ................................... 125 xiv Figure 57. Km 58. Figure 59(a). Figure 59(1)). Figrn-e 60. Figure 61(a). Figure 61(b). Figure 62. Figure 63. Figure 64. Figure 65(a). Figure 65(b). Asequeutialsehematiediagramoffuel LIFrmageacqursruou SebematiediagramoffirelLlFimagingfield—opties, including firelinjeetorandopticallyaceessiblequartzcylinder. Tbefinal besrnsteeringmirror(M3)movesthelasersheetthrough thequartzcyliuder The Chrysler 2.4 L double overhead cam direct-injection engine head. - Asebematiediagnmofthe2.4LCbryslereuginehead showingtwo intake (A), andexhaustCB) valves. Fuel injectorislocatedintbecenterm,andtwosparkplugsare sinratedinthelefiandrightcorners(8)... - Imageacquisitiou experimentalsetup foraetualmeasuremeuts in real engine operation. Chrysler engine head, field - optics (left side), ICCD detector, optical quartz chamber includingmirrorandpistomandAVLengineareshowu. .............. Apictureofsufaeemirrormormtedata45°mglewiththe axisofpistoumotiou. A schematic diagram of a side view of optical dumber-(above); elongatedpistomsurfacemirrormudquartzeylinderareshowu. WiredieOrifices.- - - - -- ......................... Sequentialfuellaserinducedfluoresceneeammgesafier fuel mjecnou(t.,) (visualizatiouwasmadebyusinga 5% DMA/95%isooctanefirel) - -- - - TheeontornmapofthefirelLthnageintensityinFig63. .......... Liquidphasefirelsprayvisualizationusingalaserinduced exciplexfluorescencevisualizatioutechnique (5% DMA-5%1,4,6-TMN exeiplex system was used). .................. Vaporphasefuel sprayvisualizatiouusingalaserinduced exeiplex fluorescence visualintiou technique (5% DMA-5% 146—1'MN in 90% isooctane exeiplex system - -- - 128 - 131 - 134 -- 134 . 135 137 137 .140 - 143 .144 . 148 151 Figure 65(c). Contour map of the liquid phase fuel spray in Fig. 65(a). ............... 154 Figure 65(d). Contour map of the vapor phase fuel spray in Fig. 65(b). ............... 157 Figure 66. Figure 67. Figure 68. Figure 69. Figure 70. Figure 71. Figure 72. Figure 73. Figme 74. Figure 75. Figure 76. Temporal evolutiouofthe liquidphase fuel spraywith the distance firm: the tip of the injector; (a): 30, (b): 50, (c): 90, (d): 120, (e): 150, (f): 170, and (g): 190 pixel, respectively. ..... 161 Normalized liquidphasefluoreseeuce intensitywithrespect tothedistaneefiomtheiujectortipatfixedtimeafterfirel injection; (a): 0.78 ms, (b):1.11 ms, (c): 1.56 1118.01) 1.78 ms, (e): 2.0 ms, and (t): 2.44 ms after start of injection. 161 Temporal evolutionofthevaporphase firelspraywith dredistancefi'omthetipoftheinjeetor; (a): 30, (b): 50, (c): 90, (d): 120, (e): 150, (t): 170, and (g). 190 pixel, respectively. 163 Theoretical calculation of droplet evaporation;” (a): small droplet, (b): large droplet, respectively. 166 Qualitative size-dependent droplet evaporation; (a): small droplet, (b): large droplet, respectively. 166 A plot of ( 1-P/P_)‘ as a frmctiou of time (vaporiution time). (a): x = 1/3, mall droplet, (b): x = 2/3, relatively large dropletanalysis. - - - - - - - 168 Aplotofpressmeversuscrank-angledegeeusingaCbrysler 2.4L 16-va1ve. The engine speed was 600 rpm. 170 Asebematicdiagramofanin-cylinderfirelvaporcah‘brafion experimentalscurp.Thefinalbeamsteeringmirror(M3) movesthelasersheetthroughthequartzeylinder. 172 A typical ofDMA vapor phase calibration image (5%DMA in isooctane solvent : premixed & preheated) using a 0.004” orifice and 80 psi fuel delivery pressure.(0.004" - 80 psi) ............... 173 Aplotofisooetanemassflowrsteasafimetiouof appliedpressure; (a): 0.003 ,:(b) 0.,004” and(e): 0.005” orificesinrespectively. 175 AplotofuormalizedintensityofDMAvaporphaseasafrmctiou of crank-angle degree for a various equivalence ratio“); (a): 1.09, (b): 1.,031 (c): 0.954, (d): 0.862, (c): 0.779, (f): 0.682. and (g): 0.563, respectively. - 175 xvi Figure 77. Figtn'e 78. Figure 79(a). Figure 79(b). Figure 80(a). Figure 80(b). Figure 81. Figure 82(a). Figure 82(b). Figure 83. Figure 84. Figure 85. Figure 86. Fluorescence intensity as a function of equivalence ratio“) for various crank-angle degree (intake process). Fluorescence intensityasa fimetionofequivalenee ratio (0) for various crank-angle degree (compression stroke)- lu-cylinder liquid phase firel distributions um’ng a 5%DMA05%1,4,6-TMN090%isooetane exeiplex system (fire! iryecnon start. MAIDCandfirel uy’ecrian duration :.1 5m). An example of cyclic variations in Fig. 79(a). In-cyliuder liquid phase fuel distributions using a 5%DMA05%1,4,6-T1\4N090°/oisooctane exeiplex system Gite! iry'ection start : 60 ATDC and fuel iry'ecn’an duration : I .5 ins). - An example of cyclic variation of liquid phase fuel distribution in Fig. 80(a). In—cylinder liquid phase fuel distributions (contour-map) using a 5%DMA05%1,4,6-TMN090%isooetane exeiplex system (number indicates 10" g / pixel) (fire! injection start : 60 ATDCandfiiel ity’ection duration:l.5 err). - - In-cylinder vapor phase fuel distributions using a 5%DMA05%1,4,6—TMN090%isooctane exeiplex system (fire! iry'ection start. 60 A IDC and fire! iry’ecrian duration :.1 5 ins). An example of cyclic variation of vapor phase fuel distribution in Fig. 82(a). - ln—cylinder vapor phase fuel distributions (contour-map) using a 5%DMA05%1,4,6-Thfl~1090%isooctane exeiplex system (number is the equivalence ratio (6) 05¢! iry’eetion start : 60.41DC artifice! iry’ection duration : [.5 ins). Connectiousforlaserindueedexeiplex fluorescence experiment” - Timing diagram for ICCD detector system configuration.'27 .......... - 177 --177 180 181 183 -l84 - 186 188 - 189 -190 --201 . 202 A plot of gain versus potentiometer setting for the ICCD detector. . 204 xvii Figure 87. A plot of pulse width versus potentiometer setting for the FG—lOO pulse generator. _ 206 Figure 88. A plot of gate pulse delay time versus potentiometer setting. .......... 208 xviii Chapter I. Introduction 1.1.BackgroaldandMotlvatioa Overthelastdccadeorso,thescopeofcondensedmanerphysicshasexpanded inmantmberofmn-uadidmalmAsmeelecuonicandstrucnnlproperfiesof sofidslnvebecomeweflmdunoodunereahasdevelopeiforexamplemmebehavim of“sofi”condensedmatter,aswellascomplexanddissipativedynamical systems. Fmthummephenommamceconsidaedthedomahofchemimmataidscienfists, mdengineushavebeenprobedudifl‘aentlevelsofmdamdingbyphysicists. For example, the discovery and development of high-'1‘c superconductivity, carbon buckyballs mdmbeemdevencrysmlgrowmhaveengagedmmaeasinglymmfidiscipflnarymix ofhvesfigatasCmnremwiththemausingbreadthofcmdensedmanareseuchhas been an increasing involvement with engineering applications. Intheprecentreeearchmethodsofexperhnentalopticalphysieshavebeen directed toward addressing a highly complex problem: the dynamical behavior of a polydthWO-phansymwunafiqtudkmjeaedmdahighpresanemroughm orificeitbreaksupintoafieldofsmallliquiddroplets.ThefieldisnotstaticMas evaporationocctns,theliquidandvaporrapidlyevolveintimeandspace.Thegoalof meprmtmchiswdevelopmemodsfmfimfifingmdmtupredngusinghsa- Mmdflummfiedynmiaoffieflfidinjecfimmdvapodnfimprmm methodology draws upon the fields of optical physics. excited-state chemistry, and mechmicalmgineuingflhaeismimpmtantmfmstudymgmishighlycomplex system. It is well known that the distribution of fuel vapor and liquid within the combustion chamber ofan engine significantly afl'ects its performance and the formation ofpollutmts.‘ A lean-burn Spark-Ignition (SDengine is an attractive concept in engine dedgnbccamehfidfiflsbothenvhmmenmlmdlegislanveemissionnmurementsby improvingengineefficiencymdfinlecommyfThaefaeitisvayimpummmconml themixhuefmmafimfmthelem-hnSIenginebecunetheair-finlmixingprocessm thecomhnfimchambahasadommntefl‘eamthembsequemwocessesofignidm. flameWmdpoumantformafionmdaleanflowfitelmairmfio)opemfing condifimsIheaimofthemimneinjecmdSImgineismprovidebenaconuolovathe mixingpmcessmdmimprovesuafificafionoftheinjectedmminthecombusdon chamber.UndaMdhgthemixhgprocessmmecombusnonchmbais,thaefore.one ofthekeyfacmrsmthedevelopmmtofMafifiedchngemgines.Hencesomemeansof quantitatively analyzing the air~fuel mixture formation is necessary. Asthemdershndhgmdgovaningoftheseprocessesisacenualnecessityinthe fieldofcomhnfimsymmaeisagrowingdemmdforpracncdmokmdtechniques macnnflyseethefilhngofthecombusfionchambermdthemixingoffilelandair. Thaefmewimafinfimmdimagingtechniquesmsideacombusfimchambahavebeen Wuadiagnosficmolbccunedlewchniqmshouldbevayeffecnvemmeding thedimhnimoffinl-airmixnnesmdmmbusdmchmtaisficsmaimofmissmdy istodevelopnew,viablediagmsficchemicalsystemsforvapor-hquidphase vinalinfimbuedmanexcipkx(excitedshtecomplex)fwmedb¢weendim&yl-or diethyl—substituted aniline (DMA or DEA) and trimethyl-wbstituted oaphthalenes (Minamhsuchasisooctane.”TheLaset'lnducedExciplex Fluorescence(LIEF) visualization system allows two-dimensional fluorescence images of the liquid and vapor-phase fiielwithtwodifi‘erent color fluorescence emissions fromthevapor and fiqtndThecalihafimoftheflmrmencehnensitythenpamiGthedetammafionof wncenuafimmapsoffiqtndmdvapmphasefinl.Applyingmisvinnlizafimapproach minta-nalcombusfionenginesfortmderstandingofliquidandvapor-phase fuel dim’huimshasyieldeddemfledmformflimmsprayMevapomfiommdthe evolution of the resulting vapor concentration field. 1.2. Application of Exelplex System Theexciplexvisualintionmethodistheimagingtechniquebasedonlaser induced fluorescence (LIP). Currently, most sumessful applications of exeiplex visualization system as an in-situ, noble diagnostic optical technique are typically based on its 2-color fluorescence emissions from the monomer (vapor phase) and exeiplex (Hquidphase)wi¢propalydifl‘aentemissimbmds.Flumucencememodsofi‘ame possibility of minimally perturbing, real-time, twn-dimensional measurements of liquid mdvapmflowmofimTheexciplexvinnlinfiontechniqueisexpectcdtofind appficafioninobsavingmmformafimindieselorsparkignifionengineswith specuaflywefl-npmandflumescenceimagesobminedfiommemmomamdexciplex constituentsdiasolvedinthefiiel.Additionally,therearethreeareasofrelevancetospray combustioninwhichenciplexfluorescencehavebeenusedasapotentialthermometers tomeasuretemperature profiles in isolated droplets,"’ sprays,"mdimpinging sprays." musdinameanuemenmofthetempauueofthefiquideithammisolateddropla orhfidlspraysprovidesimpormntmfmmafionabommethumdumsponprocesses within sprays. Such understanding can lead to more effective design of spray injectors and combustion chambers. Although there is a shortcoming in exeiplex visualization systems due to fluorescence quenching in an oxygen environment"" for properly conditionedaunosphereseg. nitrogen,""inettgases,oralltanes(whichuevery inefl'echw' Climbers). its Mafia“ m mm ' 1.3.0verviewofPresentResenrch 1histhesisisomlinedasfoflowsAfiercmcludingdiisinuommychapter,me backgromdandtheoryrelevanttothisshulyisreviewedinChapterll.Thedevelopment ofanewexciplexvisudizafimsystemwhichisappmpfiamfmimagingmegasolme basedfinldisuihmmmacombusfimchambaismepfimarycmmhmmofmisstudy. Therefore, its photo-physical characteristics, such as steady-state and time-resolved fluorescencespecuomopy,aredescnbedmChapterm.ChapterNcmsistsofthe cah‘brafimmethodsofthemmomermdexdplexflumescencemnsitywime theooncenfiafimoffirelvaporandfiqtudAsecfionisdevotedtodescnbmgthetwo difi‘erentproccduresbetweendirect-andindirect-pressm'e measurementMaabsorption) mmevapmphasecah'brafimChaptanescnbesvapmplnsecah’brafiompmficmady inammmedenginecondifiomandmapplyingtheLIEFtechniqwmdirectmjecfion gasoline engines to observe the fuel distributions. Conclusions with recommendations are followed in Chapter VI. Appendices are included to provide detailed information of instrument calibration. Chapter 11. Background and Theory 2.1.llevlewofflowvlanallzationlnethads 2.1.1. Fnelspray vianalizaflonnaingflnorescenceandotheroptlcalmethods Recenfly,sevaalstudiesdescnbedmethodsofmalyzmgthetwo-dimensimal fieldim'hlfimmmtandcombusfimmgines.”"Thaehasbemcmsiduableprogrem mthemodelingoffilelspraysforuuamlmmhmimenginesnmdmtheappficafionof expaimenmltechniquesmspmymeasmementsinsidetheenginecyfinderP'“ qufielenfiemderstandhgofthewapaafionandmixhgpmcessesmenginesis still incomplete. Most measurements have concentrated on the suitcase and development of the liquid spray. Photography has been the principal technique used?” including such modern developments as endoscopy. 3‘ In addition. Laser Doppler Velocimetry (LDVY’J’ mdholography“havebeenused1hemajorhmimfionofmostphotographicmemods usedmenginemtdiesismnmeygivespaceimmmneadofspafiaflyresolved information. An alternative approach which is capable of yielding 3-D information is to mmmmmmafimmMasaieson-Dsecfiom.Two-dimensimalpiames havebeenreponedofpremixedflmesmengines37'”spraysmengineg“mconfined sprays“andgasjets.“Asecondflmitafionofthemeastuementscuriedomonspraysis thatthemhniquesmeithanmsensifivemfinlinthevapmphmumtheuseof phmogaphydhadowgraphyetcsmreeqmflysensifivemmefiqtudmdvapmphases. asinthecaseoffluorescenceJhmthesepuatemenauememofthehquidmdvapm phasesofafuelsprayusingopticaltechniquesisverydifficultbecauseofthelackof spectroscopic characteristics which differ significantly for fuel molecules in each phase. Johnstm”invesfigatedsponnnemukannnscanermghncmcludedthatinmeC-H ngimneuBOOOcm"thaewaeWcimdifi‘aencesbetweenthespecua ofthevapormdflqtudmaflowflreudiscrhnmafiomCWiflrthemetechmmeidi thegasemnpropmeinjecfimhewasabletomakegasplmeair/filelratio measurements.”“) Usingacombinationofvisiblelightscatteringandinfi'aoredlight scaueringmdabsorpfiomChraplyvy”deishkofi‘etal.“wereabkmsepmatethe signalfi'omeachphaseforantmconfinedspray. More recently, alaser-induced fluorescence (LIP) techniquehasbeenusedto visualize in-cylinder air/fuel mixture distribution.””' With this technique, spectrally sepmatedflmrescenceimagesothuidmdvaporphasefiteldismbufionscanbe obtained. The technique is, therefore, efi'ective md helpful in analyzing the mixing V process in the combustion chamber of engines. Melton has shown that spectrally sepmatedflmrescenceemissimsfiomhqtudmdvaporfiwlcanbeobninedbyadding anexdplexformingdopmtmthefinl’mnhevaporismappedbyflmrescencefioma monomawherusthehqtudismggedbymend-dlifiedflumescenceoftheexciplex. Thefuel,whichisahydrocarbonsolvenghasnosignificamfluorescememissionofits own, and thus its evaporation is tracked by the exeiplex-forming dopants. Quantitative fuel vapor concentration or even semiquantitative estimaion requires that the fluorescent makerwapmmesatnearlyssmenteasthebtdkfiiel.Theimfiflimplemenmfimofthe technique by Melton to visualize a spray fiom a hollow cone injector,“ and subsequently byBardsleyetal."°"inaninternalcombustionenmereliedonnaphflialenemP)and N,N,N'N'-tetramethylop-phenylene~diamine (TMPD) as the exeiplex forming constituents in decane-based fuel. More efforts have focused on developing quantitative calibrations for the exeiplex method relating fluorescence intensity to mass concentration offinl.Smhaqtnnfimfivemtupraafimoftheflumescemimagesisposablebecmue flmrescencemtensityisdirecflypropmfimflmthemammmenuuimoftheflmrescing compmmdlnifialqtmfimfiveappficnfimofmexciplexsystemwasdonebyFehooet al..“ Melton ‘5 developed calibration procedures based on the calibration of light abmtpfiomthmyieldsmdodrerphobphyaicalpmameters.flowever,those proccdtuesmqtfiredthateachpuameterbecorncflydaammedmdthaeforeme ovaaflacuuacydependedmmewcmacyofasaiesofsepuammmet al.“developeddireacah‘brafimproceduesfmthehqtudmdvaporphase.1bese procdnesmvolvedtheduectmmememofthefluorescencemtensityofaknown ammmtoffiqtddmvapor.However,mtheh'vaporplnsecah'brafiomtheymeaaued absorbmceofTMPDasafimcfionoftemperanuemddienusedthemolarabsorpfion coeficient(s)ofMDfiomBerlman“tocalmlatethevaporphuemolar concentration. Their vqror phase concentration of TMPD exhibited a significant disaepmcy,byafactor4,whencompcedwithdreduect-presuuemeaauement conductedbyFelton.“ WhiletheMD/naphthalenesystemisappropfiatefordieselandgasnubine fuels,itisnotappropriateforthe20t0215°Cboilingrangeofgasolinefuels,inasmuch asthenormalboflingpointsfornaphthaleneandTMPDareinthemngeonOOto300° C. Specifically, a low-volatility marker in a highovolatility solvent will cause a significanmdaesfimafimofthemififlfinlevaporafimmakeypufineterinengine ignifionmdperformmce.Fmthisreason.aurentefi'ormhasbeenfocusedonthe development and calibration of exeiplex systems that are appropriate to automotive gasoline fuels. Melton" has developed an exeiplex system using fluorobenzene (F 82) anduiethyhmineUEA)misooctmefinl.Thisexciplexsystemisexpcctedmbe vhmflycoevapmaivewithsolvem(finl)whichisboifingmthemmpaanuemnge70° CtollO°C. However,itmaynotbeusefi|lforprncn7caluseasanexciplexsystem becmueflieflmrescenceemissimfiomthemommamdexdplenovalap.Thuaitis difiaflttospccuaflyresolvethevapaphaseflumescencefiomfltnofmefiqmenme otherhMShimiatetd.“devdopedmexdplensystemus’mganathalene(NP)md N,N-dimethylani1ine (DMA) in gasoline fuel. Indeveloping theircalibration, they tool: a miqmapproachmneadofpruenfingdtenvapmphaseuh'brafimasaspecialmapof concenuafimatheypresenteditmthefmmofaspecialmapofequivalencemfios. However,theydidnmcahbratethefiqtudphasebtuuseditsfluorescencemshow whethaornotliquidfirelwaspresent. Themamdrawbackoftheexciplextechniquesisthefluorescenceqmnchingby oxygen."" Therefore, inmostengineapplicationsanitrogen environment is required. NwmlessapplyMgthistechniqmmmeimmcomhufimengineisquitehelpfiflm WWandvapmplmefinldism’MfimsDemfledinformafionmspmy mammdmeevohrfionofthereaufingvaporconcenUafimfieldwomd helpbothmenhmcefimdamennlmderstandingmdmimproveengineperformmce. 2.1.2.Flowvelocltyfieldvisnahntion(l"l'V,PlVandMl'V) Opticalmethodsarecennaltoquanfitafivelymeann'ingthemotionoffluids, especiauythosethatarenubulenLTheabilitytomeastuetwo-dimensional velocities over an extended planar region has become available with the advent of particle tracking velocimetry (PTV) " and the particle image velocimetry 0M.” Both PTV and PW mqrfinthflphdogaphsbemadeofpmficlesecdedflowifluminmedbyasheetofhght Thephothsmusbeinterrogatedpostenmiymdumaficaflyifthemethodism beprncficaLForthecaseofFTV,the1ightsmuceismtflfiplyptdscd,readfingin encoded particle tracks (i.e., multiple images). Because individual particle tracks must be idenfifiedmdmehdisplacemenmmeasuedmepmficleseedingdensitymustbelow enoughmidenfifyindividmlmksmambigmusly.Mprufice,thisreadtsinme velocitydahdensitybehgbothlowmdmndommeNomytwoshmtfightpulses musedrenflfingmamcordedimagepairforeachpmficleJncmuastmthe seedingdensityofPIVcanbeincreasedsufficientlythatthevelocitycanbemeasm'edon acloselyspacedregulargrid. hdrePNappfiafiomaflrfidisseededwithmflhmasomefimesbflfionaof particles depending on the vohtme of interest (10,000 particles per cc of fluid).’l A sheet oflaserlightilluminatesasectionoftheflowandthereflectionoflightfi'omthe particles idenfifiesflteirposifimAmbsequentlasersheetof lightrecordstheparticles’ positions at a later interval. By comparing the photograph, the particles’ positions are correlated with highly sophisticated computer algorithms thereby reflecting the velocity ofthefluidfordefinedgroupsofparticles.ThePthechniqueispowerfulbecauseit instataneomlymessuresthevelocityofafluidatmanypoints-akeymeasurementto myfltfidphysidstmengineer.Nevathelesnthemchnithammydrawbacksaflof whicharisefiomdleneedtomeastuetheflowvelocitywithparficles.First,particlesare illuminated by a plane of light and mbsequrmt illumination relies on the particles staying withinthelayersothattheymaybeilluminatedatsomelatertime.Second,theparticles 10 havetheirowninertiaandtheythereforemaynottracktheflow,especiallywhenit changesmddenly.1hird,pmficlesmaynotgommueasofintereu.Forinsnnce, particlesdonotgointoareasofhighnubulenceorinmeasnenrslufaws. ThereccntdevelopmauofMTVMolectflarTaggingVelocimetry)byagroupat Michigan State University’2 has elimimted the problems of PN when the particle mukasmreplacedfithhuninescentmpramoleafles.thV,naflowisseeded withsupramolecules.Agridof1aserlinesisimposedupontheflowtoproduceaglowing supramolecular grid; the intersecting lines of the grid define points that follow the flow forabfieffime(fiommiaosecondsmmilfisecondsdependingonflowspeed).The glowinggridbythesupramoleculesisimagedwithCCDcamerasasitconvectswiththe flow.1tisthereforeeasymrecordhowthegriddefmmsinfimem¢bymeasningdie disMncemddirecfimeachgridintasecfionuavelsmdknowmgmefimedehybetween eachimagethetwovelochycomponenminthegridplmecanbecalculatedmdthe cmrespmdhghubrflencethifientheReymldstdmevmficitycalanatedA secondcamaaaflowsonemrecordthedurdvelocitycomponenLAswithPN,MTVis an instantaneous velocity measurement, but it is non-intrusive. Since the supramolecules mpmtofdreflow,afltheproblenuassociatedwi¢pmficlesmdtePNtechniqmue eliminatedbytheMTVtechniquelheuickinimplemenfingMTVcomesinthe supramolecular design - the grid must exist long enough to follow the flow. Since fluid flowsueslow,hmhacenttacaswithhighemissimqrmhmyieldsmdalmgexcited satehfefimes(>milhsemnds)thumnaquenehedbywuawoxygenuenecessary. 'I'heMl'Vtcchniquehasallowedonetoprobemanyproblemsthatwere previously elusive to the fluid physicist or mechanical engineer. These include the 11 independmt velocity fields of liquid, and solid. in two phase liquid/solid flows.” the leading edge problem at airfoils,’”‘ and a quantitative measure of complex flow of swirl mdmbleaotafionalflowabotnthecyfindaaxismdperpendiquarmthecyfinder axis. respectively) withinthecylinderofanenginen'"andotherflowissues in internal combustion engines.3 2.2.le 2.2.1.8pectrsscopy Specnoscopyisthegenaaltermforaflthetechniquesmatmvolveflleexchmge ofenergybetwearelectomagneficradiafionandmatter.1henvomostfamfliar wbcuegofiesofspecumcopyueabsmpfimspecnoscopy,mwhichenagyismsfened fiomtheradiafimbthesamplemdemissimspectoscopyjnwhichfiempleloses energy,andthisenergyappearsasradiation.IngeneraLexcitedmoleculesmaylosetheir excimfionenergyverympidlybylmdergomgaradiafionlessuansifiontoalower decuonicstateofmesamemulfipficity.”1hisprocessismfaredwasanintemal conversion (1C). Figure 1 shows that internal conversion is followed by vibrational relaxation (VR)tothegrolmdvibrational state.Theexcitedsinglet molecule (8.) may alsoreuunmthegrmmdstate(so)withtheemissimoffluorescence(F)mitmay mdagomter-systemaossmgaSC)mtheuiplasmteT..1nmr-systemanssmgmfas to aradiafimleutmsifimbdweenthefleofdeulfipficity.Becslueofmespm mtachangemvolvedinintasystemmmemteconsnmfordlisprocessis 10’2to v’=2 [C 1 tsetse.) s. a l in : w-o F v T. YR - E ——-b ------ > s, l Figure 1. State diagram for a molecule showing the various intermolecular processes resulting from the absorption of radiation.” 13 10“ (sec")asfastasforinternalconversion.Thetriplet moleculescanphomhorescea’), mlmdergommter-systemcrossingmthesosute.1heradiafimemiuedmamsifion between the states of the same multiplicity (i.e., singlet-singlet or triplet-triplet Umsifims)ucafledflmrescence,mdthemdiafioneminedinamsifimbaweensma ofdifi'aanmulfipfidtyiscafledphosphaescalce.”Phosphmescencefifefimesm generally considerably longer than the fluorescence lifetimes because radiative transitions between sues ofdifl'erent multiplicities are qumtum mechanically forbidden to a first approximafim.”Figlne2isasimpflfiedmpresenmfimofmeimamoleuumprocesses that may seem. 21.2.Qnenchthechanism A. Steady-state (intensity) qnenchlng Thehrtensityoffluorophorefluorescencecanbequenchedbygromd-state quencher-fluorophore reactions (static quenching). and by excited state quencher- fluorophore reactions (dynamic quenching). Static quenching is asmmed to remlt from the formation ofa non-fluorescent quencher-fluorophore complex in the ground state. Shifisoftheflumophmeabsmpfimmwithaddedqlmchuprovideevidenceof mhacomplexformafimAnodlatypeofstaficqunchingisofienobsavedathigh quencherconcenuafimsduemdleerdstenceofmnumbasofquencher- fluorophorepahsinwhichthequencherisclosemoughtothefluorophoreto inmnnemnlyqlmhimexciwdsute.”TreatmentofmistypeofqumcMgisless straightforward. However, itcanbedistinguishedfi'omquenchingduetouuegromd- 14 sl 9 I _. E x \ Intersystem Crossing interml Conversion \ ‘ k'sc’ 10‘ - to” SJ film-.1038" X .5. : . T' 33 : Fluorescence I a : ks. 10‘ ’ 109 S" Phosphrescenee ' Intersystem crossing 1:9, 102. 104 84 k’m 10‘ — 10l2 S’I FigureZ. Abbreviated version ”ofthepreceding Fig.1. 15 statecomplexformafionbecmseitdoesnotprorhnechmgesinthefluorophore absmpfimspeMmTheaendy-stuepamemwhichrespmdsmtheaddedqmnchais thefluorescenceqmnunnyieldfiawhichisdefinedastherafiooftherateof fluorescenceemissiontotherateofabsorpfionmhotonsouflphotonsin).1he fluorescencequannnnyieldindleabsenceofaqlnncherfirfiisdefinedbythe mechmianasshownbelow.AandA‘metheflmrophoregrormd—statemdemitfing~ stateaowestexcitedsinglet), respectively. kAandkparetherateconstants forphoton absmpnmmdflmrescenceemissiomrespecfively,mdk.ismemoffirst-mderme constanmfmnomradiafivedecaymodes,anhasmtanalcmversimmdmtasystem crossing. nAandnA.arethentnnberoffluorophoresinthegroundandexcitedstate, respectively.Theratemechanismofthisprocessisgivenby A + Irv aA‘. Rate = k4n‘ A‘-)A +hv. Rate=yr,,. A" -»A Rate =k,n,. Since steady-state conditions exist, we can assume: ktna=09+Una~ (hr/db =0 (1) Rearranging Eq. (1) gives: n,.=k,n,/(k,+k_) (2) Theoris formallydefinedas: d, = (fluomcence rate) / (absorption rate) = k/l‘e / k‘n‘ (3) 16 Combining Eqs. (2) and(3) gives the familiar form forthe fluorescence quantum yield in theabsenceofquencher(¢,°): #=*r/(*I+U . ('0 Whenqucncher[Q]isadded,mothaA‘decnymechmisnismwposablewherek‘is hacond—adamtecoanaquenchhgmdmeWQjSspswdofirst-mda. A‘+Q-9A+Q, Ratc=kJQInp (5) Noqu.(2)becomes: "4': And/@+hv+kJQ’) (6) CombmingEqs.(3)md(6)givesmexpressionfortheflmrsscenceqmnnmyieldinthe presenceofamwncha,providingdynmicquenchhg(Eq.5)istheodqunchhg mechanism. it“s/0941.44.12!) (7) Dividing Eq. (4) by Eq. (7) produses the familiar Stern-Volmer equatiom” pf/d,=l+k,[Q]/(k,+k_) Inthemorefamiliarformofthisequationfi'andeparereplacedbypandf',the miensifiesoffluorescencenagivenwavelengm,mtheabsencemdpresenceon, respectively,andtheterm[k‘l(k,+k_)]issetequaltoKmtheStern-Volmerquenching constmt.” FIF=1+K,,{Q] (8) 17 [fthisquenchmgmechanianaloneobminsaplmofP/Fveram[Q]islmearwithm mterceptoflandaslopeequalmKspWhenqlnnchmgcanalsoocmubygromd-nate complexformation(static quenching),anotherrenction must be’consideredintheoverall quenchingmeclutnistnt'l A+Q22AQ -) AQ‘-)AQ+heat Whetelcistleanenas. K = [442]... / ([AI... [2]...) (9) Kistheequilibrilnn constantforthe formation ofthe "dark” complex, AQ.1fboth static mddynmicquenchingmocaurthhemtensitymfio(F/F°)cmbeexpressedasme fiactionalreductionduetoquenchingofA‘ (dynamic)timesthefiactional reductiondue to complexation ofA (5 static)!” F/F‘=(1+K,,[Q])"f The reaction fean beexpressed in the following way, given the definition ofKin Eq. (9): f = [A] /([A] + [AQD = (1 + KlQD" Thuaitfollowsthat 1 F/P’=(1+st[Q])"(l+K[Q])" ‘I‘his,intlnn,canberearrangedtogiveamodifiedformofthe Stern-Volmerequation: P/F=(1+st[Q])(l+KlQ]) ‘ <10) 18 Eq.(10)predictsanupwardcurvatln'eoftheplotofl‘°/Fversus [Q] intheeventthat bothstaticanddymmicquenchingareoccurring. [fquenchingonlyoccursbythestatic mechanism(k‘,l(,v=0), Bq.(10) simplifies to: P/Fa-li-IqQ] (ll) Eq.(ll)alsopredictsa1inearrelationbetweenl“/Fand[Q]withaninterceptofl,as dossEq.(8)daivedforfllecasewheredynamicquenchingaloneisocanring. lnthe formacaseflleslopeeqmlsIthfleinthelanereqmlsthThus,onecannot detummewhahaquenchingisaaficordynamiconthebasisofasinglefinemSMm- VolmerploLIfsuchplonm'eobminedasafimcfionoftemperanueadetuminafion mightbemadefiomthechmgemslope.maeasedtemperauueofiencausestheslope (K5,)mmauseifqmnchingisdynmicmmaeaseswimwywhflemeslope (K)shoulddecreasewithincreasingtempaatmeiftheqmnchingisstafic.” B.Tln|e-resolved(dynamlc)qnenchlng Whenflmrescenceisexcitedbyaptdsedmtherthmaconfinuoussomcethe decreaseofflmrophorefluorescenceaftertheplflsenormaflyfouowsasingle exponmfifldeayhmhfimmfifefimeofflumescmdefinedasthefimeforme fluorescencesignaltodecayto1/eofitsoriginalvalue,isgivenbyEq.(12)inthe absemeofaquncheranqu.(lB)inthepresenceofaquencher. t" =(’9+kJ" (12) t=(k;+k.+k.[QI)°’ (I3) l9 Dividing Eq. (12) by Eq. (13) gives another form of the SternoVolmer equation, which applies only if quenchmg' is dynamic. t"/r=l+kJQ]/(k,+k_) suit/[Q] (14) The Stern-Volmerconstant (K37) obtainedfi'om steadyestate measurements is equal to k,r”.Thusiffifefimemeanuemennueposslblethevflueofk,canbeexuaaedfiom plotsoqu.(l4)orEq.(8)andthevalueofthelifetimeintheabsenceofquencher(r°). Itisimpmnmmnotethatfifefimemeamementsuenotafi'eaedbydleformafionofa gromd-smte'dnk'complex(AQ).Cmsequenfly,fifefimemeasuemenmcmbeusedm separatedynamicquenchingfi'omstatic. 2.2.3.1-1xcitationandelni-iondynamles Foraqumfiufivecah’branonofin—cyhnderfitelmixnueformafiomfllemeastued fluorescencemmnsitymuabecmlaedmtheconcenuafimofmeseededmolecules. Themlafionshipbdweenflmrescencehensitymdmecmcennafimofmeflmrescence mediumcanbedeterminedbymemsoftheumbert-Beerhw.Theuansmiuedlaser mtensityLwhichrenumfiompropagafimmamifmmflmrescentmediumofthickness b (cm) is defined by the following equation; A lo/L=¢XP(8¢b) whaeloisthemcidemhsamtanity,eistheabsmpfimcoeffident(cm"mr’)ofthe mediummdcisfiecmcenflafioflmoboftheflumescentmediumThenlaxafionphase oftheexcitedfluaescentmediumischmactaizedbyradiafiveandnon-mdiafive processes. Representing the rate constants (3") of the radiative process and non-radiative 20 processbyhmdhrespecfivdyfihefimedependannumbadensityofmemdemflesm meexcitedstatenn,.(t)isgivcnhye no (t) = I!» M-[ktfielu whit-em. isthemunberoftheexcitedmoleculesformedinthemediumperexcitation pdsethefiuorcmenceinmnsityatnmetisgivenhy, It(t) =ktna~¢xp(-lkt+k.]t) (15) Theflmcencemtensityperplflsemdpamitvohuneisthmobminedbyintegrafing Eq. (15): I = fund: == (haunts) lob" (l-expl-ecbl) (16) rhefirstparcnmcnceltmmmsq.(ln)nmeaummcmcequannnnyieldmmeahsence ofquencher(¢,°) definedintheprevioussectionastheratio offluorescence emissionto themeotaheotpnontpmom/photonsin).rhtasq.(lo)cmhetewnncn; I=¢t° lob" (l-eXPl-ecbl) (17) Thefinotetcmcesignalpcthscrpusemdpertmnvohnnecouectedhyeachpixelofthe detectorarraycanbeexpressedas St = [nfl‘l/4nlltt°lllob" (l-eXPl-ecbm (18) wheren ismecollectionemciencyortheoptict. a isthedetectorefficiency, a isthe solid angle oflight collection. 'I'hefirstparentheticaltermdescribestheoverall efficiency ofmeopncemddetectot.secondandthndtttmsdetcnhemeqmnnnnyieldmdme total number of species molecules per unit volume. 21 2.3.userinducedflnorescenumll') Fluorescence is the emission of light (photons) from an electronically excited weofmatommmoleaflehLIFJheurgetspeciesabsorbsenergyfiomahsermatis nmedsothmmephomwsenagy,hv,isexncdyequdmtheenugydifi‘aence,AE,in elecfificmlhepmenugyisthenmresmmcewithAE.Theloweuaateisthe gromdelecuonicnate.0nceexcited,thespecieshasseverdwaysmrenunmthe ground state (dc-excited). Firstly, if the molecules are already in a resonantly excited state, the laser energy can stimulate de-excitation. Secondly, it may spontaneously release energy (fluomcence) back to the grand electronic state. Thirdly, the excited molecule maytransferenergyinanon-radiative fashionthroughcollisionswithotheratomsor molecules. These transfers are reflected through changes in the vibrational or rotational energy levels (or a combination thereof) represented by W!) and 11(1), respectively, whaeJisthemtalangularmomennmquannnnnlmber.Addifionany,energymaybe absorbed by the molecule, resulting in elevation to a higher electronic dissociative state. A thorough analysis of molecular spectroscopy is provided by Banwell.‘2 ThemathematicalandphysicaldescriptionoleFcanbehcn'bedbyatwolevel modeLasshowninFingherateofstimrflatedabsorpfionisdenotedasWu. SfimulatedmdsponmnemnemissimfiomtheuppakvelisrepresenwdbywnmdAzp Thecomsimalcplenchingtamsmyolmedmgethamddenotedasqz,.Aflmtesmm s". The population [molecules] in the lower level, before excitation, is N? . Alter 22 wuNl “’21": Ale2 Qlez stimulated stimulated spontaneous collisional absorption emission emission quenching Figure 3. Two-level model for laser-induced fluorescence. Wu and W1. represent the rates (3") ofstimulateddisorptionandemission. A2. istherateofspontmeousemission, ansz. istherate of non-radiative quenching. 23 resonantexcitation, N, + N,=N{'. Therateequations forthe population ineach level are givenby 3 W12N1-('721+A21+Q21)N2 52!; Jr W __.l. .. dt ’ leNl +(W21+"21+Qzl)N2 Atthepukofthelasaptdsethepoplflafiommeatsteadysutemdthepoplflafionof theupperlevelisgivenby W N = B N 2 ‘721 + All +Q21 l Under conditions of a weakly perturbing laser (low irradiance), the lower level poprflafioncanbeasmunedtobeappmximatelycmstant;mdbothW,zandW2,are small.Thentheupperstatepopulationcanbeexpressedas W = l; 2 "n +ta N: (19) Theflmrucencesigmhseenbyadetecmrjdphomnyispropmfimdmthefiacfimof theupperstatepopulationthatspontaneouslyde-excites sf¢A21°N2, or, using Eq. (19) and We A s new Nf.[———ll— I '2 421 +921 24 where the parenthetical term A,/(A,,+Q,,) is the fluorescence yield, 1“,, which is the fiacfionofthetomlexcitedstntepopuhfionthatde-excitesviafluorescence. Themteofsfimulatedabsorpfioncnlbewrittentoreflectthecouplingofthe laserwiththeabsorbingmoleculss ”.2"? = ”chit". where 3,, is the Einstein 8 coeficient for absorption (cm’ocm"s.l"), 1,, is the laser spectral irradiance (chm’ocm"), n, is the number density of the absorbing species (cm' ’),andVisthevoltnne(cm’)ofgasilllnninatedbythelaser.TheEinsteinBcoeficimt descnbesthesuengthofcoupfingbawuntheuppermdlowalevehsothmmetem 1,.8,z expresses the probability for absorption ofa photon by the coupled molecules dluingthetimedurationofthelaserpulse. Thefluorescencesignalperlaserpulsecoflectedbyeachpixelofthedeteaor arrayfi'omavolumeoffluid,V,intersectedbythelasercsnbeexpressedas” Sf =[flb‘z‘3 '(anrVC)'Fy '(lvBu wherenisthecollecfioneficiencyoftheoptics,sisthedetectoreficiency,0isthe solidangleoflightcollection, fgistheBoltanmnpoplflafionfiacfiomandeisthetotal number of excited species. The first parenthetical term describes the overall efficiency of meopncoandmedetectmmhesecondpmmmeficnlteemdetcnhetmemlnumheroi species molecules within the vohlme Vg. Theabsorption linewidthistypicslly describedbyaVoigt profile. whichconsists oihothLotentzianandnopplet(oeussian) profiles.The Lorentzianprofileistheresult of the radiative decay rate of the excited state, which consists of the spontaneous decay 25 rateplustherateduetomolecularcollisions(whichareafilnctionofgasdensity). The Luauzimpmfimofmefineshapemaeforeisdepademmthemmm andconisionpcmeterhespccnucmtplmgotthelumlmeehapewimmethsotpnon fineshapeismflectedbytheovalapmmgraLngwandI,=Ig(ld. wherelisthelaser WW/M%W0fWY-WOWWEWW g(v) = I¢,(v)¢,,(v.r.p)av ,, _ whereo.and¢p,.arethelineshapefimcfionsofthelaseranddleabsorption, respectively. Thefluorescencesignalisthen Q S] = asGJ'NTVc ~15; ~13!2 I¢,(v)t¢*(v,T,P)dv. v Atafinosphaicpresane,&ehsalhewidfiistypicaflymmhgreatammmeabsmpfion finewimmfingmmovalapmtegrflmatapproximatesflleabsmpfionhnewidth Thusflahnoqlhaicpreaanedleenfinabsorpfionspearalbmdcouplesmsome pmfionofthehserspecualbmmthereisfiuledependenceofthesuengmofthe fluncencesignflmfiemrmmdhgenvhmmentHowevermspresnuehcmsethe absmpfimlmewidmmmnmmemuessedmnnberofcolfisimhmiscasethe absorpfimfinewidthcsnexceedtheluafinewidth.1hecoupfingofthehsaspecufl distributionwiththeabsorptionspectralbandisineficient. reatltingintheexcitationof fewermoleafleflthaefmeadeaeuedflumescencesignflcompcedmthesignalat aunosphaicpresnue),andthecmmlingbecomesafimctionofthecombusfion environmentintemperannepresstlreandcollisionparmers. 26 Laserdiagnosficsimprovestmderstmdingofcombusfionphenomenabecause meyprovideremote,spsce,fime,mdspedesorssolvedmuauemeanfcombusdon prametas“LasaMIcedflmrescencetechuquesueweflnutedmthedetecfimof mncfivespccies.”*lheyuewidelyappfiedadiagnosficmokfmmvesfigafiomoffinl distributions in internal combustion engines, atmospheric chemistry, combustion and phmamwhaeadeflfledmdushndingofcahhchemicdreacfimpathwaysis desuedfickhredt”mmarinsthedevelopmemofdifi‘aemLIFmahodsmdtheir appficafionbfltemeannemeMofspeciesconcamafimsmdtempemuuemcombunion environment. Using two-dimensional LIF schemes, a wealth of lmique information has beenacquned;"itmayofienbemficiemmmeanaemeapproximatemncenuafionand relative spatial distribution of the molecule under investigation. There are situations where concentrations of reactive intermediates should be measured quantitatively with filebeuanamabkacmnacy.Thisismqldre¢forexmpleforamemingfidcompmison of experimental rhta with the predictions of combustion models, including detailed chemicalreacfionschemes,e.g.aflamemode1whichsimu1atesflleformafionof pollutants. LlFisoftentheonlydiagnostictechniquewhichcanprovidethedesired information without perturbing the combustion process. To obtain absolute concentration ofimpmmmmtumediatesfiomthemasuedflmrewencesigndsanmblemegiesfor cah'brafimmustbedevelopedmdtenedofleninrathersimplecombusfion mvhmmmbefmemeycmbeappfiedmmmuemedmofmedesuedmy. 27 2.4.1'herlnodynamies Thevaporpresuneofaflfiqrddsvarywithssnuafimtemperauueinessenfially mesamemmner.Thedependenceofmeanuafionpresuuemthetemperanaewiflmw be developed from theoretical considerations. The generalized relationship that results is almvafidforsofid-gasmdsohd-flqfidphasechmmOnebegimbycalaflafingthe entopychmgeofasmnplembshnceduhgaphasechmge.Thisentopychangein termsofthevariablesv(molarvolume(V/n),wherenisthenumberofmolesandsisthe entropy)anchlmberqnesentedas sews a V However,formyfirstorderprocessmvolvingachmgeinplnsethetempaanueis constantdluingthephasechange.Thereforedieaboveequafionreducesto “(i-J,» The quantity (oi/aormn be replaced using the Maxwell relation (at/at)? =(d’lmv. Hence eels Theterln(a‘P/mvisflleslopeofthesanuationclnveatagivensanuafionstate,and thisquantityisindependentofthevohuneduringachangeofphase. Consequently, the partialderivativemaybevwittenasatotalderivativefl/dT,anditmaybemovedoutside themtegralsigndlningthemtegrafimoftheaboveequafiomlntegrafionleadsm z "41' 2 01' s -s £=[_2__L] (20) dT vz-vl whaethealbsaipm“1”md“2”representmetwophasesmtheprocsss.Fuexample meymayrepresenttheunwed-vapmmdmed-hquidphasedwingavapofinfion process. Theentropychangedlningaphasechangemaybeevaluatedfromthefirstand secondlaws. Fromthesecondlawdr=&/T,andforaconstant-presslneprocess(such asaphasechange)thefirstlawforaclosedsystemis&1=&.Thus dr=cfllIT and s2 «sI =(h2 -h‘)/T. Eq. (20) thenbecornes dP III-h! All (2'1") .-. 1'0:2 -v1)=7'z; (21) This equation is called C'Iapeyron equation. It is generally valid for any phase change whichoccusatconstantpresnuemdtempamae.Fmafiquid-vapmplmsechmgemis equationmaybewrittenas 95-1%. dT- ”I: (22) IngeneraLAhandAvm'etheenthalpymdvohmechmgssbetweenmynvosanuafion satesnthesamepresuueandtempaahne.NmethatdnClapeymeqmfimpamitsdle evaluafionofenflnlpychmgesforphnechmgesfiomalmowledgeofonlvaTdata. For liquid-vapor and solid-vapor phase changes, Eq. (21) can be further modified by 29 introducing several approximations. One considers only the liquid-vapor phase change, but the results are equally applicable to solid-vapor plnse changes (sublimation). For liquid~vaporphase changes at relatively low pressure, the value of v, (gas phase volume) ismanytirnesthesizeofv,(liquidphasevoltnne).Thusagoodapproximationisto replace vmbyvfintheaboveequations. Also,attheselowpressures,therT relation for thevaporcloselyfollowsthatforanidealgaev.=RT/P.Bymakingthesetwo successive approximations in Eq. (22), we find that (0’ thg (”-1212 h “P M (23) P 312 Eq. (23) is fi'equently called the Gaucho-awn equation. Integration of this equafimdependsonmevaiafionofhbwimtempmuleaanaflvariafimofpressae (ortemperature)ischosensothatthechangeinhtovertheintervalofintegrationis small, then integration yields 1nP=—’1Rg-(-;-,) +c <24) whaeCisaconsuntofintegrafionihismdicatesmnmevapmpresnueofahqludis vaycbselyanexponenfiflfimcfionofthesahnafiontempaflmeThegenaalformof theequationisalsovalidforsauuationdatabelowthetriplepointinthesublimation region. 3O 2.5.Dropletvaporizntion 2.5.1.0eneraleharacterlstlcs VaporinfionofliqlfidhydroccbonfiielisaprereqlfisitemcombmfiomAnyfilel mvaporizednthecombustmexitiswasedlnaddifiomitmayhavethedeuimenml efi’ecuofaodingmaalmfachlnlbuelyvapaizeduthecombunmexnmayhave inslficiemfimemhInAgaindlaeistgeplusposmbkdeposifimorendothermic decompodfimDheafitelmjecfionmtoacombusfimchamberisencomteredinawide mge of applications such as industrial heaters, diesel and direct injection gasoline mginajetenginengasuubmeaaclhehqludfinlmjecfimmdtsmmeformafimof dropletsthroughatomiufionmechanismsthatdependonthesystemoperafing cmdifimalheindividualdropletprocessesareofprimaryimpmflnceinspray comhnfimsinceinmanymsnncesthedropletsueweflseparntedfiomeachother. Dmpletvaporinfionisthaeforemimpmuntpmametermdtemodelingoftheabove systemsmdcmbetheconuoumgfacmrforenergyconvasionrates,mixnnerafio distributions, and overall combustor behavior. Present understanding of droplet vaporizationandspruycombustionissulnmar'iaedwellinreviewarticlesby[.allv,’2 Sirignano,”andFaeth." Asketchofthedropletevaporatiooprocessisprovidedinfig. 4, forthe hypotheficalcasehwhichapmefinldropishstmflyinfiofiwedmmagasuelevmed tempaaumAttypicdinjecfiontempaauuenthefilelconcenaafimutheliquid mfaceislow,mdmmeisfinlemasdiflildmfiomthe&opeulymtheprocas.0nder 31 T T Y. o" ‘ ,’ I Y s r ' t D D U smut-rm "" . Lass-turns: .1 ""' Figme4. Sketchofthedropvaporizationprocess.” Where'l‘isthetemperatme,r,isthedroplet radius,v,istheradialvelocity,de.andearemassfiactionofambientgasandfilel, respectively. 32 thueconditions.thedropletheatsup,tnuchlikeanyothercoldbodyplacedinaheated environmenclngeneraLtemperatluesarenotlmiformwithinthedroplet. withthe muimumuqludmpmnebenedumeaufaeeumeuqludtempmmfiseame rateofmastmsfamcreasesasaremhofhighafinlvaporcomenuafimatmedrop’s wface.1hishutwoefi’ects:(l)nhaeuhgpufimoftheenergyreschingthedrop nufacemustamplytheheaofvaporinfimoftheevaporafingfiandahheomward flowoffitelvapormthebmmdxylayermmerateofheatumsfermthedroplet Thisslowstherateofmaeueofthefiquidaufacetanpaannemdhtermthepmcess temperatlaesbecomemorelmiformintheliquidphase. Eventually,astageisreached whaeafltheheatreachingtheaufaceisufifizedfmtheheatofvapofimfionmdthe dropletstabiliaesatatemperattne calledthe“wetbulbtemperature”." 2.5.2.81mplemodelofdropletvaporlzation A. Slmplemodelofdroplet The simplest analysis ofdroplet evaporation has been reviewed by Faeth" and canbetmderstoodwiththefollowingamunptions: (1)Thedropletisasslunedtobesphericallysymmeuic. (2)Thegasphaseisasnmedmbeaqlmsi-stadyconfinmwhichalwaysadjustsmthe steady-statemlcuueforthehnposedbmmdarycmdifimsatenchmsmntoffime. (3)Fluidpropafiesxeasnmedmbecmstantwithpropatyvahwsdetermmedatm appropfiatereferencecondifionl‘hefiqludfilelisasaunedtobeasingleplue component. 33 (4) Therelationshipbetweengasphasefuel concentrationandliquid temperature is given bythevaporpresslnecorrelationfortheptueliquid, e.g., theClausius-Clapeyron equationSln'facetensioncorrectionsareomitted. (5)Theambientgaseshavenegliglb1esohrbilityindleliqludphmeandooly fuel is difl‘llsingfi'omthesurface. (6) Theradialmotionoftheliquidsln-faceisasslunedtobesmall. (7)1'hepressureiseverywhereequaltotheambientpresslue. (8)Radiationisneglected. (9)Massdiflilsimisnpresentedbyanefi‘ecfivebmuydifi‘usimhwmdtheDtu‘om efl'ectisneglectedintheheatflux. (10) Reaction efi‘ects, such as fuel decomposition, are neglected. For convenience here, theLewisnumberwillbeasstnnedtobelmity. ltmustbeexpeaedthntheseasmmpfimsbecomemvafidatverylowpresauewelowl atm),forsmalldrops(ontheorderoflumorless),atpressuresapproachingthe thermodynamiccfificalpointofthefilehandinthepresenceofveryllnninous flames.“ B. Dropletvaporlnatlon a. Heat-transferantrolledvaporhatlsn During vaporization, relatively volatile solutes are probably rapidly heated to theirboilingpoints,aflerwhichvaporizationofthematerialocctns. Iftheboiling point ofthemnedalisappredablylowerthmthembienttemperanuecmdtwfionofheat fiom the ambient gases to the surface ofthe solute will likely be the rate-limiting process 34 forvapodufiomGiventhiscmdifimmevapofinfionratewfllbeinfluencedby ambiemtempunuueandthethermalconmmvifissoftheambiemgasesandsolme vapor,jtntasfordesolvafion.Thevq)odnfimmeeqmfimcanbenpresentedby” D,’-D’== ,t (25) whaeDisthem'opletdiameteratmyfimetafiaitsennymtoaten-secfiomDoisme mifidrhopletdimeta,mdk,isthevapodnfimnte.8ecamethisprocessisheat- mafacmaofledthevapofizafionmek.dependsonthefltamalcondmfivityofthe gasesnmmdingthedroplenthembienttempaanuemdsevualothapmmeters.As themmpaauuedifi‘aencebaweenmeboifingpomtofthesolmemdthembiemgases become less, mass-transfer control could become the rate-limiting mechanism.” b. Mass-transfercontrolled vaporization Massetransfer controlled vaporintion of a droplet would be limited by the diflinionofgaseolnsohneawayfiomthedroplecdlevapofinfionratewmddthen dependuponthedifl’usion coeficientofthe solute vapor, thesolute volatility, andthe ambienttemperature.Forthelargedroplets, thevaporinu'onrateequation wouldbe describedbya relation similar in form to Eq. (25).” lnthecaseofverysmallrh'oplets, mass-transfercontrolledvaporization would follow Eq. 26 and 27 below:” «int/d: =k,D’ (26) D, - D = k,t (27) wheremismemassofsolmelostmningvapofinfimthek’sareconstants,andDrefers to droplet diameter. Eq. (27) follows from Eq. (26) because the mass of the droplet is 35 proportionaltoD’. Eqs.(26)and(27)applyonlytodropletswhosediametersaresmall (lessthan0.l um)compmedtothedifi’usionhyerstnrmmdingthedroplet.Thediffmion layerisontheorderofonemeanfi'eepathlength.” 2.5.3.1(1netlcsofdropletvaporlntion A. Gasphaseeqnatlons» Thebasicconser'vationequationsforthegasaregivenillkefs.”‘” Underthe previousasannptionsforfltedropletinSectionZJtheybecome: Conservation ofmass: %(pzvr) =0 (28) Conservation of species: grim, Y, - %]]=o (29) Conservation of energy: i-[rz (pv'Cp(T- TJ-kg-H = o (30) wherev,isthemastfiacfionoramhientgasotfitel.Sinceonlythefitelhasnetmass mimn'shomreptetenthmnydimtnvityorthefielwimmmmgasphase species; similarly, the fuel specific heatshouldappear in Eq. (30). By definition xi; =1 (31) 36 The malysis only considers two species, fuel and ambient gas. Therefore, through Eq. (31) only one comervation ofspecies equation must be solved. Integration of Eq. (28) yields rzpv' =n’rf/41r=colm. whnetitfismemtalfitelvapotiunonmteormedropmhehotmdmyeondifionsue Thefiwlcomenuafimmdtempaanneatthefiqludmfacemrelatedmroughthevapm pressurechnacteristics ofthe fuel A final bormdary condition is supplied by the insolubility assummion, which implies that themassfllutofambientgasisaeroattheliquidsurface d7 r=r’; purl:l —pD’-;r‘-'-=0 Withtheproblemposedinthismmner,lrnowledgeof p, Yr...’ Toand T’provides Yr’ rilfandtheheatuansfermtetothedroplet.Aheatuansfercoeficient,which 8 includesmasstransferefi‘ects,canbedefinedasfollows “(filo-r.) Solution of the equations then yields the following 37 41:3," = ln(l + By) (33) Nu=%=2|r(l+8,)lfly (34) where ByisSpalding’smassu-msfernumber B.=IY.=(Y..-Y..yu-Y..) (35) For high transferrates where Yin-’1’ Byis aconvenient drivingpotential forthe definifionofamassumsfercoefiidemforthediflinimofonegasmroughamgnant gas. In] =4m':zK'(YF’ -Yp.)/(l-Yr,) Substitution from Eq. (33) then yields the Sherwood number 2K‘r Ska-7:2 1+By)/By (36) At lowmasstransferrates, By -)0, Eqs. (34)and(36) yield Nu=Sh=2, whichare thefamiliarvaluesforasphereatlowmasstransferrates. B. qunidphaaeeqnatlons IntestswithmspendedlargedropsD>1000um,anlunberofmvesfigatorshave obsa'vedcirclflafionctarentswithinthech'opasitevaporated'm“ Iftherateof chaflafimisrapidthedroplettempanuuewinbemhfivelymifmmatachmstantof time, in spite of the lowthermal difl‘usivity of most liquids.“n However, others have mademeamaementswhichmdicuematcirculafimisstabflizedforandldropsand 38 appreciable temperature gradients are present within the drop throughout its lifetimes.” '“Theconservationofmassofthetkopletliquidyields d 4 z , affirm/Fm, <37) Employthqs.(33)md(37)mefindsmexpresmonfortheevapmafimrateconsant as follows ,) (38) Forthecaseofampidlymixingdropnhefiqlfidtempaanneislmiformmdwehave ntc --=4nr :I-(r‘ -r)-tit III: (39) ThelasttermmEq.(39)reprssentstheenergyrequiredmevaporatethefilel.Theinitial conditions for Eqs. (37) and (39) are Eq.(39)canberearrangedsothatthewetbulbstatecanbeintupretedasfollows Jr "Uh s ._.’_.= f‘ __L-1 (40) dt MC 8 p; y where a, =cp(r‘ -r‘)/h,‘ Byefimfimthedropletwmperanaemlongachangeswithfimenthewetbldbsme andfromEq.(40) ’ 39 providing a relationship between concentration and temperature. A second relation is provided by the vapor pressure relation, Eq. (32) allowing solutions for Ymand T .” Ifdleambienttempetann’eishigh. Sn andifYFo =0,wehave Yr” =Br /(l+BT) SinceB,isaconstantatthewetbulbstate,Eq.(38)indicatesthatl(isaconstantlmder thesecondifimaasidefiomvmiafionsmlemconvecfimforthenmchculafingdrop, theenergyequationfortheliquidis at a a 2&1 3=+3(' ‘) <4” 61' r—O, a— 251' 2 . r=r:(t), kfr 3:4121' {Ta-TQ-mlh, 8 The conservation ofdrop mass relationship, Eq. (37), is unchanged. Given the surface tempa'ature, Eq. (32) provides Yr, , and Eqs. (32-35) yield the heat and mass transfer rates. The sohuion proceeds by integration oqus. (37) and (40) or (41). Analytical solutionsarenotknownformeseequations forrealisticvaporpressurerelationsand numerical integration is normally employed for drop life time calculations. 2.6. Exciplex (excited state complex) fluorescence Aclassofextremelyimportmtcoupledrenctionsinvolveexcimersand exciplexes.Mmymolemfleswfllnmasocimeorreactwitheachotherinthegromd stateHowever,mceoneofthemismanexcitedsute,itwinmdugoreacfionswith othernormflymn-rencfinggrmmdstatespeciestoformnewmnemexcitedsmte species.1hisisnmapammentchemicalrencfiommdwhenmenewexcitedspecies decaystheaiginflgrmmdsmtespeciesmeformedlfthemacmntisagromdsme molemfleofflteametypedlemwspeciesisanexcimucxcimdmdimahlfmetwo spedesuedifi’umttheyuemfledexciplex(excimdmcomplex),dmoughifmemo speciesiuesimilar,theterm"mixedexcimer'issometimesused.'°‘ The absorption and fluorescence spectra of organic molecules dissolved in nonpolarsolvenmalchastypicalfuels,areviltuallyidenticaltothespecuaofthesame moleculesinthevaporphase.However,insomecasesitispossibletoreactthe fluorescing excited state molecule, M', with an appropriate ground-state purmer, G, to famesecondflmruchgspecieeM—G‘,knownasaciplex(£‘).1heflmrescenceof thisexcitedstaecomplexE’isdwaysred-shifledadmrespectmmnofmeexciwd monomersmissimM‘becausetheexciplexhasalowerenergydrmmbormdM’Jhe basicshemialmechfisnismeamefmanexdplex-bandsynemamzmsincemeyafl confiinthreebasiccomponents;ammomer(M),agrmmdstatereacmnt(G),anda solvent which serves as fuel. The exeiplex is formed in the reversible equilibrium '°‘ M‘+G e—eE‘ 41 whereM‘isthefirstexcitedsingletstateofthemonomer,andE‘istheexciplex.The ethbfiumofeqrmfimwhichgovamthefwmafimoftheexciplexmthehqludhnnot thehpstconfiofledbyadjmfingthecomennnfimofmegromdmereacmnflG). lhewwdengthofexciufimsmncedetammeswhichexcimdelecaomcsmteismched diningtheabsorptiontransition. FiglueSillusn'atestheindividualstepsoftheexciplex mechanimlntheliqludphasedleeqlnhbriummaybeshifiedfartothefightby inaensingtheconcenuafionomedG.UpmvaporizafiomtheconcennafionoftheM dewfllbereducedsothatM‘emisdmwifldommateMoreoverJheeqluhbfilmis temperanuedependentshifiingtotheleftwithincreasingtemperauues. Accordingly,the exeiplex is less stable at elevated temperatures and the exeiplex fluorescence emissions atmnuateerespectmthatformemmomer.Theenergyofmeminedphmonfiom thefluorescenceisgivenby £2 .1 E: where It is Planck’s constant, c is the velocity of light, and A. is the wavelength ofthe emission fluorescence. Sincetheexciplexemisfionhasalowerenergyitsemission occlusatalongerwavelengththmthatoftheexcitedmonomer.Figlne6shows the exciplexmgydiamFmthisreammmelasaindwedexdplexflmrescence (LIBEtechniqmthevapmphaseflumucenceisdommamdbyM‘mdmefiqlfidphase emission ischaracterizedbythe red-ahifiedemission of the exciplex.’”°“m 42 l. M+UV ‘—"M‘ 2. M’+G —*E‘ Green Fluorescence 3. M G Liquid Phase CONCEPI‘ l, 4 : Lasetexcitation ofmonomerin liquidorvapot 2 : Exciplex (excited state complex) formation, liquid phase only 3 :Exciplexemission flmctmceoccmsingreen 5 : Monomeremission fluoteccenceoccurs in blue Figure 5. The mechanism of exeiplex liquid and vapor visualinfion systems." 43 AH~10-20Kca1 \ _ AH M*+G \ \ ,” 1 E" ’ \ I ‘ I ’ ENERGY I I I Figure 6. Exciplex energy disarm!" Chapter [I]. Experimental Methods (1): Development of New Esciplex Systems 3.1.Chaprerovervlew [nameyofpotenfiflexdplex-basedvapodfiqmdvisnliafimsystemsmain selecfimcritefiafomsedonthesnongmomtempaanneexciplexemissionwitha reasonableseparafimfiommemmomeremissionmdretenfimofmbaannalexcipkx emissimutempaannesashigthOO‘C.Mmyexisfingposubkexciplexsystemsm were investipted. However, none of them completely satisfied our criteria. Some yielded strong emission eficiencies but spectral separation between monomer and exeiplex was notsuficientmrviceversa. Thischapterdescn'besnewvaporfliquidvisualiutionsystemsbasedonan exeiplex formed between dimethyl- or diethyl- ntbstituted aniline and trimethyl- substituted naphthalenes in an isooctane solvent. Characterization methods employed steady-state and time-resolved fluorescence emission spectroscopy. Among the many syaems and formulations investigated in this study, an exeiplex consisting of 7% l,4,6- trimethylnaphthalene (l,4,6—‘I'MN) and 5% N,N-dimethylanilme (DMA) in 88% ismcmnewufomdmhemebeasystemfmlaser-Mudexciplexflumescencealfin vimflzafimuaedmobsavemixnneformanonindieaelorspukigmfionengines. Wonofspecuanywensepuatedfluorescencefiomazmonomerinthegaseous phasemdfiomexciplexmmegasolmefinlmqmresthatmeexcipkxfmmingdopanm have boiling points within the distillation range of gasoline (20°C to 210°C). Tempaannedependentmummenmofmeexciplexflumeecencehaveshownthatme quenching mechanism leading to exeiplex formation is sufficiently favorable to permit 44 45 theexciplextoexistattemperaturesapproaching 150°C.Theseresultssuggestthatthe DMA-TMNexciplexsystemshmddbeusefidasadhgnosficfmtheviwalinfimofgas and liquid distributiom in flows. 3.2. DcvdopmeatJLChancterbatioa ofaewexelplexsystems 33.10ptlcaleharacterbatloa A. Sample preparation and apparatus 'I'heTMNisomerswereobtainedinpm'eformandusedasreceived. 1,4,5-TMN, 1,4,6-TMN were purchased from Carnegie-Mellon University. 2.3,6—TMN was purchasedfiomAldrichChemical Company, aswasDMA. DEA. andisooctane(2,2.4— trimethylpentane) of spectrophotometric grade. All chemicals are reported here on a weight basis. The normal boiling points '°’ of 1,4,5-TMN, 1,4,6-TMN are 140-142°C and 145°C, respectively at low pressure, and the melting point of 2,3,6-TMN is loo-102°C, those of DMA md DEA are 193-194°C and 217°C, respectively. AbsorptionspectraweremeasmedonaOlis-modifiedCaryU- spectrophotometer. Emission spectra were measured by a F4500 Hitachi spectrofluorimeter or a highoreaolution instrument in the Department of Chemistry."0 Theexcitationwavelengthwas308nmandemissionwasrecorded300nmto$50nmat scanspeedsof240nm/min.Theexcitationandemissionbandpasswere5 or 10m The PMT was a Hamamatsu R1104 or compm-able tube with applied voltages ranging fiom 40°to700V. B. Spectroscopy ofthe sample Theexciplexesmloyedinthissmdyare formedfi'omthecomplexesbetween dimethyl. or methyl-substituted aniline (N,N-dimethylmi1ine (DMA) and N,N- diethylaniline (DEAD with 1,4,5-, 1,4,6- or 2,3,6-trimethylnaphthalene (1,4,5-TMN, 1,4,6-TMN or 2,3,6-TMN, respectively). Figure 7 shows the absorption and emission spectraofmeanilines.Theabsorpnonspecn'aofDMAandDEAaresimilarand charactuizedbyabroadhandwithlu=298 and303 nm,reepectively. Excitation in thisabsorptionmanifold(1_,._=308nm)producestheintense fluorescencereproduced in Fig. 7. The absorption spectra ofthe TMNs, shown in Fig. 8, are typical ofnapthalenes. Thebroadbandsfeannepmgressimsinmewbrafimsoftheuapthyldngzmthis “brafimalprogreedmisnvealedbymeshupfinesuucnneovedaidonmeabsorpnm profile. ManyofthephotophysicalpropertiesoftheTWisomersarenotwell known. Notwithstanding, the energy ofthe intense fluorescence is relatively invariant for the three isomers with 1.“. = 343, 342, and 339 nm for 1,4,5-1'MN, 1,4,6-1'MN and 2.3.6- TMN, respectively, upon excitation with wavelengths A... < 330 nm. The 308-nm excimfionfineofaXeClexcimalawiscapableofpromtcmgeithermflmeorThm excited states. Thus. the exeiplex formed in the experiments results fi'om the interaction ofelecn'onicallyexcitedTMNmonomerwithgl'omdstateanilineorvice versa. The hydrocabmfiwhisoomdoesnMabmbattheXeCIexciufionfiequencyasshown inFig.9,therefore,itisphotophysicallyinertinthesestudies. 47 «(8) (b) x H \ u' (a) (b) 1. \ / \_ L L 1 -.l ..___ Relative Absorhanc mum: W3 “IOU 270290310330350370390410430450 Mum Figure 7. Absorption and emission spectra for (a) DMA and(b) DEA monomers in isooctane at roomtempennm WWI WERE , A u 4 . 270 290 310 330 350 370 390 410 430 450 Mom Figure 8. Absorption and emission spectra for trim¢hyl.naphthalenes (a) 1,4,6-‘1'MN, (b) 1,4,5- TMN, and (c) 2,3,6-TMN in isooctane at room temperature. 48 3.5 I’TrT—riiVTTfiI'Tr‘l—FVTIITWTYTTYTT'O[rrYjITo 2.5 L A liLLLLllIAl—l44LLLLLl[1114J414114LLLL44L ahaothancateu.) ..b .. tn III‘ll!I"IIIIIWI‘IIII]IIUI‘IFIIIIII ..LLL 4.1. LLLJJJ—LJJJ—LLL LLLLHJ. 444.1 ). -0.5 100200300400500600700800900 W (urn) Figure 9. Absorption spectmm of the isooctane (2,2.4—trimethylpentane) solvent at room temperature. 49 C. Qaaatumyieldofsample lnorda'tomeastueqmyieldsofmeDEAandTMN-isomeramchaslAfi- TMN, 1,4,6-TMN and 2,3,6-TMN, the absorption and emission spectra of these moleaflesweremedFigmes7md8showthosespecnalreadts.Ansampleswere premedmmisoocunesolventlnmiseapefimemaflemissionqaecmweremcorded from 300 nm to 550 nm with a F4500 Hitachi spectro—fluorimeter. The excitation and emissionbandpasswalenm. The absolute fluorescence quantum yield (Q) as a {motion of excitation wavelength (A) is the ratio of the number of emitted fluorescence photons to the number of absorbed phototu,"o w) =11» ”.00 (42) whacMUismephomnmtemityofflmrescmcemleAflsmemmnsityofradiafim absorbedTheabsabedphMMintensitycmbecdctdatedfiomtheBeer-Lmbathw, 1.1.61) = MIDI 140"] whereI,(/1)istheincidentphotonintensity,s(/D isthemolarextinctioncoeficient (M'cm”),cisthemoluconcentration(M),andlistbeopticalpathlengthofthesample (cm). Because numerous factors govern emission intensity the measurement of absolute quantum yield is dificult. These include variations in source intensity as a function of 50 fime.variafionsinthcspecnalresponseofthedetectorsystemasafimcfionof wavelength, bandpass error, difl'erences in collection eficicncy of the collecting optics as afimcfionofabsorbanceandinnerfilterefi‘ectsueflections, polarintion,and reabsoption-reemism'on efl'ects.'°’ Accordingly. relative intensity measurements are prefaredmdthequmnmyiddistypiuflyobmmedrelafivemmeemissimmtensityof aknownsampleatagivenexciufionwavelalgmmintegratedareaunderthe flmrescemeemissionmisproporfionflmthemnlmtensityofflumescemfight emiuedbythewlmiommdmismmispropmfionalmlgyecllfmefluorescence emission spectra oftwo solutions are measured with the same apparatus under identical conditions, the ratio of the integrated fluorescence intensities"° is given by, IMAM/u Dram = D, I Dr = Ase! M» ' (43) where}, and f, are the quantum yields ofthe sample of interest and reference, respectively, and A, and A, are the respective optical densities defined by egg-x5 . The mngratedmtensifiesshmddbecanaedforthesponspomeofmedaectmsystem However,whentheemissimwavelengmsofmenfaencemdsamplemesimilu,asis thecasehere(theinvesfigatedmommaemissimmaxhnauebam333md343nm), mistemcmbeneglectedlnmemeaaremenmqumnnnyieldswerereferencedm DMA,“ whichlnsaknownfluorescencequantmnyield of0.11 incyclohexane. Because wewishedmmasaethequannmyieldsofmummomersintheisoocmnehydrocarbon solvent, Eq. (43) needs to be corrected with a factor of (n,/n‘)z to account for the 51 difl'erent refractive indices of the sample and reference solvents.“ Rearranging in terms of the sample quantmn yield gives d. = MA.D.n.‘/A.D.n.’] ' (44) SincetheabsolmequanmyieldofDMAmcyclohexmeisknowmdtevalues ofthe other DEA and the Ma are simply obtained from Eq. (44) (n,=l.426 for cyclohexane and n,=l.391 for isooctane). The fluorescence quantum yields of the monomers. measured for 308 nm excitation. are shown in Table 1. These high fluorescence yields areadhectqmnfimfivemeanneofmemmnseflmrescencethmwmactefizesthese species. Table 1. Relative Fluorescence Quantum Yield of Investigated Monomers nm 1,4,6- 342 0.065 1.45. 0.069 .6— 0.071 ' 34 DMA 0.1 l 3.2.2. Optical properties of exeiplex formers A. Room temperature measurements A-1. Photopllyaics of DMAoTMN exeiplex system Molecules in an electronically excited state M‘ may be highly polarizable and can interact with other polar or polarizable species. An exeiplex is produced when M‘ 52 associateswithareactingpartneeroyieldacomplexMQ)‘ more stablethanthe individualconstituentsM‘orQ. Iftheexcitedstate’sreactingparmerisadifl'erent molecule Q,thentheexcitedstatecomplcx (MQ")iscalledanexciplex."""2 Atypical scheme fortheoverallexciplexprocessisdepictedbelow. M‘+Q _____, (MQ‘sE‘ L—o M+Q ‘___, meanelectronicstructmeperspecfivetheHOMOsofMananrefilledand thusmaeismappreciablegr'omdstaeimMmAccmdingly,meabsenceofm appreciablemfionbetweengrotmdeandQexcludesstaficquenching mechanisms of M’. However, excitation of one M’ permits a bonding interaction to residtfi'omtheinteractionofthepartially filledbondingandanti—bondingfiontier orbitalsof M‘withtheHOMOandLUMO, respectively, of QasshowninFig. 10. The consequences of these energetics me several-fold. The exeiplex complex (MQ)‘ is mbfliuerespecth‘mdhencetheflumeacenceofmeexciplexisnd-shified floor that of M‘. Moreover, (MQ) is tmstahle and therefore, the decay of the exeiplex excitedstatereturnsthesystemtoadissociativegrumdstateM+Q Finally,the exeiplex is formed at the expense of monomer by the direct reaction of the electronic 53 The Chomsky ot Exclusions (E16910! =9 Excited Coma-a) n—~I.~.-—-K air-9+- eff-c a it "H—fi‘ l a J +...+,..+ +‘L Figure 10. The photo-chemistry of DMA-1,4,6-TMN exeiplex system. $4 mmmouasxdplusym .......(0) mm) was “ES” t:1.3~ tan“ Figure 11. The grand-state configuration of the DMA-1,4,6-TMN exeiplex system in isooctane solvent, at room temperature. 55 exdtedshteMdeqwnchingisthmadymmicprocessdependingdirecflymme concennafimonJhechemicalarmhmasweflasthephmophysicsofmegromd stateconfiguration oftheexciplex formationofDMAand 1,4,6-‘1'MN, are shown in Fig.11. :- DMA-(memexdpbxm Inaseriesofewerhnents,eitherDMAor2,3,6-TMNwasusedasamonomeror gromdstatereactant(quencher)tofindoutwhichconfigmationyieldsabetterexciplex. In the DMA-(2,3,6-TMN) exeiplex system, monomer (either DMA or 2,3,6-‘1'MN) weightpacenmgewasfixedatS%mdquenchermolarconcenuafimwasmcreasedfiom 0%to 10%. Figure 12 showsthefluorescencespearumof 5% DMA monomerandits change upon adding 2,3,6-TMN in isooctane with 308 nm excitation. DMA monomer solution exhibits fluorescence quenching as the concentration of 2,3,6-TMN increases. Withomaquencher,mewavelengthcmrespmdingmpeakemissimintensityof5% DMAis343nm.Asthemolarconcenuationofthequencherincreases,theDMA mommerflmreacencemtensityat343nmwasdecreasesgreaflyandanew,broad emissimbmdwudevelopsgndnflywithmemuhnumemissionintensityaromdwl nm.ThisnewemissionfeahneisduemexciplexfmmafionbetweenDMAand23,6- TMN. Further increase above 10%(w) 2,3,6-TMN did not improve the exeiplex emissionFigure 13representshowthemonomerandexciplexfluorescenceintensity change with respect to 2,3,6-TMN concentration. The 5% monomer intensity at 343 nm 56 unlaslonlnhnshfleu) § § § § § ""l""l""l""l‘7_" O Figure 12. Thefluorescencespectrmnof5% DMAandthefluorescencespecuauponthe addition of 2,3.6-TMN in isooctane at weight ratios of (a) 0.1 %, (b) 0.5 %, (c) 1.0 °/o, (d) 3.0 %. (e) 5.0 %, (t) 7.0 % and (g) 10.0 %. The fluorescence intensities of monomer and exeiplex are relative to each other. 0 2 4 6 8 10 96“ of 2.3.6“ Figure 13. Fluorescence intensity changes of the 5% DMA (a) and the resulting exeiplex upon the addition of 2,3,6-TMN in isooctane (b). (a): monomer intensity change at 343 (am), (b): exeiplex intensity change at 401 (nm), respectively. 57 decreasesexponenfiallyandnewemissionawl nmduetotheexciplexincreases logarithmically. The exeiplex intensity dominates at values of 2,3,6-TMN concentration greaterthm 1.6%. Exciplex emission seems to be fully developedat 7% 2.3.6-TMN. TheexperimentofFig.12wasthenreversed.DMAwasaddedasaquencherto anisooctnesolution of a fixedweightpercentage(5%) of 2,3,6-TMN monomer. Figure 14showstheemissimmtensityofa5%2,3,6-TMNwithrespeammecmcenuafimof DMAW‘Mamencher,&ewaveleng&atpeakemisfimmmmityof5%2,3,6-TMN was 347 um. Similar to the 5% DMA-(x)%2,3,6-‘I'MN exeiplex system, the 2.3,6-TMN monomerfluorescencemtensitydeaeasesmdanewbroademisfimbandgradmfly developsarmmd400nmwhentheDMAquenchermolarcmcenuafimmcreases.This newemiss’onfeatureisalsoduetoexciplex formationbetween2,3,6-TMNsDMA. Fm'ther increase above 7%(w) of the DMA concentration does not improve the exeiplex emissionThemtensitychmgeofbommemonomermdexciplexflumescenceufixed wavelength is shown in Fig. 15. Therearetwodisfinctdifl'ermcesbetween5%DMA0(x)%2,3,6-Mand5% 2,3,6-TMNO(x)% DMA exeiplex systems. One is the difl‘erence in emission intensity, mdmeothais“aos&ulk”,thespecnalwavelengthsepmafim(AA)hetweenmonoma and exciplex emission. At fixed concentration, the DMA02,3,6-TMN exeiplex system hashigheremissionintensities, forbothmonomermdexciplexrelativeto2,3,6- TMN-DMAThefmmerexciplexsystemhadasfighdyluguspecn-dsepm'afionAA thanthelatter; 58nmfortheformer,and53nmforthelatter.Theseresultsareevidence that the DMA02.3.6-TMN system has better exeiplex formation than the 2,3,6- TMNODMA system. 58 § 'é’ atnlaalonlntsnalty(a.u.) .. to 8 8 IIll'llll'lllIllllllllll’lrrrl 100 50 o L 300 Figure 14. The fluorescence spectrum of 5% 2,3,6-TMN and the fluorescence spectra upon the addition of DMA in isooctane at weight ratios of (a) 0.1 %, (b) 0.5 %, (c) 1.0 %, (d) 3.0 %, (e) 5.0 % and (t) 7.0 %. The fluorescence intensities of monomer and exeiplex are relative to each other. lrlll'Illllllll ll lllll'l o LL1_I_1_I AJLLL4LLLLLL1 IIJ_LL4LALLLIIIJAJJ 0 1 2 3 4 5 6 7 8 WetofDMAquanehar Figure 15. Fluorescence intensity changes of the 5% 2,3,6-TMN (a) and the resulting exeiplex upon the addition of DMA in isooctane (b). (a): monomer intensity change at 347 (nm). (b): exeiplex intensity change at 400 (um), respectively. 59 b. DMA0(1,4,6-TMN) exeiplex system This section only concerns the 5% DMA-(10% 1,4,6-TMN exciplex system since itisknownthahatfixedconcenfiafiomdieDMAsTMNsystemisabenerexciplex fmmermanTWODMAaexplainedahoveinSec.eFigme16showstheabsorpfion mdemissimspecnaofDMAmdlAfi-ThmmisooctaneTheabsmfionspecmunof theTMNistypicalofnaphtlnlene.Excitationintliisabsorptionmanifold(1_,<330nm) produces inwnse fluorescence (s.’°'~ 0.065) with A“ ~ 342 nm. The absorption specuumoftheexciplexformmgcomtapmtDMAJscharactaizedbyabroadband with}...~298mthecompotmdfluorescesatamaximumof334nmwitha quantum eficiency of0.11. Figure 17 shows the evolution ofthe exeiplex emission with theadditionofl,4,6—TMNtoS% DMAinisooctaneatroomtemperatme.Theintense fluorescence of the DMA monomerismarlredlyafl‘ectedbythe addition of 1,4,6-TMN. Themommeremissimissignificmflymdwedwiththeaddifionofitscongenermda new,intenseemissionappearsatlowerenergy,consistethhfmmafionofa DMA0(1,4,6-TMN) exeiplex. It was fund that solutions with molar compositions of 5% DMA and 5% (to 7%) 1,4,6-‘1'MN give the highest fluorescence yields of the exeiplex. Asthe1,4,6-TWconcenuafimincreasenthe5%DMAmonomerintensityat343nmis reducedgrestlymdanewexciplexemissionatNSnmdevelops. ltisobservedthatthe exeiplex intensity is dominant at 1,4,6-TMN concentrations greater than 3%. By 7% 1,4,6-TMN, the exeiplex emission seems to be fully developed. The intensity change of both ofthe monomerandexciplex fluorescenceufixedwavelength is shown in Fig. 18. 1 E 4 1 c1 .1 '1 '1 .4 J 1 .4 ’5‘ .— v 1 d i '1 4 --< .1 i tslstlvaahswhancuau) 2's 3 .. .. ' WWI—LlJJ—LAILLLJ i‘fl‘lwm a 0. a0 0 s .0 .0 Q z.- . . .‘ U n C .O .0 .0. . JtLl—l-'.‘J rauLLL4111411 .. wavelengtMnm) Figure 16. Absorpn'on and emission W for (a) 1,4,6-TMN and (b) DMA monomers in isooctane( 0.8 MA) at room temperature. m I I 1 T rrTj rT T r‘U—‘I’ I T TT I; § § unlaalonlrthnaltytau.) a Ujil'llIFIITUI'IIII'IIYIFFIVI 0 Figure 17. Thefluorescencespectnnnot'5% DMAandthefluorescenccspectrauponthe addition of 1,4,6-TMN in isooctane a weight ratios of (a) 0.1 %, (b) 0.5 %, (c) 1.0 %, (d) 3.0 %. (e) 5.0 %, (t) 7.0 % and (g) 10.0 %. The fluorescence intensities of monomer and exeiplex are relative to each other. 61 Since ultraviolet excitationiscapable ofproducingeitherDMAorTMNexcitedStfltc-‘I. theflmrescencewasmonitmedat400nmutheexciufimwavelengmwasscannedto assessthedomhnntabsmbingspedesthuludswexdplexformafimTheexcimfion profile of the exeiplex matches tint of the TMN: 1.41.4,6-TWOBMA) =- 340 um; W8330nmforisooctanesohlfionscontaining5% DMAand5%TMN. Exciplexemissionisobservedatwavelengthswell intothenearultravioletdueto pronotmcedabsorptionatthssehighconcentrations. ltisprobablethatexcitationofthe DMAwfllalsoladbexciplexformafim.However,TMNisthestrongerabsorbing spedesmdieUVatmehighcmcennafimsneededfmexciplexformafion(absorbmce (A)=4.12at388nmfor5%TMN,whereasA=1.41 at388nmfor 5% DMA) thereby obscuring absorption by DMA. Finally, the attenuation of the excitation profile at higher wavelengthswellbelowtheabsorptionmaximmn ofTMN, isalsoduetothehigh absorptivitiesofsolutionsatl..>388nm.Forthesewavelengths,alltheabsorption occm'satthefiontfaceofthecell,outofd1elineofsightofthedetectionoptiesofthe instrument. c. Geaueldiscassioaofesclplesresalts TheinmnsefluorescenceoftheTMNmonomersismarkedlyafl’ectedbythe addifionofthembsfinnedmflines;simflarly,DMAmDEAflmrescenccisaltered significantlyinthepresenceoftheTMNs. Foreithersituation,themonomeremissionis significandynducedwithmeaddifimofitscongenamdanewfintenseemission appears at lower energy, consistent with the formation of a DMA-TMN exeiplex. It is formd that solutions at compositions of 5% DMA and 7% TMN give the highest 62 fi' 1 r s lfi‘r' fT—j T fiTfi f"! T '— E 1 .1 .. E' 1 5 ~ . s. L J I I I T. J I- 4 C I :_ “:1 _‘ i 1 .T' 1 I l bL I 11LJ_A L_.l 1 LL; 14 LLmlLI‘ 0 2 4 6 8 10 %wtof1.4.0-TIN Figure 18. Fluorescenceintensitychangesofthe5% DMA(a)andtheresultingexciplexupon the addition of 1,4,6-TMN in isooctane (b). (a): monomer intensity change at 343 (nm), (b): exeiplex intensity change at 405 (nm), respectively. y f1 1 fr Is T T l V f* ~ 500 _- 1 ‘e__ I 1 i 400 ‘_' 1 g 300 Z- -Z i (D) j i 200 1' (61m ‘."’°--§ - 100 E- t o . 300 400 450 500 What) Figure 19. Fluorescence spectra of(a) 5% DMA and the resulting exeiplex fluorescence spectra upon the addition of (b) 7% 1,4,6-TMN, (c) 7% 1,4,5oTMN and (d) 7% 2,3,6-‘1'MN in isooctane. Relative specs-a1 intensities are presented. 63 fluorescence yields of the exeiplex. Moreover, as observed from the relative spectral profiles of Fig. 19, the DMA-1,4,6-TMN exeiplex system has the higher fluorescence peakunissionmtalsityandlugaspecualsepmafion(M)betweenmonomerand exciplex.Thewave1engthtdsepeakexciplexemisdmmtensityis405nmfm7%1,4,6- ‘1'MN,401nmfor7%2,3,6—TMNand382nmfor7%1,4,5oTMN,respectively.inorder masseuwhahaDMAorDEAismeheMexciplexformingconsfinentwithmeTW system,theexpaimemdFig19wasrevasedmdmflineaddedmmisoocnnemlufion of a fixed concentration of 1,4,6-TMN. Figure 20 shows the results for the addition of 5% DMA and DEA to 5% solutions of 1,4,6-TMN. As observed in the complementary expefimentdesmbedabovethemtenseflmrescenceofmemmomer,mmiscaseTMN (A,_=350nm),isattenuatedefficientlyinthepresenceofthcanilinewiththe concomitantyowthofalowenergyexciplexfluorescence(1_=402,400nmforDMA and DEA, rewectively). As is evident fiom the relative exeiplex fluorescence intensities, DMA forms a slightly more stable exeiplex (higher intensity) and better spectral resohtfim(A1)isobservedwithregudmmemonomeremismomThesamereadtwas obtained forthe 2,3,6—TMN and its exciplexcongeners (i.e., 5% DMA or DEA) shown at Fig.21. Theserendtsareconsistaitwithexciplexformafionasarestdtofthedonor (DMA)-acceptor (TMN) interaction between proximately disposed constituents. Because theDBAhasmorecarbonsontheaniline(diethylversusdimethylofDMA),itis staicaflymmebulkydeMAmdthusitmmeweaklyassociateswiththeTMN monomer,dierebyacc0tmfingfortheslighflyweakermdblueshifiedemission 'T r'fiwr' 311' I I' I T I T11 ' s rfi nT‘lfifi .T alLramJu4+ o 8 I'llI‘T—lUT[Urilf'TUlellIljfi'T'ITII \ L‘ F )- i' r. L. Figure 20. Fluorescence spectra of (a) 5% 1,4,6—TMN and the resulting exeiplex fluorescence spectra upon the addition of(b) 5% DMA. (c) 5% DEA in isooctane. Relative spectral intensities are presented. m Fj' r—Tfifi r I rfir r? I T [—1 '7 fit I I 1' TT 4 350 E- .3 -.~ E 1 3 300 L- 1 3 E : 250 .- ‘3 b 4 g 200 :- ‘3 h d L. .1 150 L ‘1 - 1 I- d b u 5 3 5° :' ..s' T. : r ‘ o .- LiA Figure 21. Fluorescencespectraof(a) 5% 2,3,6-TMN andtheresulting exeiplex fluorescence spectraupontheadditionoflb) 5% DMA,(c)5% DEAinisooctme. Relativespectral intensities arepresented. 65 Table 2. Monomer and Exciplex Emission Wavelength of the Investigated Exciplex Systems. , Exciplex Systems (% wt) Monomer Peak Emission Exciplex Peak Emission Wavelength gag): Wavelength (.0: s! l 5% DMA-7% 1,4,5-TMN' 343 382 5% DMA-7% 1,4,6—TMN' 343 405 5% DMA-7% 2,3,6-TMN' 343 401 5% l,4,6-TMN¢5% DMN’ 350 402 5% 1,4,6-TMN05% DEAF 350 400 DMA: N,N-Dim¢hylaniline, DEA: N,N-Diethylaniline, TMN: Tfimethylnmhthalene. The % Wt fortheisooctmesolventis'88%and"90%,respectively. of the TMN-DEA exeiplex complex. The emission wavelengths of monomers and exciplexes investigated are shown in Table 2. A-2. Exciplex qaeachingaaalysis Theexcitedsmtemonomerreactswithitsexciplexformingpmmerinoneoftwo pomblewaysintamsofaDMAexcitedstatedretworeactionpathwaysare 0MP + TMN aDm + Tm (dymmic quenching) (45) 0554‘ + Tm é [pm 4» TMVI’ (static quenching) (46) ForEq.(45),theexciteduatemommerractsuddri8pumamabimoleadureacfiom andtheexcitedstateissaidtobequencheddynamically."3 Alternatively,asdescn‘bedby Eq.(46),meexcitedstatemonomermayreaabyitsassociafimmformmexciplex (cafledsmficqmnching).Fmbommedynamicmdstaficquenchingpathways,the 66 excited state concentration of monomer is diminished, therefore, the monomer flmrescenceisefl'ecfivdyauemmequuenchedwithmcreasingcomenuafimofits exeiplex-forming partner. The quenching processes are quantitatively described by a modified Stern-Volmer’treatmentm 10/1 = (1 +KSJTWXI +KITMNI) (47) whaetherafiooftheintensifiesofDMAfluorescenceintheabmandpresenceof TMN,(Io/I),isrelatedtotheconcenuafionofaquencherviatheStern-Volmerconstant, szmdmequihhfitmconstamKTheStern—Volmercmsmmatheproductofthe quenchingrateconstantandthenaturallifetimeCKw-sk‘ro). ltthereforedescribesthe dynamic quenching of DMA’ by TMN (i.e. Eq. 45). The overall quenching reaction is augmentedbytheassociafionofDMA‘andTMNudescnbedbytheequflibrium constanthorEq. (46). Ifbothstaticanddynamic quenchingprocessesareoperative, Eq. (47) predictsan upwdcurvatureof the plot of loll versus [TMN]. Conversely, if qmnchingocausbyonlyonemechmianagv=0mthecaseofstaficqmnchingand K=0fordynamicquenching),thenEq. (47)reducestotheform [on --= 1+K'rrMN] (48) togivealinearrelationbetweenio/Iandfl'W]withmintaceptoftmityandaslopeof K'; forthedynamicquenchingK'=KwandforthestaficquenchingcaseK'=K. 67 a. DMA-(256M) exeiplex system Figure 22(a) plots [o/Iratiosobtainedfi'omthevarious DMAemissionbandsof Fig. 12vm2,3,6—1Wconcenfiafion.1heplotfoflowsasquarehwandtheupwmd Wisaclemhdicafimthfitheexdplexqmchhgmechflunconmbommc anddynamic contributions. A fitoqu.(47)tothesedatayie1dscoeficientsk. =37.6M" andk,=2.7M.'.Thenatura1lifetimeofDMAinnonpolarsolventssuchasisooctaneis 2.4ns."‘ [fitisasmnnedthatdredymmicquenchingrateconstantisdifiilsionafly controlled (kq :- lo’ it"s"), which is reachable for an exeiplex system, then lg, is calculatedtobe2.4M,-‘ whichisingoodagreementwiththesmallerconstantkz obtainedfi'omthefitoqu.(47),andthusidentifiesitasthedynamiccomponentofthe quenching process. Accordingly, the equilibrium constant for exeiplex formation was K =37.6 M-‘thercfore,theDMAquenchingmechanismwasdominatedbyexciplex formafionThisremhisenfirelyconsistemMmmeappearanceofbfiduexciplex flmrescencewimmaessmgqmnchingofmmomaflumescenceinthesesynemslfmis weremtdiecasethendreDMAflmrescencewmddbeqtenchedenoobsavafionof a red-shifted emission. Figure 22(b) shims a 5% (2,3,6—TMN) monomer quenching feahnewidrtheaddifimofDMAinisoocminconcastmmereadtsaboveaplmOf theMemityrafioaJDislinemindieDMAconcennafionThiscleulymeamthatmly oneofthestaticordynamicquenchingprocessesisoperativeinthissystem.Afitoqu. (48)tothesedatayieldsava1uefor1(’=22M'.1tisbelievedthatthissystemcanbe explained with the static quenching process only, otherwise 2,3,6-1'MN fluorescence would be quenched with no observation of a bright red-shifted emission. 68 TTrTrr'I—rrT—TI—ITTT—rlfiilfljfiOTTTT (If!) 25 _— a E i1 _ 15 - 4 - I I 10 :— (b) j 5 E- i ' 1 0 ”indium“Ulurrihuua....J....L1.u.‘ 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 MolarOonesntratlonMotOueneher Figure 22. A modified Stern-Volmer plot with respect to the molar concentration ofthe quencher; (a) 2,3,6-‘1'MN, (b) DMA quencher, respectively. The solid line is a fit of Eq. (47) and (48) to thae data. 35 —f rT r—l T'I—T'j r1 fi—‘rfiT'T fiT T—TI Ffi f1 30 11111 I." IILI1LL111AALIJLL111J1 o l L_1_J_LLI_1 g1 PLJ L1 #4 LJ LJ__L_1 0 0.05 0.1 0.15 0.2 0.25 molar eoneenuatlon (I) of the 1.4.04" quencher Figlne 23. A plot ofthe [.11 (hits ofFig. 17 versus the concentration of 1,4,6-‘1'MN in isooctane solvent. The solid line is a fit of the modified Stem-Volmer Eq. (47) to these data. 69 b. DMA0(1,4,6-TMN) Exciplex System Figlae23plots1flrsfiosobtainedfi'omdrevariousDMAemissionbandsof Fig.17 versus 1,4,6-TMN quencher concentratiom [1,4,6-TMN]. The plot follows a smnrehwmdtheupwudanvannemdicmeeaaficmddynamicquenchingAfitoqu. (47)totheeedatayieldsconstantsof70.1M"and3.8M'.ThenaturallifetimeofDMA innmpolrsolvennanhuimocnnerZAnsAgamifitisasumedmatthedynamic quenchingrateconstantisdifl'usionallyconn'olledmz10?M"s“)whichisreasonable foranexcipleasystem.thenK,iscalculatedtobe2.4M',whichisingoodagreement withdresmallerconstantobtainedfiomthefitoqu.(47).AnequilibriumconstantofK = 70.1 M“ suggests that the DMA quenching mechanism is dominated by exeiplex fomafionThismhisdsoenfirelyconsimethhmeappearmceofhightexdplex flumescencewithhaeashgmnnchingofmmomaflmrescencemmesesystemslfmis wenotmecasethenDMAflmrescencewmddbeqwnchedwithnoobsavafionofa red-shifted emission. Among the investigated DMA-TMN exeiplex systems, the DMA-1,4,6-TMNsystemhasthehighest equilibrimnconstantm), meaningthatitisa stablemdprowceebemerexciplexformafionasindicmedbypreviousrendts. A-3. Fluorescence lifetime measurements Lifetimes of DMA andthe DMA-1,4,6-TMN exeiplex weredetermined bytime- correlatedsinglephotoncotmtingwithaninstrumentintheIASER Laboratory.The fimecmnlatedphmmcomfingset-upwasmodifiedfiomapreviousconfigmafimm in ordertodecreasethetransittimespreadofmonochromator, andconsequently improve 1 70 the overall instrumentresponse function. The SPEX 1681 monochromator was replaced withambnacfivedoubkmmochromatmfiomCVILaser.merebyeflmmafingtheuseof a 10mmmaskonthedifiacfiongrafingThe220psrisetimeofthedetectorwas improvedto 156psbyreplacingthe 12 umR1564UlllMPC-PM1‘G1amamatsu)witha 6 pm R3809U MPC-PMT (Hamematsu). The shortening of the detector’s risetime necessintedacmcomitmushmwningofthe9cmiManddehyloopofmeflmrescence channel of the CFD (Tennelec CFD 454). Finally, a TAC biased amplifier (Oxford InstrumentsModelTC 864)replacedtheTC862,changingtherangeofthesmallest temporalwindowfi'omZStoSnsTheexcitationwavelengthusedtomeasmethe lifetime of DMA monomer and DMA-(1,4,6TMN) exeiplex was 308 nm, obtained by doublingtheRGGdyeoutpmofaCoherentModel7020yelaserpumpedbyamode- locked Nd:YAG (Coherent Model 768). The emission was monitored at 390 nm. For highly absorbing samples (> 5% DMA-1.4.6-TMN concentrations), emission was coflectedatthefiontfaceofthecuvettewithamicroscopeobjecfivemaling,25.0522) directed through a polarization scrambler (CV1 DPL-10-450.0-700.00), and focused onto themonochromatorslit. Lifetime measurements of isooctane solutions of DMA and 1,4,6-TMN parallel thereeultsofthemady-statefluorescenceexperiments.Figure24showsatypicalplotof thefifefimeeofmonomaflmrescenceof5%DMAmddreexciplexemissionsasthe concenuafimofl,4,6-Thmisinaeasedma5%molarconcennafion1hroughomme fifefimemeanuemenhDMAmonomerconcennafionwasheldconstantat5%inthe isooctane solutions. The emission decay (I... = 390 nm) exhibited an exponential decrease in the absence of TMN with the value of r= 1.59 us. With increasing 71 concentrations of 1,4,6—TMN, a biexponential decay was observed, with the DMA componemofthehfefimedrorwningaswmddbeexpectedforadymmicquenching process. Thetemporaldecayofthelonglifefimecomponentremainsalmostinvuiant (r= 11.0 t 1.5 as) with increasing 1,4,6-TMN molar concentration, but its relative confir’hfimwmeovaandeayproceamlhesereadmueconsinentwimthe formation ofanexciplex. Since the exeiplex decays by unimolecular dissociation as desm'bedonp.52(aunmmisedinTable3),itsflfefimeshmddbeinvariantwith1.4.6- TMNconcentration. B. flightemperataremeasaremeats Themamintueuindevelopmgthesecxciplexsystemsistovimalisegas-fiquid distribtnions in flow vaporization applications. Since we are especially interested in amomodvegamlineengineemishasledmtobeginmvesfigafingtheflumescence qtmchmgproceaofexciplexsymmsudwuedmmpummA“good”exciplex systematmom—temperanuedoesnotguaramathatitwinbe“good”atelevated temperature environments, e.g., under operating engine conditions. Mon exeiplex emissimsmqmnchedflhightempaahnememthemamauyacfivatedproceescafled “thermalquenching”."‘ FortheDMAsTMNsystemtobeuseful forvisualizinggas- liquid distributions in engines, it must produce a favorable exeiplex emission at high temperatures. llllllllllllrllllllllll'lIl _200 11.1.LL.4L...1...I..V:. -2 0 2 4 6 8 10 12 lltetlma(na) Figure 24. Time-resolved emission fluorescalce spectroscopy of (a) instrument rise-time response, (b) 5% DMA monomer and (c) 5% DMA05% (1.4.61'MN) exeiplex in isooctane solvent. Table 3. Picosecond fluoresceme lifetimes of 5% (wt) DMA and DMA-1,4,6-TMN exeiplex as the wt % of 1,4,6-TMN is varied' ' The solvent was isooctane and the We wn maintained at 28 (i: 0.5)°C. The lifetimes weredetectedstl.=390nm. 73 B—l. Sample preparation and apparatus Temperatlne-dependunfluorescencemeasmementsoverarangefiomZO-ZWC werepcformedwidrmappmahnconsisfingofaPyrexrmmdbonomandlcmqmrtz awene.Thedesignoftheceflpaminedtheexciplexmbeprepuedmderhighvacmm condifimsMebyprecludhgoxygenuaposmhlequmchhgmmceinphMophysical meaauemenuofmommermdexciplexflumeecenceiiqmdsaddedmthemmdbonom wueWydegassedbystandardfieeae-mp-dnwmocedmthemmdbonom waswaulatedviaaKonusqtfick-releaseTeflmvdve.Thefiqlfidsamplemsvaanm mfaredmmecuvenemdisolatedbysealingasecondeteeqmck-mleaseTeflon valvebetweenthetwochambasThemeuepmfionoftheceflcouldbedemchedfiom mermmdbonomsothathcouldeuflybeinnommmahightempaanueappuams showanig.25,whichwasdesignedmpemittempaanuecmnoloftheflmtescence menemningspecnoscopicmeaanemeanhesamplechambawasarectangtdar aliminlmblockcomprisingupperandlowersecfioos.Asquareslotmthelower aluminum block served as a sample holder, which was enclosed by the upper block. Openings on titres sides of the lower block permitted absorption or right-angle flumescencemeeauementsmbeperfmmedihedummumblockwasheatedbytwosets of75Wcmfiidgehedasmdtheirtemperahuewuemonitoredwithathermocouple (Type J,OmegaMode1199).Thealuminum blockswerewellinsulatedwithmachinable sofidahmhrmsificetecaminThewvettecmddbemslatedvancaflybyLScmm posifioneithadrefirpfidorvaporphaseregionsofthemveneondreopficaxis. Temperature-dependent exeiplex emission spectra was measured on either a F4500 Hitachi spectrofluorimeter or a high resolution instrument.” For the latter, the excitation 74 wavelengthwas308nmandemism’onwasrecordedovera300-500nmrangeatascan speedof240nm/minwithaHamamatsttR1104PMl‘astlredetector.Theexcitationand emissionbsndpasswereSorIOnm. a-z. Temperatnre-depsndent fluorescence quenching Exciplex formation between DMA and 1,4,6-TMN was preserved at elevated tempamthigmeZGshowsthetemperannedependentexciplex fluorescencespectra of 5% DMA-5% 1,4,6-TMN sohltions of isooctane at 28°C to 150°C. The significant blue-shift of the fluorescence profile with increasing temperature indicates that exeiplex formafionisdrmnaficallyanenuatedAshifioftheethbrimnfiomexciplexto monomer is expected on the basis of entropic considerations. More quantitatively, if one asameethatmeexciplexemissionintensityispropmfionalmthecmcennafimof exciplmdlendnmwedmmnsityofmeexciplexemissimbmdprovidesthe temperannedependemequflrbfilmconsmmsandhencefieeenergyforformafion ofthe exeiplex relative to K = 70.1 M‘| at 28°C. Theexpression forthetemperature-dependentequih‘brimnconstant 1((T), isgiven by: _ [5‘] __ __ = _ K(T)-[M,][Q]-¢XP[ AG/RTI ex1’14“! TAD/RT] (43) -- Thermocouple (b) Figure 25. The heating chamber for high temperature exeiplex fluorescence measurements llLl 11 1111ui11L1111J111JLL1111 gamma...) §.§§§§§§§ wnvanulnherlto‘ (eef') Figure 26. Fluorescence spectra of isooctane solutions of DMA. 1,4,6-TMN (5% w) at (a) 28°C, (b) 60°C, (c) 100°C md (d) 150°C. The fluorescence spectral intensities are relative to each other. 76 where [E‘] is the molar concentration of the exeiplex-pair molecules in [DMA-1,4,6- TMN]. [M’] and [Q] are the monomer (DMA) and quencher (1,4,6-TMN) molar concenUafiommpecfively.AGistheGrbbsfieeenagy,Acheenthalpy,andASthe entropyofthesystem. . Infanthetempa'annedepmdaaexciplexemismonisdreconm'bufionfiomthe pmeexciplexmdthemonomerflmreecence.1nordermfindmndrepmeexciplex cmmWMonlthemtegrmedemissimmMsityofmeexciplexwasdaaminedfiom aGmnsimbmd—shapemflysisofmespecnaushowanigzalnmisanalysisthe pmeexciplexemissionwasenfirelysepmatedfiomthepmemonomeremission. Two difl‘erent central wavelengths for the Gaussian band shape were employed; 25x10’ cm’I and 27x10’ cm" for the exeiplex and monomer emission. respectively. As temperature increasedmeintegratedmtensityofpmeexciplexdeueasedrelafivemmepme monomeremission.Theinitia1molarconcentrationsofDMAand 1,4,6-TMNusedin the 5% DMAeS% (1,4,6-TMN) exeiplex system were 0.3173 [M] and 0.2258 [M], respectively. Using these values with the room temperature (28°C) equilibrium constant (K=70.1M"),theexciplexconcenn'ationcouldbeobtainedwiththevalue of0.2189 [NammomtempaanneAsstthhemtegramdexciplexmtensityispropmfional mtheconcenuafiondhectly,tharthetempaannedependemethbfiumconstant relativetoK=70.l M"at28°Ccanbeca1culatedandrecordedinTable4asafimction oftempaahae.lnmdamobtlinemhllpy(mlndanropy(AS)ofmesystem,The values of ln(K(T)) were fittedusingEq. (48). Thevan’t Hofl‘ plot of thesedata shown in Fig. 27 yields AH as-16.9 keel/mo! and AS = 47.9 eu.1t should be noted that calculated 77 ‘rrrllrFI—lrthIIIIIIIU 1 [J 1114gLL1 L 1 2.2 2.4 2.5 2.3 3 32 3.4 3.6 10% (11K) Figure 27. Van’t Hofl' plot of equilibrium constant for exeiplex 5% DMA-5% (1,4,6-TMN) fmmafionlheaciplarcmcenuafimswaedauminedfiommemtegmedmtensityofme deconvoltaed exeiplex fluorescence bands of Fig. 26. Table 4. Temperature-dependent equilibrium constant K(T) Temperature (K) Exciplex-pairs (M2) Equilibrium Area of the exeiplex Constant (M') fluorescence(x 10‘) 301.1 0.219 70.1 12.11 333.1 0.177 4.63 9.81 373.1 0.093 0.29 5.14 423.1 0.033 0.02 1.84 78 enthalpymdenuopyvdllesmlmperfimimlheasstmpfimthatdleexciplex flmrescencewupropmfimalmconcennafimmaynmbejusfifiedsincethetempaanue dependenceofnon-ndiafivedecaypednoftheexciplexhasbeenignordm However, thenegafiveennopyofformafionisconsistentwithdimerintionofmonomers Mmeova,meenmalpyofformafimismhnewimmatobservedfmodrerexciplexsof naphthalene (typically between 10-20 kcallmol),"°"' and the entropy of activation. droughlmge.iscmsistentwimthm®savedfmsomemmaficexciplexes."’The smafla'valuehaeprobablyrentltsfiommepresenceofdueemahylgroupsonme naphthflenermgTheeemoiefieeprovideabmfiermexciplexfmmafionbysteficafly hinderingthecofacial arrangement of the DMAand 1,4,6-TMN tr-aromaticrings, which ismeprefamdgeomenyfmexciplexfmmafionTherendmpresmtedhaemdicmesmat metempaannedepeaneasnementsofmeexciplexflumescencemdicatemmme quenching mechanism leading to the formation of the exeiplex is suficiently favorable to permittheexciplextoexistattemperanneeapproaching 150°C. 3.3.Appllcafloaofexciplesflnoreseencetothermometry Theliqtddphasetemperanneinthefirelspraysisoneofthemostimportant factorsafl‘ecfingthespraydevelopmengmixnneformafionandcombustion processes. In agashnbineoramnomofiveenginefilelsprayinjectedintoahotgasannosphae evapmmeebyhemmferfiomdseammdingair.FuelvqofinfionfinVairmixing. ignitiondelay,andthepoeitionoftheignitionstartinfuelsprayafl‘ectthespray combustion characteristicsandtheexhaustemissions significantly, andare {motions of theflquidphasetempaahneinthefinlspraysminordermmderstandthespray 79 combustion process, investigating the temperature distribution of the liquid phase is important. However, non-intrusive and highly space and time-resolved measurements of dleflqlddphasetempaannemfinlsprayshavenotbeenmeaanedfmgasofinebased filel.Themeanuememofthetemperanaewithinadynamicsprayenvironmentis dificluLThermocoupleeprovidepointmeamementsblamaydistonthelocal flows significmdy.Laserfightscemu-mgtechniquenmchasthoseusedmmeforward scattering (Malvern) and phase doppler instruments,"' have provided significant information on rh'oplet number densities, size distributions, and velocity distributions, but have been virtually insensitive to droplet temperature. Recent modifications to the phase doppler techniques (“rainbow scattering") may lead to temperature information,‘22 but these techniques will remain ftmdamentally point or line-of-sight techniques. Fluorescencemethodsasdescrlhedinthissection. ofi‘erthepossibilityofminimally perturbing, real-time, two-dimensionalmeasnementsoftemperatme within dropletsor spraysAsigmficamweahnessofmeflumescencemedrodisdieirsensifivitymthe environment, particularly to quenching strongly by oxygen in the combustion atmosphere. However, as long as a pre-combustion environment is considered. the fluorescarcemethodcmbeusedasa“mermometer”mthetemperanueflow-fields. 3.3.1 Overviewofexclplex thermometry The applications of exeiplex fluorescence to thermometry are well extensively reviewed by Melton.’ Although the photophysical processes lmderlying the commonly observedshifiofmeexciplexemissimbmdmdlorterwavelengthsasmemmpamme is increasedare notwell understood,"‘thetemperatmedependence ofthe fluorescence 80 quanttunyieldcanbeusedforexciplex thermometry. Zhang'23 andAlexander developed mdlmedafllnxucemthamometerwhichexploitedmedecreasemthefllmrescence quanhmyieldothodamineBwidrincreesingtemperanaeoverthenngeZWCto 185°C.Melmnmdcoworkas'”developedaseriesofflumescemmamomeuysystems basedmmephomphysiceofeacipkxenpafiuilmlythosefmwhichmevemissimis significaflymd-dfifiedwifimspeflbtheM‘emisfimmmeseexciplexflmreeccnce mamomeUymmemonomerM°mdtheexciplebeommmrescemdsincethe mfioofM‘cmcennafimmE’concenuafionismpfllnn-dependcm’eifllermmugh meviscodtyofmemlvunaoledmficfimitxmmroughmeeqmlibfium cmmnghmpamnemamodynmichmit),therafioofmeimwsityofemission fiomE‘tothatfiomM’canbecah‘bratedasafluoreecencethermometer.Smfil '2‘ confimedthanasprediaedfmhdknon-evapmafingfiqmdsitwasindeedposmblem usemexciplexflmrescencethamomaabasedmtheintermoleculmexcimerfmmed betmenexcitedstatepyreneandgrmmdstatepyreneindecanetomeannethe temperature withint1°C overtherange 25°C to 91°C. Theblue-shifi ofthe exeiplex flmrescmcespecuacmddflsobeexploitedmdevelopexciplexshifimamometas.’ Thisshifiumeanredbytherafioofthemmmitymachosmspecnalbandpassonme shorterwavelengthsideofdieexciplexemissionbmdmdntmachosenspecual budpassmthelmgwavekngthddecmabobemtphiteduadlamommr335chnm aal.'”havedsodevelopcdaflumeecenceshifidrumometawhichmakceuseofthe blue shift in the fluorescence of BTBP [MN-bis (2,5—di-tert-butylphenyl)-3,4,9, 10- paylenedicuboximide]inmahnolmminaaloflumetempuanueincreases(15°C- 70°C). They achieved temperature resolution of 1-2°C. 81 3.3.2. DMA-TMN exeiplex thermometry Analysis of the temperature sensitivity of the exeiplex emission involves taking theratiooflell..."mlhispm'amemrcmlbeassignedinseveralways.Onecanratiothe inmnshyoftheexcipMpeakmthatofmemmomermrrafiomeintensifieeinaband aromdthesepeaksoalthoughdledemucefimisarbiuary,mismaybedonequite acmnatelybypcrfmmmgGmmfmfimmmeflmreecencespecnaoveracaminwindow abommemaximmmdevelopsomemfirelydifl‘aentmmpaannechmactaisficmhas thecorrelationof theshifiinthefluoreecencemaximum(1_)withtemperantre.1n Section 3.2.2. temperature-dependence of the exeiplex fluorescence quenching of 5% DMA-5%1,4,6-TMN system was discussed in detail. As temperature increased, the flumescencemmnsitydecreasedgreadymdmeovaaflemissimswaeblm-shifiedas shown in Fig. 28.1n terms ofGaussian band shape analysis, both ofthe intensities (i.e., shorter wavelength side (In) and longer wavelength side (15)) within the exeiplex fluorescence intensitywereobtainedandrecordedinTableS.Theratiosofthese intasifieecmddbemedasmexciplexthamometawithtempmhuemgefiomZPC m150°C.Asincrasingtempaannemechmgeofmeflumescmcerafiofln[IE/lu])was greatlydecrasedThemhdfineismebestfitwithmexponendaldecayequafiomy=a ap(-bx) + c with the fitting constants, a=l6.6l, b=0.024 and c=-0.24, respectively. Figure 29 shows the calibration result of 5% DMA-5% (1,4,6-TMN) exeiplex thermometryin90%isooctane. 82 41o r—r'rIrrjrrT'lTjfir'firTfiv-Trfirj :r-i 4 £395 E- .3 E (I) 3 E 390;- : _ 335 :- a; E— a» ‘3: 375 5— _. _, 370 Lu ”duhiumwsludlu: 20 40 so so 100 120 140 160 Mare) Figure28. 1hechmgeofwnlflmrescenceintmsity(a)mdcmrespmdingwavdengthchmge atpeeltintensity(b)inFig.26withrespecttotemperanue1'hetotalintensitiesarerelativeto eachother. 16 *T Fwfii ‘ f‘ r '4 T 7 ‘ T r' '1 ~ 14:— “j 12 E -3 10 :— ‘3 ‘ "' :1 .3. 8E- '1 : j 4 _— 1 2 E— '3 : i 0 s 0 50 100 150 200 MW) Figure29. Thefluorescence intensityratioflnfl/Lflofexciplextothemonomerwithrespectto inausingtempaanneTheflumescmceintmsideswaecalqduedbuedeansimbmd shapeanalysisThesolidlineisadIafitequation; ln(l.ll.)= aexp(-bT)+c,witha=l6.6, H.024 ltd c-—0.2421, respectively. 83 Table5. PmemonomerGdeexciplexfluoreecenceOE) intensityandtheirratiosas afimcfionoftempaflme.Bo¢ofthemtensifieewereobminedbyusmgdreGaussian bandshapeanalysis. Temperature Areaoftheexciplex Areaofthemonomer (K) fluorescence (x10‘) fluorescence (x10‘) lane/1...] rid [Id 301.1 12.1 1 0.0032 8.24 333.1 9.81 0.29 3.53 373.1 5.14 1.20 1.45 423.1 1.84 1.68 0.095 Chapter IV. Experimental Methods (ll): Physical Characterization of the Exciplex 4.1.Chapnr0verview Thischapterpresentsthedetaflsofhowasampleisprepmedandcah'brated quanfitmivelyforbothoffiqlddandvaporphasefinl.1tbegimbydescnbingdre phomphysicsofLuerlndlnedExciplexFlumescencefllfinvinmlizafimwhose flmreecenseewueemployedmmecah'hafimFmdmennlopficsisdumbedmexplam howtheimageisacquiredandproccssedThemoasignificmtsecfionisdevmedto descnbmgmehmddmdvapmphasecahbrafimproceamhasmlafingflumescence intensitytotheconcentrationofliquidandvapor, particularly emphasizingtwodifi'erent procedneebawemdhea-mdmdheapressnemasnementmmevapmphasecahbranon InthnSecfiommecah'brafimreadtsfmbommmdmdvapmphmmdiswssedmdemil amimeaudddifi‘mbenveendirea-mdmdheapresunemeammentfmvapor phaseconcenuafionareemphasizedThefinalsecfionofmischaptadesmheethe expahnemalmemodsfmacnmlmaauemmtsoffielLlEFimagingmanumzcylmda andinamotoredengine. 4.1.1.8scbiexe-imionialqnidandvapor phases Theabsmpfimandflmreecencespecuaofmmolemdesdissolvedinmn-polar solventnwhutypicalfinlsrevirnnflyidenficslmmespecuaofmemmemolemdesm thevaporphase.However,insomecasesitispossl‘bletoreactthefluorescingexcitedstate moleadeaM°,wimmappropfiategrmmdstatepmma,Q,mfmmasecondflmrescing species, M-Q‘, called an exeiplex (E‘). The common mechanism for exeiplex formation is 85 the reversible equilibrium“ which governs the concentration of the exeiplex complex. A genealschemefordleovaaflexciplexprocessisshownmSchemelmSecfim322. lnthehqtddphasedleethbriummaybeshifledfartothefightbyadjusfingthe concennafionoandQ.Uponvapodnfion,theconcennafionomedefllberemwed sommM°cmisdmwifldommneMoreovendleemdhbdtmistemperanaedependenn shifiingmthebfiutempamaemmAccmdingly,meexciplexislessaableat elevamdmmpaanmmdmeexciplexflmrescmceemissimisanenuaedwime thatfmthemmoma.meesereasms,inmeLasaMlcedExciplexFluoreecence techmMmevapmphaseflumucenceisdommamdbyM°mdmehmudphaseemissionis characterized by the red-shiftedemission of the exciplex.”"°‘ Because the exeiplex forming dopants,MandQ,aretypicallydissolvedinanon-fluorescingsolventwhichservesasafirel. theflqtfidpbasefinlisdisfingtushedbythedisfinctemismmfiomtheexciplex, E°,and emissionfi'omthemonomaM‘wiflmcktheevaporafingfilel.Figme30showsthe fluorescence emissions fiom the vapor and liquidphase using a 5%DMA05%TMN exeiplex synnninfimsolmlmmmduquidphmammdomodby the monomer (DMA) and DMAOTMN exeiplex pairs, respectively. With the 308 nm exciufimthemaximrmmmnsitywavelengmsforthevapmmdfiqlfidm344(nm)and 408 (nm), respectively. ~ . a i . 1 - .. i 0-3 :' 2‘ 0-3 r '2 I 1 0'4 t' 1 : 1 0.2 __ j - 4 O P; a 1 1 4L1 L_t 4__l____ - i_. 300 350 400 450 500 Embalonwavelengthoun) Figure 30. Vapor and liquid phue fluorescence emissions (A: vapor phue, B: liquid phase). 5%DMA05%1,4,6-TMN exeiplex system in 90% isooctane solvent was used. 87 4.2. Optical setup for two phase fluorescence intensity calibration 4.2.1.6eneralfleld-optics Agencralschemaficdiagramofacah’brafionexpahnemisshowninFig3l.The senmtmlizedmeampksymexcimalasa,hsaenagymeter,fightshcetfmming opficnmtmsifiedCCDdeteaor,mddetecmrconuofla.Theexchnahsamsuscdasme excimfimsmuceopaafingatawavelengthof308nm(XcClfine)wimamaximummwer oflSOmleulseatamaximmrepetitionrateonSOHz.Lightfi'ommeexcimerlaserpassed through a custom~made variable beam-attenumor with removable quartz discs for adjusting thepulseenergytoadeeiredleveLandafieldstoptocutofithefi-ingee.Itthenpassed through the beam-splitter for the energy meter, and a field-optics consisting of fused-silica cylindrical focusing lens with a 300 mm focal length and a cylindrical recollimating lens with a-37.5mmfocallargthtogeneratea800micronsthick(FWI-N)and20.6mmhighvertical sheetmdfinaflydlesamplemmesamplesystem.Theflumescencefiomthevapor(or liquid)wasimagedbyaquartzcameralens(NikkorUVlens, 50mm,f/11)ontoadigital computer-controlled image acquisition system (Princeton lnstnunents ICCD-576x384) eqdppedwifiagamdimagemtmsifialmmehsu-MwedexciplexflumescmceaIEF) imaginghsumdcamaagfleueaflelecfiodcaflysynchromzedmacqlnremeimage. FlmreecentimagesnkcnuemferredmcomplnermemorymdwfinentodiskTwo bud-pesfilfiswueemployedhflsshflxfmmevapmphanabmd-passfiltermw LaserModelF35-355-4),centeredat349.1nmwith37.8anW1-1M,wasusedandforthe liquid phue, a long-pass filter (Corion LL400-S) was employed. The relative transmittance ofthcse beam protector 3.13 sample ICCD ”Ct: detector . c: :> L2 field-optics c: := Ll detector power meter - controller attenuator exeimer Pulse laser (X eCl) . , ggerator computer Figure 31(a). Schematic diagam of calibration experiment of filel vapor and liquid droplet 1 H i:- L)1(fl [1(6) Figure3l(b). Schematicdisgrmnoffieldopticetomakeasheetoflight.(a)and(b)areinitia1and finalbeamwidths, 6.6mmand800pm,respectively. L.md14arefilsedsiliceconvexandconcave lens with focal lclgthf, 8300mm); =-37.5 ms, respectively. 89 1m11.fi—'Y7 ff‘rlr '7] F J sob -: Z 3‘ - I J 60— " J I A [a 3 l u— ’ -o 40— i .. l- - i a a - ,’ J . i .1 20" .t .. - / J o" .. ”111..-“; 300 350 400 450 500 Figure 32. Bandpass filter % - transmittance (A: for vapor phase. B: for liquid phase). 90 bandpassfiltasueshownhFigHJheflumeecencehbnsityMwasexnamedfiomme imagesmdmrmdiaedwimmelmpulsehnensitymdMgainmdfinaflycmnlated tothespecieeconccntration. 4.2.2.!‘1eldofviewof1-age WhenmedfoerimensiomlimagingapphcafimsPrmcetonlnstrmnentsICCD camerasmedeeignedtoopficaflymplicateastandard35mmSLRcameraThemain difi’erencebetwunthetwo,however,isthesiseofthecollectedimage-orthefieldofview atagivendimBeforecalculafingthefieldofviewatagivcndistance,itisusefultofirst Figure 33. [map'ng Field of View 121 91 i D:distancebetween,imagemdfocalphncofcamera(Focalphneofcamaais46.5 mm backfiomfiomdgeoflanadapta),B:dinancebawemobjecfivelemmdfocalphneF :focallengthofthelensused(50mm),S : lCCDdiagonaldimension,O : fieldofview coveredatadistanceD(orportionofobjectthatisimaged). . Thenthefieldofviewcanbecalculatedae FD where M=-——-2- (0-5) s 0‘14 Asanexample, formlCCDcemerawitha‘Ihomson7883,eachpixelhasaspatial reeolutionof23umx23umwith576x384arrays.1herefore,thediagona1dimensionof thefocalplaneof the ICCD camerais 15.9 mm. lnthevapor-liquid calibration experiment. thedistancebetweenimageandobjecfivelenswasset1127mm,and50mmNfltkor objectivelensona18mmextensiontubewasemployed. Withdreseparametel‘s. the magnification(M) ofthe objective image andthe field ofview(O) can becalculated as M: 50‘(127+18+465) . 0593 1272 15.9 o - 0593 = 26.810mn) Byanalogy.dichorizontalandlongitudinalfieldofviewoftheobjectiveimagecanbe calculated as 14.84 mm and 22.34 m, respectively. Therefore, a callhration image area 92 (liquiddropletorvapor)approximately39ttmwidex39umhighwasfocusedontoasingle pixel(23pmx23um)ofthedetector. 4.3.1ateneity-concentrationealibration 43.1.Vaporphasa Thevaporphasecall‘brationwascarriedoutintwostepsusingasampleofN,N- dime&ylmfline(DMA),thevapmphasemarka.Thefirststepwumdetaminemevapm phaseconcaluafimvemmtempaanneofthesmnple.Thesecondaepwasmdetemme flumescmcemtensityvmsvapmtempaanne.Usingmeconcmnafim-temperanne mlafimshipfiommefirststep,memlafimshipbawemtheflmrescencemtensitymdme vaporconcentrationofthesamplecouldbeobtained.Therearetwomethodstoobtainthe vaporphaseconcenuafionOnemethodistheabsmpfionmeasmementnlggeuedby Rottmno et al.‘2 For example, in order to calculate the vapor concentration of TMPD, Melton meannedabsorbmcesofTMPDasafimcfionoftempmanuesandemployedtheBeer- Lambatuwwnaehernedtnemolulbeorpnoneoemnentnnimpneiredirom Bulmm“1heo¢amethodproposedhaeisfiedirefl-preesuemeanuementexphhedm SecfionB.Thissecfimismamlydevmedindesmhmgmemam&fierencesmthevapor phasecalibrationprocedmesbetweenthetwommods. 93 A. Absorbance Measurement (Indirect pressure measurement) a. Samplepreparationandapparatu Forthevaporphasecahhrafionasmallsampleofoxym-fiee. pure DMAwas prepacdushgasmndndfieeu-ptmp-mawmmdmsatedmmeqtmflmfimeter ceflwithastemanlemmlightpnthlalgth,pmchasedfiomNSGPrecisionCells Before placingthesmnplehtheWanabsmbancemeauaememwasmadeontheempty menewimmespecnophmomaer.Thiswastheblmkngnalmatwasusedmmbuactme efl‘ectofthecuvettewallsfi'omthesubsequentsamplemeasmements. Temperature-dependent fluorescence measurements over a range from 30 - 200 °C, wereperformedwithanapparatusconsistingofal’yrexrolmdbottomanda 1 cmquartz cuvette.‘lhedesignofthecellpermittedtheliquidsempletobepreparedlmderhighvacmun conditions, thereby excluding oxygen as a possible quenching source Liquid samples were addedmmermmdbonomusmgamiaohmrsyfingefinedwimalomchhypodermicneedle sothatthesamplecouldbeplacedonthemtmdbottom.keepingthewallsopticallyclear. OxygmconuinedmmeDMAsolufionwasmmoughlydegassedsevaalfimesbysmndmd fieeae-pmnp-thawprocedtuestomakesurethelewasnomoreoxygenthan 10" Torr. The DMAsolutionwasfi'oaenwithliquidnitrogen,evacuatedto 10‘Totr,andthawedwith warm tap-water. Approximately 200 microliters of DMA liquid sample were then vacuum umsfamdmmewvenemdisolmedbymasecmdxontesqluck-mleasemflmvalve betwemthetwochmnbersTheawenepmfionoftheceucmddbedaachedfiomthemund bonomsomnitcmddusilybemnodmedinmahightempaanneappmamsdesignedto 94 permit temperature control of the fluorescence cuvette during spectroscopic measurements. ThesampleheafingchamberisshowninFig.25andmcn’bedinSecfion3.2.2. b. Calibrationpmesdure Thefirststepwastodeterminedrevaporconcennafionversustempemnuefi-om absorbance measurements with a UV-VIS spectrophotometer (CARY 1E Varian Model). maeesingmewmpaameofmeamplemcreasesmevapmcmcmnafimbymcreasingthe samration vapor pressure. The temperaturecontrolled sample heating system was mounted in mesamplecompmnnemofthespecnophommeterfordlisptuposeasshowninFig.34. Measlnementswelemadeata10°Cintervalsfiomroomtemperannet0200°C Thespecn'ophotometerwasaflowedtowmmupapproximatelyonehomandaemed withsamplechamberwindowsoftheheatingsysteminthesamplebeam.’lhisremovedthe efl'eaofthewmdowsfiomthemaanemenm.Themfuencebmmwasempty,i.e.airwas themfmiheamplewasthenplacedmmesamplechamberoftheheaterand absmbmcemeannemeanaeukenafierthetempaanneemdhhratedmeachmterval. Afiersubtractingtheblank signal. concentration was calculated usingthe Beer-Lambert’s law A(T)=st(TJ whaeAistheabsorbances isthemolarabsorptioncoeficient,bisthepathlength(cm), andCisthetmknownconcentrationofDMA vapor. 95 Figure 34. DMA sample absorbance measurement instrument using a temperaturecontrolled sample beefing system with spectrophotometer. 96 lhesecondstepmmevapmphmecah'brafimwumdaammeflmrescencemmnsityvenus vapmconcentafimusingmevacmmsuledDMAmmple.Againtemperanaewasusedas aspecificparamctertovuythevaporconcentrafionofDMA. Thereforeatagiven temperanaeonepairofdamflhloofDMAflumescenceemissimintensitym)and WadcmddbeobminedBycmrehfingthepahsofdamflammthedammus of(T,,C,),whichcouldbeinferredfi'omthefirstcalibrationstep,thedatapairsof fluorescmcemtensityvmcmcennafimahcomnaspecifictempaatmT=Thcan thenbedeveloped. c. lmageacquisitionandanalysis lmagaofthefluorescenceweretakenperpendiculartothelasersheetat approximately 10° Cincrements withtheintensifiedCCD detector.Abandpass filter (CV1 LaserCo. Model F35-3554), centeredat 349.1 nm with 37.8 nm FWHM. was employed to coneaonlyvapmphaseflumeecence.Agencralopficesenmmacqlfirevaporphase flumsscenceisshownhFig35Jhesystemwusymhrofizedbyextmflyniggefingme ICCDdetecmrfiomtheexcimalaserthroughtheFGolOOpulsegeneratorandtheST-130’ demamconnofla.Swaalimageswerenkmflenchcommnafimlevelusing10expoaues perreadanimmckpmmdalbcacfiommdstoredmfllecomptaeniheavaagepMse energyofthe 10pulsesmostneerlyassociatedwitheachimagewasrecordedforimage proceuhglaterflhedeteflmhudwmpmmflasmeadjustedbymn-nunpomnfiomm bmuenmfinemwimmenanmitsmmedialface.Cah’brafimsfmmesepotenfiometers, alongwithotheropaafimflmfomafimmmsisttheuser,meconminedmmeAppendix. 97 Figure 35. Experimental set-up for determining fluorescence intensity of DMA vapor phase (sample heating system with insulation removed). 98 HumescmcemtmsityMWeremwdfiomtheimageemingmeCSMAmfime" A rectangularregionofimereetmOl) containingthecuvetteimage(excepttheedges)was definedmdtheavuagehnensitypapixelwucdallmedlhemgimofmmreuremaimd mesamefwaflimageaTheavuagemtanifieewerenmmalizedbythepldseenergy, cmrectedfmdetecmrgammdcouecmdfmtheanmuafimefl’ecmfimequmtzwindows mdmrvehewaflthaflthehflnsifieswereplohedrendfinginacah’brafimmeof flmreecmcemtensityvmmassconcenuafimofDMAvapm.Theconcennafimoffiel vaporwastheninferredbuedonthemixnneratioofDMAtofuel. B. DirectPresenreMeaeurement a. PreeeureSensorCahhratiou TheMPXZlOO-D series deviceisasilicon pieeo-resistivepressuresensorproviding ahighlymmmdfinecvdmgeompm-direalypmpmfionflmtheappfledpremue. Thesensmisasinglemmohthicnhcondiaphmgmwithmesuaingaugemdamin-film resistmnawmkmtegraedm-chip.1hissensmisnmpamaemmpmsmedova0°Cm85° CandabletomeasmepresaueoverOtolOOOton.Priortouseinthevaporphase cah’brafimexperhncntthesensorwuceh‘brfledusingavacmmptanpandaBmanm capacitance manometer (MKS lnstmments Model 14581-184000) with a 0-1000 ton- meaauemmtmge.Figme36showsaschcmaficdiagamofasensmcah‘brafim.Avammm pumpwasusedmdeaeuetheahpresnnemthesensorataIOOtorrmtervaLmdthe mmwedahpresnueasweflassemmolnptavomweremeaanedsimultanemulybya Baratron manometer and multimeter, Vacuum ump Valve 1 Pressure sensor - sealed with high temperature epoxy Valve 2 um] Baratron capacitance manometer Figure 36. A schematic diagrun of MPX2100-D pressure sensorcslibrmion experiment setup. 100 w ,..J TTITYIIIII‘U‘TITIrrrrrTrIeflTIIWfir-r I+ SensorOutputtth) 0 ..ulurruLuus.l.iunL-.urLuulmus.1. 5,: 0 100 200 300 400 500 600 700 800 ApplisdPtsanuruaon') Figure 37. MPX-ZlOOD presstne sensor calibration result. Solid line is the least squares fit to the eqmtion y a -0.0524x + 39.6, max. deviation of 0.343. 101 respccfively.Atthelowestpresnne,thescnsorwasaflowedtoleak.therebyincreasingthe presmueupto70070rr.Figla'e37showsacah'brationofthesensor.lnaseriesof cah'brafimathethmwuemdterepeuablemdshowedafinarmlafimshipbetweenme sensoroutputvoltageandeppliedpreauu‘e.‘1helinearregression fittinggives y=-0.0524x+39.6 with maximum deviation of0.343 and was used in the vapor phase cah’brationtoobtaintheDMAvaporprasure. b. Experimentalapparatusforpreseureseusorcalibration Tbefirstprocedurewastomeasurevaporprcsmredirecflywidlreepectto temperature. For this measurement, the high temperature sample cuvette was extensively modifiedasshowninFig.38,mmeasluetheDMAvaporpressmeandtemperamre simultaneously.TwoPyrextubee(¢=1l8")wereattachedtothemainstemtoallowthe insafionofsminlees-steelmbmg(¢=lll6”)thatwasconnectedextanaflymthe pressure sensmmdmamomwlaTheKomjomawhichiscomposedonyrex(¢=1/8')mda metaltube(¢==l/l6”),wasinsertedbetweentwoPyrextubesandasensor,andthensilver solderingwasprovidedtomakesurethatitwasleak-tight.Al-typethamocouplewhichisa phlg-typecmfigmmimsensifivemmedpmddiamedwimmmlessesteelmprotectme sensorpmtwasmedbmeanuethethempaanueinsidecuvenemdcmctedgauge mdicator2.AMPX-2100Dprmuesensmwuemployedmmismmmment1hissensm hadavacuumsideportands1/l6"-stainleessteeltubingwasinsertedintoitandsealedwith hightempaahneepoxy.1hetempaahneofthenmpleheefingchambercolddmenbe increasedupto200°Candbewell- Vacuum Pump Quartz Cuvette é» Liquid Sample 102 I f j— Heating Chamber ‘— Insulator Figure 38. Vapor phase calibration instrumentation Pressure Sensor Temperature Sensor 103 msulamdwimmeduminumsihcuecaamialhemdmmhfimwasneceesarymavoidthe evaporuedDMAvaporcondensafimflmgthesminlemaeelmbmgwhichwasmthemom tempaanuecnvummentlherefmeitwaswrappedwithheafingmpemdmainminedat consentemperanue(150°C)Mgmemuanemem.Thistemperanaewasconstmfly mmhmedwithdlamocmmlegaugeindicatml.Wimtheaidofmemstnments,meDMA vaporpresslaeandtemperatmeweremeesmedsimultmleously. c. Calibratiouproeedures Thevaporpreesnewasmeusmedat10°Cintervalsfiommomtemperanaet0210° C.OnceaP-Tphasediagrmwuenabhshed.themolmconcennafion(n/V)couldbe determinedusingtheequationofstateofanideelgas(PV=nR7).PandVaregasvapor pressureandsamplecuvettevolume,»isthenumbcrofmoleeofthesample,andRandTare gasconstant(~ 8.311/molelo “temperature, respectively. Thisdirectpressure method dhecflydepmdsonmeequafionofmtefmanidealgasdhaemetwomaincfiteria'” for theapplicabilityoftheidealgaslaw.First. ifthevaporpresslueflefismuchsmallerthan thecriticalpressureP,i.e,P,< Controller I ‘_ Tontput variable I I Pulse Invert .4 t. Fuel Inlection | System I l ‘ to | Fuel Injector l Oscilloscope ‘ I Sync. _--_.____________ __’ Field Optics ICCD Gate Camera A Detector Figure 57. A sequential schematic diagram of fuel LIF image acquisition experimental setup. [29 acqldsifiomwhaammmexpomefimeimagepixelsizemdnumbaofuquinngimage fiamesynchroninfimwithexcimerhser,wereprevimlymemodzedbytheimage acqldsifimmfiwue(CSMA)mmewmpruuhmddiskmroughtheFG-100prdsegenaator and ST 130 controller settings. Fuel injection systems for 2.4 L direct injection gasoline mgimwueprofidedbyChryslafmmesmdyoffiqlfidmdvaporphsefinldismbufions hintanflcomhsfimchambasThissystemconsisedoffirelmjeaaanddehygenenmr box, which provided fuel injection signal out (to), delayed signal out (n), filel injection duration. and fuel injection fi'equency. Except for the fuel injection signal out, all output signalscwldbeadjufledwifialO-nunpounfiomaa,thaebymemquhedexpefimenml parameterscmrldbeeasilysetup.ATektronicslOOMszigitaloscilloscopewasusedto monitorthedelayedsignalout. Thespecnophommeuicgradeimomesolvemistypicalofagamlinefirel.Pme isooctanesolventdoeanotabsorbUV-VlSlight,therefore,forthefuelLII-‘imagingaseedis necessu'ymprovidethefllmrescenceemissionwiththeUV-laserexcitafion. Forthe fuel LlFimegingNfi-dhnahylmnioemm)wasmedasmemeldopehtrhaeuem reasonsforthischoice.FhsttheDMAhasahighquantmneficiencpraOJl),therefore, immlafivelyhighflmcenceemisdmmtmsitymkeshusymfianliufinldismMm SeconiweahadydevclopedfimhnfimsystemsbasedeMAdeMN-isomaswith misoocnne-buedfirel.1hisprefiminnyexpefimemwuthefiraaepmchecktheLlF imaging acquisition and the validity to use the DMAOTMN-iaooctane exeiplex system for twophasefuelvisualintions. Forquantitativeapplicationot‘thefuelvaporandliquid visualization using a LIEF txhnique, 5%DMA05%1,4,6-TMN exciplex system was employed in an isooctane solvent. A Nitrogen environment was provided inside the quartz [30 cyfindermprmeaagainsflmrescmceqmnchingmsmoxygendmingmeexpaimentme finlmjecmrwucmfinumnlyfireduarepefifimmeofwflzmdprommmmdseszone forfirelinjecfion(t.)andtheothaforhnageacquisifionatafimedelay(t¢)withrespectto thefilelinjectionfimeFuelinjectiondurationwassetto4.0ms.Theexcimerlaseris externallysynchrnniaedwiththedelayedsignaloutaoofthefirel injector. Forthe experimenupaformedudthaconfimmuswavelasa,thegatedrnafimconuofledthe exposlnefimeGatedurationwasvu'iedbetweenZJusandll pal-lowever,forthe measuementusingtheprflsedexchnahsu,%al8mpldsemnfimmdapproximmely fewmflmrescentfifefimemememelybriefexpounefimewhichismdepmdentofme camuagnemnfimwasachieved.Withduscmdifimmeacquhedimagssm“fiozen”. SincetheexpostnetimeoftheICCDdetectorwassetasl.0secduringthemeasurements, eachacqlfiredhnagefiameisexpmedmIOlasushmmeissnldy,wesamedehyed signalout(t.)fi'om0to5.04msasanincrementof0.56msor0.llmstomimic600/lSOO rpmengine cycleasan2°ll° crank-angledegreeinterval. respectively. A detailed schematic diagram ofthe fuel LIP imaging field-optics setup is shown in Fig.58.Aprdsedexcimerlaser,308nm(XeCl),istheexcitafionsomce,withamaximum poweroflSOmleulseatamaximumrepetitionrateonSOHz.Thelaserbeamisformedby twopairsofcylhdricalfinedsificalenmmmasheaoffightapproximatelymumthick by50mmhigh.direcmdmmequmucylmda.Tominimiaereflecfimlosseganmim mdlenseswuemfi-mflecfim(AR)m¢e¢Thesheaoffightpassedthroughmecenterof mefirelmjectmwhichwaslocaednthecemuofcyhnderaxisThefinlfllmrescence emissionwasimagedbyaquartzcameralensmikltorUVlens,50mm,t711)ontoanlCCD camera, which is a digital. computer-controlled image acquisition system (Princeton 131 Excimer Laser 308 nm 17 ns pulse width 5-10 pulse/s Ml AN M2 Entrance Stop I: :3 Power Meter :1 e E Ll ICCD camera 576x384 pixel c: 3 L2 B.P Filter -— L4 L3 ¢ [X A A / M U V U Quartz Cylinder with Fuel Injector Figure 58. Schematic diagrun of Mel LIF imaging field-optics. including fuel injector and optically accessiblequartzcylinder. Tbefinal besmsteeringmirroralfl) movesthelasersheetthroughthe quartzcylinder. I32 mauumenm)equippedwithagnedimagemtensifia.ThePfincamharumenmumaahas ahighdymmicmge(lfi-bitdigifiz¢imxlowmise(cmleddeteammminimizedark mummmmmmhmmnmumyammm). high qlnntum eficiency (212% at 300 nm), adjustable gain (51 to [00 mum/Wm).mdadjundueehcnonicgafing(3.5nsm80ms).0nediadvmmge ofmecamaasywmisimrelafivelymodeuspafiflresolufiomwhichisfimiwdbyme S76x384-pixeldetectorandfirrthcrreducedbyafactorof20r4bythefiber-optic-coupled mansifia.Anothadisadvanmgcisthelmgimagereadomfime(aboisfmmmbinned filflfimexsommorethmoneimagecmbeobnmedpainjecfioml'heCSMAsoflware can produce mono clu'ome (gray-scale) images. However. false-color processing (assigning a colormcatespondmmmmfitymgefisofienusedforhnageenhmcementlnom appficafiomeachpixelintheimagecorrespondstoabinbyz.Forin-cylinderfuelLIF imagingthefinlhjedor,hsamdcmaagatewereaflelecfiodcaflysynchronizedto acqlnrethehnageuaspecifieddehyfimeafiafirelinjecfiomThefirelinjeaorwas typiullyfiredmnfimnulywaO-Mncondsbefmemfingdmmsifiommdtypicdly eitherfivelO-luer-shfiavaagedimagesorupmsomdividmlimageswaemkem mafaredmcomplnermemmy,mdwiuenmdiskWidrdrisconfigmafion.onlyone imagecmddbenkenpufinlmjecfimsmceoxygenisagoodflmrescenceqmnchafor mmafichyrhncubmtheexperimentsweredoneinanitrogenenvironment 133 4.4.2. ln-cyllnder fuel LIEF imaging In a motored-engine A. Englaeasseatbly&opticalcllamher Theengineassanblecmsinsofasinglecymmotmedenginewithmelongated pistonconuiningaqummstonaownOnmpofthisismoumedaZAHter4cylmder directmjectedCln'yslerheadwhereonlymecombusfimchamberwasbeingused A sepuuehnakesynemwasdedpedmdehvaaprehamdatomizedmixmmtheengine forthecalibration. Alltestsusingthisexperimentalsetupwerenmatamotoredspeedof 600rpmwithawideopenthrottle(W.O.T.).‘IbebaseengineisanAVL 530 diesel uankcasemdcylinder.1bishasbeenmfiaedwitha100mmsuokecmkshafiwhich machesmesuokeofmecyhdahadmusedfleadvmhgeofusingmisengineismm itprovidesabdmcedefingmdreciprocafingmamThiscmkcaseisdmfiuedwimm addifionalflywheelwhichaddsaddifionalinafiamkeepaconsmmwmforasingle cyfinder.1hecyfinderheadisa2.4literdoubleoverheadcam€hryslerdirectinjection prmmypeJtcmmimfomvdvssmospukplmmdacenuaflylocuedhighpresnne prototypeinjectorasshowninFig59. InthisexpaimargonlythenumbertwocombusfionchamberwasusedAnopfical combustion chamber inchrdingChryslerenginehesd.quartzcylinder,elongatedpistonand ICCDdetectorisshowninFig.60.Thelifiersofallothercylindershavebeenremovedsoas onlymacnmethemquiredvdvesmcylmarhadwummmedabovetheAVLengine mmwmdammcyhndumdneelspacawuphcedmbameheboreand stroke of the combustion chamber are 3.444”, and 3.937”, respectively. This gives a single cylinderdisplacementof0.8litersandacompressionratioof9.4tol.Thequartzcylinder wasmadeofGE124fusedquartz.whichhasatransmissionof92%intherangeofUVlight Figure 59(a). The Chrysler 2.4 L double overhead cam direct-injection engine head. @000 MO. Figure 59(b). A schematic diagram of the 2.4L Chrysler engine head showing two intake (A). and exhaust (B) valves. Fuel injector is located in the center (F). and two spark plugs are situated in the left and right corners (S). I35 I 2' I . ! Figure60. lmsgeacquisitionexperimentalsetup foractualmessurementsinrealengineoperation. Chrysler engine head, field - optics (left side). [CCD detector. optical quartz chamber including mirrorandpismandAVLengineareshown. 136 thatisused.1'hecylinderwaslOmmthichwhichprovidedenoughstrengthwithoutlosing laserpower.Thepistmenensiomwhichisahollowtubewithaslotcutinonesidefor accessofastationarymhror,wasbolteddhecdytothetopoftheAVLpiston.Thiswas machinedfrombillet70705‘l‘651aluminmforahighstrengthtoweightratio.‘l‘hepiston extasimconhhsaquutpifimaownmdkrdm’fiinhismaQialisaflmpolymer withalowcoeficientoffriction.Italsohasatemperaturerangeof(—400Fto+500F).These propafiesmkeawefl-nutedsedingmataiflfmtheqMcyfinda.Theqmpistoois alsomadeofGE 124filsedquartzandisl.78cmthick.Thisquartzpistoncrownmakssup about65%ofthepistontop.Thismaltesitpossibleforlargeportionsofthecombustion chambertobeviewedatfimeswhenthepistonisclosetoTopDeadCenterCTDC). Thisis themaxhnummncmbeusedbecmueofmeneedfmpistonnnglmdsfihemhrmthatwas usedmgamaccessmthecombusfioncambaisafirstnufacemhrorwithUVenhanced allmmumcoafingforgoodUthluUmmissiomwhichwaspluchasedfiomEdmrmd Scientific.1‘hemirrorisanellipsem0lmtedata45degreeanglewiththeaxisofpiston motionasdlowninFig.61.Thisallowsaplaneviewofthecylinderthroughthequartz piston. Thedoubleoverheadcsmsofthecylinderheaduedrivenbyapowertakeofl‘fiom meAVL,viadnglegeubeltMey.Fwahgnmentpwposes,exunnmswaeaddedmme cmslnfis.ThisenfireassemblywummmedbyalShorsepowavuiablespeeddirect amemwmrJheoflmgofthecylmdabeadissepamefiomthecmkmeoflmgsystemof memamedengine.Thereuonfmthisistheheaddoesnotsituthesamemgleitwas dcsignedforwhichcrestesoildrainingproblems.Anoilpumpdrawsoilfiomareservoir andsrmpliesoiltoaninletpasmgeinthehead.Apresaneswitchmonitorstheoilsupplyand ifthepresslnedropsbelowasetpoint. shutsthepumpoff. Oildrainingoftheheadis [37 Figure 61(a). A picture ofsurface mirror mounted at a 45° angle with the axis ofpiston motion Direct-injection engine head ertz cylinder f = 8. 75 cm, 1 =10 cm ._ Elongated plston — Mirror mounted at 45° AVL Engine mount Figure 610)). A schematic diagram of a side view of optical chamber (above); elongated piston, surface mirror and quartz cylinder are shown. 138 mmpfishedbyapplymgasfightvacummrenmfinesmsalledinmerennnpassagesof thecylinderhead. B. lstalresystem Aspecidinmkesystemfordreexciplexcah‘brafionanddirectmjecfimtenswas neededbecuneofmeqlenchmgofmeexdplexfllmrescmcebyoxymThissynemdm needstobeabletomessurethemaasflowmteofnin'ogenandfilehandcreatea homogenmmixnuefwthecahhafiomTheMesystemfmthecah‘brafionofthe exciplexvapmphasewnsissofamuogenbdmcesystemehmflowekmentaFE), intakeheaterassembly, fuel injecfionandmixingaoneandanintalteplenlun. Similar homogmomchmgemixingsynemshavebemdevelopedmmepaafwemissimtesfingof smaflenginesmdasimfludesignwuchosmformislhemmkesyaemfmmem-cyfinder dhectinjecfimmeaslnemenmissimilm’mmecah‘brafimsenmexcepttheheatamd injectionsystemsarebypased. Theniu'ogenbalancesystemworkslikeanacclnnulator.ltconsistsofa55 gallon Wwidraphsficdiaphagrnacfinguaflowbflmcmgdefiamuogenisdefivered fiomcompressedniuogennnksdlroughregulamrstothebouomoftheconminer. The omlaisviathempmrmghazmchdimeterflemblehoseflhemcogenisnendrawnmm thelaminrflowelementwhichmeasuresthemassflowrate. The laminarflow element is Model SOmth-l % from Meriam Instruments. It is ratedul7.0SCFMmd8mchesofwaterdifl'erenfid.Thiswasmomtedbetweenmup steamtubelengthofls inchesandadownstreamlengthof‘linchestoenstnelaminarflow. MMMWMIOMMWMSfimeWImm 139 mcommuldbyMefiambennueacmuatereadings.Thedifi'aenfialpresaneponswem connectedmaMeriammclmednlhemmometuwhichdisphyedtheflowmteinSCFM. Theinclinedhrbemmomflrmedwasmodelflflfiflwidramaximumcapacityofl inches ofwaterdifi’erentialwhichallowsforUSCFMtoflowtluoughtheelement. Nexttheniuogenpassedmroughahataassembly.ThehaterelemeMwnsistsofa 300-500wauheuglmreplacememdement1hishe0uasembkismomafimflycmuoued byaclosedloop thermocouplecontrollmit(API InstrlnnentsCo.).Thisheatedairisdrawn intoaninjectionchamberwheretheatomizedfuelisintroduced. Thisinjectionchamber consisbofmahmhumblockwimthemuogmentamgamdegrcesmthefilelinlet. Thefinlinjecfionsystemcmsistsofpresnninedfilelpassingthoughofifices. The fuelwaspresnn'edbynin'ogeninapresslnevesselwhichdeliversthefuelatagivenpresslne WWmThefinlmmmmrough7ummmeNMashmofl'valve, mdmenmeodfimTheonficeswenpolyuymlfinedimmdwireexmlsiondies pmmnedbyHoosier'W‘ueandDieTheconfigluationofthisorificeisillustratedinFig.62. The sizes that provided the necessuy flow rate were 0.003, 0.004, 0.005 inches in diameter. OncedreisoocnnemdDMAmixnuewasmjectedinmtheniuogenstresmthemixnne passeddroughamixingchambacmsisfingofsphflwmmdwhescreurofvuyingmesh size.ThischamberwasalsohenedusinghesfingmpeandconuonedwithaVafiacm mainthmelevuedtempannuefmfiquidwaporafimAsachechasiglngluswaspmm linemennnedurtthemixnnewasinvaporfmandnofiqtfidfilelwasonthewafls.This mixhnedrenpassedthoughminsdaedmbewithalengthmdiamdermfioofoverwm ensure fully developed flowandpropermixing. Themixturethenenteredthe intake plenum. Thisconsistedofa 15 gallontankwithpressureandvaclnunreliefiThisprovedtoprovide 140 Polycrystalline Diamond Die g ....... \ / Stainless Steel Case Figure 62. Wire die Orifices. 141 adequatevolumetostabilizetheflowfortheLFE.Themixturethenpassedthoughastraight sectionofmanifoldtotheintakesideofthecylinderbead. C. DIrect-lnjectlonfnelcoatrsl&lasersynchroahatloa Theconuolofthefinlmthehighpresureinjectorforthein-cylinderdirect injection. wasaccompfishedwithaspeciaflydesignedconfiolboxThiscontrolboxusesan munfioma360degeeDynapushafimgleencoderJhismcoderwasanachedmmecam takeofi'oftheAVLwithaflexcoupling.Thiscamtakeofi‘isgearedinternallywichd ratio. 'l'hisallowsfortheencodertospinonceforeverycompleteenginecycle,tworevolutionsof thecrankshafi. Thecontrolboxprovidesasvolt‘l'l'Lsignaloutforsynchronizationofthe laserandtheCCDcamera. Thissignalhassuficientdelaytoallowthelaserpulseand cameratocaptureanimageatanypointinthecycle.‘l‘hefilelinjectionpulseoutsignal whichisalsoadjustableanywhereinthecycleissenttoaspecialcontroldeviceprovidedby Chysla,whichopamesmehighpresnnemjeam.Thisboxdsocmmmsadigimlmdom memmmmmdmpmfiwlsyaemthatdehversfinlmmehighpresnne injectorismadeupofafuelreservoirfi'omwhichalowpresslneOOpsi)12voltpump drawsfuel. Thislowpresaneprnnpprimesthehighpresslneprunpwhichisarollerpump drivenbysléhorsepowerelectricmotor. Thesystemplnnpsmuchmorefuelthanisneeded. soaretumfromthehighpresmrepurnptothereservoirisused. Chapter V. Results and Interpretations 5.1.Z-DSectloaoflmagesoffnelspraylnatestqnartzeyllnder Inmissecfiomtheopficalmethodsdevelopedinthepreviouschaptersueusedm visualizetheinjecfionofaliquidsprayinanoptically accessiblequartzcylinder.ln Secn’on 5.1.1., the qualitative temporal and spatial distribution of the liquid spray is described.lnSection 5.1.2.,aquantitative description ofthesprayvisualimtionismade forboththeliquidandvaporphase. 5.1.1. Fuel Spray LII" imaging lnordertosimulateaéOOrpmengineenvironment, atmosphericpressuretests wereconductedat600injecfionsperminluewitha4.0msinjectiondluationinthe apparanudesmhedinSec.4.4.Theinjeamwasfiredforthreemfompreparatmy injecfiommdafierthattheflumescenceimagesweremcordedFmdrisLlFimage acquisifimthevaficalsheetoflightpassedthroughthecenterofthe fuel injector. All acqtfiredimageshadthesmneexpaimenmlcondifimsexceptthatdredehyfimet. was varied. Figwe63showsnineLlFfluorescenceimagesthatdisphydledevelopmentofa fuel sprayanditsdistriblrtionin0.56mssteps.correspondingtosuccessive2° crank- mgledegree(CAD)inwrvdsu600rpmenginespee¢Difi‘aentfalsecolorscale assignmentswereusedtoencodethesprayfluorsscenceintensities. Figure64represents theWensitycontmnmapofWemisfimflmrescencefiomthefinlspraanFig. 63, imagecapnnewasmadefi'omo.56mst05.04msatterthestartoffirelinjection. corresponding to 2° CAD and 18° CAD, respectively. At 0.56 ms alter the start 142 (a) t, = 0.56 as (b) I, = [.125 as (c) t, = 1.684 In: (& t4 = 2.243 m (e) I; 8 2.801 m (g) l, = 3.920 nrs (It) I, = 4.480 an (i) r, = 5.040 as Figure 63. Sequemial fuel laser induced fluorescence (LIF) images after fuel injection (to) (visualiution was made by using a 5%DMA/95%isooctane fuel). (a) t, = 0.36.: (b) t, = 1.126” (c) t, = [.684 ms (t0 r, =- 2.243 m (e) 1, =2.801 m 0) r, = 3.360 m (g) t, same-e (It) t. = 4.480” (i) t. =5.040m Figure 64. The contour map of the fuel LlF image intensity in Fig. 63. 145 ofmjecdomthesptayemergedlikeabufletwithadensespafidfinldistrihrfiomMostof thefuelresidedinthecenterofthespraycore.flowever,1.126ms(4°CAD)later.the sinldimwuqlfitedifi'erentThecartercorewumwdividedbytwomainbranches withmmchdedmglemmwflMmeovenimhfllet-Iikefinldisuihrfionwas completelychmgedmaspraycmewhichhadanrongfirelconcenuafimmbothsides ofthelqrperputofthecone.However,thecentralpartofthesprayinjectionlinewasnot complfielyfifledlnvhgammhweakerconcenuafiomakinmahoflowcmesprayflhe fiqlndfilelmmeouternnfaceofthespraywasfarfiomthemainfiqlndnreamwhich wasinthecenwrmdtendedtomoveapmtfiomitMoreova,thaeexistedcouisims whichcoddproduceatominfimoffinldmpletmdreintafacebetweennitogen moleculesandliquidontheoutersurface.Therefore,itwasanticipatedthatliquidphase firelwaspredommmninthemtefiorofmespraymdfirelvapmwasdismbmedweakly awayfiomdtecme.1hisisareasonableasnmpfimsincevapors-easflydevelopby Mmizafionandbythevapofinfionprocessofdrehqtudfirelonadroplet’sm mface.1helongertaflofthesprayonthedglnhmdsidemaybedlremtheefl'ectof nitrogengasflowimpingingonthefilelondratside. 'l'heimageinFig.64c,talrenatl.684ms(6°CAD)afterthestartoffuel mjecfiomshowsfintherdevelopmemofmamuqlfidnreamdownwudfiflmgmecenml pmtofmespray.However.onecmamuenflyobsavetwobrmchesofmamsneams. Onmebadingedgeofmespray,avatex-likemofimisdevdope¢Thiswuldbedmm mevaporphasebecunethefiqrddisgenaaflyleupannbedbymeambientfllndmofim thmthegasphasebecauseofimmafiaAsmjecfimproceedsmesprayflmrescenceis more intense overall, and then shows the highest fluorescence intensity near the injector 146 tipwithavalueofo.763at2.243msafierthestartoffiselinjectionasshowninFig.64d. ltwasobservedthatintheimagesrecordedattimesgreaterthan230lms,thehollow- conespraycoflapsedcomplemlyJeausemMmemofmbientairmmmemjeaedjet Walow-premuengionbmeflhthemjector.Fudmmoretheflmrescence mansitydecayedwithanlafivelyhighconcemfimmwuddtebocomofmechmber, asbeseenverycleclyinFig64. By mid-injection (up to 2.243 ms) the fuel distribution has become reasonably axisymmeniqmdtheconcenn'afiondecreasedaxiafly(fiomtoptobouom)and increasedtemporally(from l.126msto2.243ms).0ntheotherhand,imagesrecorded fiom2.8msafterthestartofinjecfionshowedoppositebehavior-atemporaldecrease while remaining axially symmetric. Moreover, as the end of injection approached. the intensity distribution became asymmetric. At the end of injection, the overall fuel concmnafiondimmishedmdthemjeaedfinlspreadmwdthebouomofthechmber, due to gravity. Atter mid-injection, a wide fuel distribution could be observed with weakerintensity,probablyduemdreevaporafionothludForexampleinFig64f,the thirdcontourmapat3.360msafierthestartofinjectionhadanintensityvalueof0234. Asinjectionproceeded.however,itdecreaseto0.073atS.04ms.Thesefeaturesare qmfimfivelycmsistmtwiththeefl‘ecmofevapmafimoffiqlndmopsfiommepenphay ofmespray,mdpmmhlymaerapidspreadingofthevapmbyhnbldentdifiilsim.m Fig.64hand64iattheendofinjection.theoverallconccntrationsweremuchlowerand theimagefieldwasfilledbythefirelatlowconcentrationsduetothevaporphase. 147 5.1.2. Feel Spray LII-2F Imaging We now examine the two-phase distribution of liquid and vapor using exeiplex imagingAquanfitafiveconcenuafionmapofthesprayisshowanigés.Thespray consisted of the 57«DMA05%1,4,6-TW090%-isooctane exeiplex system. All expaunamlset-upcondifimswerethesameasthefilelLlFimagingdiswssedm Section5.l.l,exceptthefilelinjectiondurationwhichwassetto3.0ms.lnorderto simmateaISOOrpmenginespeedthemjecfimfiequencywassaatZSHLeqrnvflemm [500 injections/minute. Additionally, two bandpass filters wereemployedto capture the fiquidmdvaporphasespraysepuflely.lmagecapnnewasmadebenveen0333msmd 2.444msafierthestartoffilelinjectionatstepsof0Jllms.correspondingto3°CAD t022°CADatl°CADintervals.Allimagesoftheliquidandvaporphasesconsistedof averagesofIOindividuallasershots.Thesprayprofilewastaltenat333usafierthestart ofinjection. In Figure 65(a) and (b), images of the liquid and vapor phase spray, respectively, areshown, while Figure 65(c)and(d) shows contourmapsofliquidandvaporphases. In Fig.65(a)and(b),thegeneralbehaviorofthesprayshowsasymmetryaboutthe injectionline.However,higherfilelconcenu-ationsexistinthelefiupperpartofthe spray,whichisbelievedtobeduetoanasymmetryinthefuelinjector,especiallyat highchjecfimfiequencieampuficulmitcmbeseenthatthehighencmcenuafims ofthevaporareassociatedwiththehighestconcenu'ationsoftheliquid.i.e.,thevapor wasevapaafingfiommdivimnlfiqtnddromehegmerdshapeofthesprayhada high the! concentration on both the right and left sides ofthe spray. However. the center part of the spray was not completely filled. apparently forming a hollow cone. (l) 333 u: (2)444 us (3)555 us (0666a: (5)777»: (6)888»: (DIMus (8)1111“: (9)1222” Figure 65(a). Liquid phue fuel spray visualintion using a laser induced exeiplex fluorescence visualization technique (5% DMA05%I.4,6-TMN exeiplex system was used). I49 (10) 1333 us (ll) 14““! (12) 1555 us (13) 1666 us (14) 1177 us (15) 1888 [LS (16) 2min (13211111.: (18)2222|.ss 150 (19) 2333 us (20) 2444 us 151 (l) 333 in (2) 444 ll: (3) 555 P3 (4)5561” (5)177” (6)88815 (7) 100015 (8)1“er (9) 1222113 Figure 65(b). Vapor phase fuel spray visualintion using a laser induced exeiplex fluorescence visualization technique (5% DMA-5%l46-TMN in 90% isooctane exeiplex system was used). 152 (10) 1333 [as (11) 1444 us (12) l555 us (13) 1666 us (14) m7 us (15) 1888 gas (lQZWus (17)le115 (192222513 153 (19) 2333 as (20) 2444 [IS 154 (1)3 CAD (2)4 CAD (3)5 CAD (4)6CAD (5)7CAD (6)8CAD (8) 10 CAD (9) 11 CAD Figure 65(c). Contour map of the liquid phase fuel spray in Fig. 65(a). 155 (10)12CAD (ll) 13 CAD (12) 14 CAD (13) 15 CAD (14) 16 CAD (15) 17 CAD (16) 18CAD (l7)19CAD (18)20CAD 156 (19) 21 CAD (20) 22 CAD 157 (1)3CAD (2)4CAD (3)5CAD (4)6CAD (5)7CAD (6)8CAD (7) 9 CAD (8) 10 CAD (9) 11 CAD Figure 65(d). Contour map of the vapor phase fuel spray in Fig. 65(b) 158 (lo) 12 CAD (l l) 13 can (12) 14 can (13) 15 CAD (l4) 16 CAD (15) 17 CAD (16)18CAD (l7) 19 CAD (18) 20 CAD 159 (19) 21 CAD (20) 22 CAD 160 In Fig. 65(c) and ((1), both the liquid and vapor fluorescence decreased from the injector upmbouommdalmdeuusedradiaflyfiomthemfimThisprobablywaspardy drummefastuominfimprocessatthehnafacebaweentheairandfinlonthem nnface.Genaafly,drefiqlfidphaseflumescenceis70-80fimeshighammmatofthe vaporphaseinmostpartsofthesprayarea. A. Tlme-depeadeatanalyslsofthespray ‘l'hetemporalevolutionoftheliquidfilelsprayisshowninFig.66.Theintensity flucnmfionsmmeearlysmgeofmjecfimcmbeexphmedbythemjectmchmactefisfics. Uponreceivinganelecu'onicpulsefi'omthecontr'olsystem.theinjectoropcnsitsvalve maximaflymdaabflizesvdveposifimmthefirstsevanlhmdredusflmrescence mtensifieswereobminedatvafiwsaxialposifimsalongdheinjecfimfinemomthe mjector-fipmmebouom)atdifl‘uemfimesafierstmtofinjecfiomAsfimeproceede¢ meflumewenceintensityofmoaofthemvesfigatedposifimsMaeasedmdreacheda maximumnearl.67msatterthestartofinjection.Between0.78andl.67ms,greater flmrescencemmnsityfiomflqlndfinlspraywasobsavednearthefinlmjectmfip.m addition. the relative intensity of all investigated positions (or pix.) was increased. However, at 1.78 ms afier the start of injection. quite difi‘erent progressions were obsavethighmmnsityulngedinancemdlowmtemityneumeinjectm-fip.Asfime wdvedfiuthafihefinlmmnsitywudeaasedmflykeepingthehmhfivefunues modauelyconnantThispheoommonmaybeexplainedbyasamingthefinlinjecfim consiasofmnepszthevdveopeningmdmevflveclosingproceamingme opening process, especially between 0.56 ms and 1.67 ms, fuel concentration 161 1.0 ‘5 T— I T: T I T T— 1 r—T fir I—T I T I—j r T rj T I‘T h , ‘. « oa— 1 \. .. '-\ g " a’.\t\ '/e '1 06- r r/ - \\\ t /. b MI .I‘ 4 {m/ r" (L"' .\ 1r 4’. q \hol’ s" . / e/ 0.2” y‘a‘o’ -l G s .1 a _a L4 L4 LJ 1 I a LJ 1*; L l L a a a 1._LJ 0 500 11!!) 1500 2000 25!) Woodhull-W013) Figure66. Temporal evolutionoftheliquidphasefuel spraywiththedistancefi'omthetip ofthe injector; (a): 30, (b): 50. (c): 90, (d): 120, (e): 150, (t): 170, and (g): [90 pix., respectively. 1 co I I U T T r I I I .. Q (h) \\ “3" \‘D. ‘\“‘\- ‘ i \ x M . ~~ *-~ 4 K ‘su---- -’--— --§ 0‘”. (c) €:::'-—--1——-—Jwvfa::-~- -Z: d 0 “o\ ;::k::—.’- .. -,L'>—g:_‘_‘ ‘CL. .. pppp ' ~‘.°-.§-~: "'-= 02- e—e---°"" (e) V, - - ' '-~-—o . . .--- -.--- --- ---.--- -- d (0 o L l L J l ,1 J an Q 100 140 100 ”Minoan-mun) Figure67. Normflizedfimddphasefiumescencefimsitywidlrespeamthedisnncefiomthe injectortip at fixed timeafter fuel injection; (a): 0.78 ms, (b): 1.11 ms. (c): 1.56 III-$01) 1.78 ms. (e): 2.0 ms. and (t): 2.44 ms afterstut of injection. 162 continuously increased, due to the high liquid mass flow rate resulting from the large inside volume of the injector. However,at 1.78msafterthestartof injection.duringthe valve closing process, the inside volume of the injector decreases reallting in a relatively nnaflamomRoffimlexifingmemjeaw.Thereamnfornl¢ivelyhighamtensityafar distance(e.g., 190 pix.)intheinjectionlinecomparedtoanearone(e.g.,30pix.)is possiblyduetothereducedfilelsuppliedfi'omtheinjectorduringthevalveclosing process,mdalsoduetothepfiorinjectedfilelwhichwaddbemovingdownwud Figure 67 shows this phenomenon clearly. In Fig. 67, the normalized fluorescence intensityofdreliqrfidwasplouedvstheaxialdistancefiomthecenteroftheinjector-tip atvarioustimessubsequenttoinjection.Individualdatasetsof(a),(b)and(c)wetalten . a10.78 ms, 1.11 ms and 1.56msafierthe startofinjection, respectively, andexhibitthe generalprogressionoffeaturesdmingthevalveopeningprocess.Data(d),(e)and(t) were taken at 1.78 ms, 2.0 ms and 2.44 ms, respectively, and indicate the general behavior of the liquid spray during the valve closing process. ThevaparphasefluorescenceintensityanditscontourmapueshowninFig. 65(b)md(d).hdividuflnlmbasmthecmmmsshowthenhfiveinwnsityn«malized bythehighestintensityvalueobservedinaliquidphaseimagetakenat 1.67msafter mjecfiomThespafidmdtempmalprog'essimsofvapmphasesprayuesimilammose oftheliquidphase.lnparticular,itcanbeseenthatthehighestconcentrationsofthe vapormmociatedwiththehighestconcentrationoftheliquid,i.e., thevaporis evaporatingfiomindividualliquiddroplets'l‘hetemporal behaviorofthevaporphase spray at different heights from the center of the injector-tip is shown in Fig. 68. The general features of the spray look the same as the liquid phase. However, the intensity 163 om rrTro‘UleITUT'Tror—TT—TTFI anrludllsenslty g 1 I 0 I '1 FT ' l ' ' ' ' l U ' ' '1 P/ O 1 \ °\ W7 ‘1\ f P/O E; / 9 \ L14114L4L1AA all Lkl GLLJLJ1111J4L11L4L4414L+1L_J_ 0 ED 1W 15D 2000 2500 mam W618) Figure68. Temporal evolutionofthevaporphasefirelspraywiththedistarcefinmthetip ofthe injector; (a): 30, (b): 50. (c): 90, (d): 120, (e): 150, (f): 170, and (g): 190 pix., respectively. 164 was relatively constant with a value of 0.01, much smaller than that of the liquid phase throughout the injection process. B. Simpkuodelofdropletvaporlaatloa hChapmerapofinfimofthefiqluddropletthroughtheheat-msferormass. mferconuouedcasewasdiscussedindetail.1herearetwomodelsforthe vaporization of the liquid droplet given by” D,’ -D’ = Ic,t (25) -dlu ldr = k,D‘ (26) D, -D = 1,: (27) whereDisthedropletdiameteratfimet,D,istheinifial-dropletdiameter,thek,’sarethe vapofinfioncmsmntsdetaminedbyacombinafionofphysicalconstmtsandbythe Nusseltnurnberofthedmplct.andmisthemassofdropletlostduringvaporization. For arelativelylagedropletsiaeEq.(25)canbeused.butforsmalldropletsEqs.(26)and (27)areprefl'ered. Itisasnmedthatthevapofizafimbehavimofthefiqluddropletismass-orheat- difl‘usion controlled as described by Eq. (25) or (27) and. further, the measured flumescenceimmsityispropafimalmthemasofvaponaeddroplaThisgivesme following relationships for the emitted intensity vs. time. 165 t = I/k’mo ‘0) 3 Do/k’(I’D/DJ ‘ =- ”540112 ‘0’) ‘3 Dal/kl (I‘d/Dov 1 = kJDo’ -D’) 3 Do"; (I-DVDo’) (50) whaekfisthepropafiomlnyconstlnrelafingemisfimmtensitymmevolumeofme vapofizeddrnpkthmdermmdamdmequafimfivefeannesofmeemissimintensity drretodropletvapodnfimaflconstantsintheequafionwerenormalizedwith 1 (It, =k, =k‘=D,=l),andDwasallowedtovaryfi'omlto0.Atheoreticalcurveforboththe largeandsrnalldropletsizeisshowninFig. 69and70. lftherelationshipbetween vaporintiontimeanddropletdiameterarelinearasshowninFig. 70s,thenEq.(27)isa betterapproxirnation forthe vaporization behavior of smalldroplets, suggesting mass~ nansferconnnL”OntheotherhmdifthevapofinfionfimeisproporfimaltoD’,as showninFig. 70b,thenEq.(l)isabetterapproximation forthevaporintion of relatively Inge lh'oplets, suggesting mass- and heatqu control.” Asecondapproachcanbetakentowuddeducingtherelafionshipbetweenband tfiomhemissionflmrucenceByasnmingthemaximmemissionfimrmenceof thevaporsignall_tobeproportionaltotheinifialdropletmass,”i.e.,l..=k.D,’and recalling from Eq. (50), that! 8 1'10} ~D’), the following expression canbe reached: DID. = (I - 01...)” (51) Substitution of Eq. (51) into Eq. (25) results in the following relation: I. .S J :5 '1 l . g 1 1 .J 1 o l L l L L a l L 1 4 J 1‘ i a IL L l L d 0 0.1 0.2 0.3 0.4 0.5 0.0 0.7 0.0 0.0 1.0 The (1-0') Figure 69. Theoretical calculation of choplet evaporation; ” (a): small droplet, (b): large droplet. respectively. 'I ' —U I I. r T r T T T—T 1 T F I f I r .l 1.1!) - 1. II"-’. a ‘ " e—--a F1 '1 1- . q 1- .... d 0.15.. ‘5‘“ 1 33. : : g 0.9- 1 : «mg-m 1 - 0.1 .. 0.8- - Gr . 1 L 1 1 1 1 4 . 1 4 J L 1 L 1 1 1 ‘ 0 0.1 0.2 0.3 0.4 0.5 0.0 0.7 0.0 0.9 1.0 Donna-tuna) Figure 70. Qualitative size-dependent droplet evaporation; (a): small droplet, (b): large droplet, respectively. 167 I - (It/01’)! = (Fl/1...)” (52) Aplotof[l-UL_]”vmtshould,therefme,yieldasu'aightlineifnormalhector mumquofinfimforhgcdropmmthemeofmafldmplm, Marion of Eq. (51) into Eq. (27) yields: 1- (ks/DJ! 3 (Fl/1.1)” (53) Inthissitmtion.aplotof[l-—l/I_]"’vmtshmfldgiveastraightline. Inordaminveefigatethcvapmiufionprocecsquafitafivelyinmemficspray mehmflmrucencc intensitywasusdThemthetoml fluorescence intensityofthevaporphuewascalculatedfiomtheimageini’ig. 65(b)asafimcu'onof elmedfimelnordertoseewhetherthechangeintonlvaporplmefluorecccnce intensitywasduetothevaporizationofthe liquidckoplets, (l-l/L.)"inEq.(52)or(53) wasplouedasafimctiouoffimeasshowninfig.71.1‘heexponentxdependsonthesize ofthedroplet:formlldroplel3(s0.l um),x=1/3,and£orlarge¢hoplem.x=2/3.The dataindicomedby(a)md(b)inliig. 71 representtheanalysis foramlldropletanda reletivelylargcdroplet, recpectively.‘l'wo solidlinec inthedatareprecentthebestfits of mmkmbemmedmmm)werefinedweflwithmexponenfial equafiomwhflednpoinu(a)wuefimedwimuimpklineunyeedmindieedngmafl dropletvaporiutioo:y=a-bx.whereyiseqmlm(l-lll_)"’,xistheelapeedtime(or vapofiufimfimexmdamdbmfitfingconmmindiafingthedmplet I68 0.7 fit I Y ff fir 1' fir T '— T T T~—I' Y I r Y— Y_fil’ I Y '— V a b 4 D 1 ( I. < 06 - - . p d t d L < J 0.5 "' .4 D 1 D 4 l' l 0.4 E 0.3 E -PIP AlA‘A 2'... 1 . C] (b) J r. i 0‘2 : ‘3 1 ’ D 1 0.1 - 1 El U j o 1 1 mg L4 L ;1 144 L 1 4 . g4 0 0.0005 0.001 0.001 5 Figure 71. A plot of (l-l’ll’...)I as a fimction of time (vaporization time).(a): x = 13, small droplet, (b): x = 2/3, telatively large droplet analysis. [69 diameter ratio, i.e., (D0,) and vaporintion rate constant (k, lDo), respectively. Although thevaporphaseflmrescencennensityistheannoftheconmhltionsfiomme vapminfimofafldifl'auusizcdroplemjtmayhecomhidedthateulymthemjecnm processonlysmalldropletswhosediamemruesmallerthanOJ um”seemtoplayan importantroleinthevaporizationproceas. 5.1.Dlrectenglleappllcatlona Whflednedirectcah’brationmethodmentionedinChapter4canbeappliedto bofitheflqmdmdvapmphasefinelspraqunnficafimwhichismambiemtempmnne mdprcsamhmaynmbcappficablefordueamcasnementofthefinlconcenuanmm aninternalcombustionchamberwithamotoredenginecondition. Sincethefluoresccnce intensityfiomthephasemukascanheafi‘ectedbymeprccaneJempaanIemdme mixnnedensitywhichmeconfinuouslychangingwimthemovingpistmmsidethe quartz cylinder, some mems for the quantitative evaluation ofthe fuel distribution, ccpedaflydmhgthemmpreasimmisnecmymthemmmedengimmndifiom hfamdtcflmruccncemtemityofmeDMAvapmphascdevelopedmChapAdepends ontheconcentrationandquantmnyieldexplicitly.Thequantumyieldofmostofthe WmoleaflaanhuDMAdependsontempaanneandpreenre.“Thaefme,the fluorescenceintensityofDMAinthevaporphaseisanimplicit functionofthose parmetasinamedenginecondinomOntheotherhanimeefl‘ectoftempaanne mdmvuiafimmfimfidphanflmrmmtmsityisnmummme the! intake process. thepressure of the inside quartzcylinder is relatively constant (i.e., ambient prustne) as shown in Fig. 72. Most of the liquid phase fuel can survive only a I70 9’ 01 T TFIIITITVIIITTT—ToilrrliIiffiillrirrrd ‘ ‘ n . 3 _ d ‘ q .1 25 - ‘ -« < . d 180CAD 1 .1 _ : m : ———-’ 1 Intake 1 0.5 process 1 0 ~ A A . preeaure(a.u.) .3 .L (fl UH! lTIllllTT'lllIllIIl‘IIII‘IIII‘WT C1 .05 411—1 _1_1__11l_1_1.11'1L11L1LLLLLLL11LLLJL‘ . O 100 200 300 400 500 600 700 “ammo-(can) Figure 72. A plot of pressure versus crank-angle degree using a Chrysler 2.4L lG-valve. The enginespeedwas 600 rpm. I71 shorttimeafterfuel injectionbecmseofthefastfiielatomizationandvaporization proceasmnmairflow.AsarenflLMgthecompressioncyclemwhichtempaanne mdpresmueuenlafivelyhigh.mostofthefiq|udphasefinladnbevapofized;its fiacnonismuchmaflerthnmatofmevapmphase.1herefore,thenensecfion. emphasisisplacedonthevaporphasefilelcah‘brationinamotoredengine. 5&I.thvaporcallbratlon FigueBshowstheexperimenhlschemaficdiagmmofanm-cyfinderfilelvapor cah'brafiothesameexpaimmmlsenmwasusedmmeaamlenginemeamuementsm mecah'bradon.$%DMAwasusedm9S%isooctanefiiel.Aperfectlymed1 uniform gaseouschargewassuppfiedmthecombusfimchamberthmughtheengineinmke manifoldlnthemotoredenginecondition. cah‘brationwasmadeusingaequivalence ratio(¢)insteadofDMAconcentration.Theequivalenceratio (¢)isdefinedasthefuel toairratioatstoichiometry.ltcanbecelculatedbymeasm'ingtherateofmassflowof fuelandair. The combustion chamber was emulated by an optically accessible quartz cylinder, shown in Fig. 61(b). A schematic diagram of the engine cylinder headconfiguration is shown in Fig. 59(b). Figure 74 represents a typical fluorescence image as a fimetion of crank-angle degree (CAD) obtained fi'om the premixed DMA vapor phase using a 0.004” odficesizemdSOpsifiieldeflva-ypresanewithGOOrpmenginespedThebfight spxbmmeimammhuynmefigmhmdfitisbefievedmbeduemthe reflecdmbaweenmemnermdwerwallofmeqmcyfinder.""mdividualimages I72 Excimer laser 308 nm 17 ns pulse width 5-10 pulse/s ‘— _ ICCD camera 576x384 pixel Figure 73. Aschematicdiagramofanin-cylinderfiielvaporcah‘bration(oracmal engine measuements)setup.‘l‘hefinalbeamsteeringmirror(M3) movesthelasersheetthroughthe combustion chamber (quartz cylinder). Approximately, 800 pm thick and 50 mm wide horizontal sheetoflightwasproducedusingafield—opticsconsistingcylindrical fused-silicalensee. (L4 =- 145.63 mm. L3 = 47.5 m. L2 - 47.3 m L1 = 300111111, respectively). I73 74(CAD) 90(CAD) IZiXCAD) 150(CAD) 180(CAD) 210(CAD) 240(CAD) 270(CAD) 300(CAD) Figure 74. A typical of DMA vapor phue calibration image (5%DMA in isooctme solvent : premixed & preheated) using a 0.004" orifice and 80 psi the! delivery preasme.(0.004"-80 psi). 174 weremadebyaveragingoverflwlasershom.Eachimagewuthenusedmcalculawd theequivalenceratioasafimctionofCAD.AsseeninFig.74,althoughthesame emuvalencemfiowumedmeflumescurcemtensifiesofconsewfiveunagesdowly moreasedasthemk-ngleadvancedfloweva,eachunagerevedsflmrescence auenuafimflmgmehsabemndirecfim(e.g,meflumescencemtensityatthelefl-hmd sideoftheimagcisslightlyhigherthanthatofright-handside).Theattenuationefl'ect wasenhmcedbyinueashgeqmvalencerafioandCAD.Howeva,thehuerhadmore efl‘eccwpeciaflylargecm. UsingtheseunageatheDMAfluorescenceintensitywasthencah‘bratedasa fimctionofCADforaflequivalenceratios.Themassflowrateofisooctanewas measmedasafimctionofappliedpressm'eandshowninFig.7S.Aswasthemassflow rateoffireLthemassflowrateofairwasalsomeasm'edatGOOrpm.whichprovides various equivalence ratios. The fluorescence intensity of DMA vapor phase for various crank-angle degrees is shown in Fig. 76. Regardless of their relative equivalence ratios. thegenaalfeahueswueqfitesimflmdmingtheinukeprocesstheflmrescence MensityWyMeasedhnMgmecompresfimprocesshmqeasedabmpuy. Generafly,thoseimapsthatcontainedhigheqrflvalencemfiospro¢nedhigh fluorescenceintensityateachCAD.lnordertoknowtherehtionshipbenveen flmrescememtauftymdeqmvdencerafiowimrespeamvuiommk-mgledegrees. medaumFig76waereanangedmdtheflmrescenceintensityploaedasafimefimof equivalenceratioatflxedCAD.Figm'es77and78representtheresultsdmingtheintake and compressionproceas. respectively. The intensity of individual data in Fig. 77 and 78 was normalized by the maximum intensity (e.g., the maximum fluorescence intensity per 175 E7 111l1j11 TIT—YTrTT. T‘ I I ' I III—I—rs [.171 LLtJJ WWW) o 3 § § § § § ’é’ § WWW L 1|1111l1 (a) (b) V I K' _. 3 O I LV (c) e - u a i e a ,W1L1111J1HL11111L11H] 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 maaatlowrabcflaoeetanflghse) Figure 75. Aplot of isooctanemass flowrateasafimction of appliedpressme;(a): 0.003”, (b): 0.004”, and (c): 0.005” orifice size. respectively. j . fiT [— I I Fifi I I’ r T hi I— r - d r ( ) /I 1 5;» ’5//: 1 - ’IA'I :- 3 ' X/d ,4, ‘4 b (B) “/lo/t/‘a ‘ ‘; ,o-xx/a I/( o J n -- c/Otéf/ ” /'/ I J - , tin-.3. ' '7’,” ,o « E .’ I ’0‘._-'.’-- ,:d’ / q ‘ ’ ..«O --v-o-V' -' ’ + 3|- r ” """" .0." O .1 ./l -0--',ao-' / : (’3'. o —-o”°--°’ 1 . :7 -—-°” " ‘ . o; ° 1 2 r L L__ L_L 1_ L L a _r_ _L 1 L4 L 0 1m m m Figure 76. AplotofnormalizedhnensityofDMAvaporphaseasaflmctionofamk-angle degree for a vmious equivalence ratio (0); (a): 1.09, (b): 1.03 l, (c): 0.954. (d): 0.862. (c): 0.779, (t): 0.682. and (g): 0.563, respectively. 176 pixelcanbeformdwheneqrrivalenceratio(¢)isequalto1.09at300CAD).Asshownin Figs.77and78,theflmrescencemtensitywasinaeasedbyincreasingtheCAD.This maybeamlainedbythegroWofdeueDMAvapamixnneduemthefisingpiston. Rishrgpistonmtheefiecfivevohmefliaebycmnesrehfivelydensemixnne. However,asobservedirrFigs.77and78,theinmnsityincrementdoesnotfollowalinear pmgressionasthemk—mgleadvmces.AsdiscussedhChap.4,thevaporphase fluoracencehmensityisproporfionalmthequmnmyieldmdfluoreecentspecies concentrationintheexcitedelecn'onicstateasshowninEq.(54) 51°C [$310154 amt-mm (54) Infacnquannrmyieldisafimctionoftemperattneandpressmei.e.,atgivenspecies concentration, fluorescence intensity is diminished as increasing temperature and prunneprobablymnmmlfisimdqimchmgbeMeenmolemdninamomredengine Mfimwmymsmmmmmmmamodynmicmm (e.g.T,P.de)ceconfimnuslychangingwiflimk-mgledegreethaeforefor direameaanememofmequivdencemfiofiomtheflmrescentimageitisnecessarym nkemmaccmmofthevuiafionmpreumMmpmuemdmixmdensityinsideme cylinder.Maeova,ushowninFig7inChapter3,meDMAabeorpfionandemission mofleovalmmaefmemaeexissspotenddsecondaryreabsorpfimofmeDMA flmrescenceemissionbythetmexcitedDMAvaporwithinthedistmcebetween excihflonmnemddietopofthepistmhead.FofloMngShimimetal.“themeamed fluaescencemtensitypapixelwumpresenteduafimcfimofmeequivdencemfioa) and crank-angle degree (9), leading to the following equation to replace Eq. (54) 177 1T1r1fiIrTTTroTT1'T'IrlI ' ' é/ a - o/‘ 5 / 0/0' ,0 p 8 r 1L1d§ “monument“ 981$UQI80067J181075 .. .srman-smbsame «morn-menus- - .smmssamunAa \\ I s \ \ § m M0“ \ ‘\ .~ T'r 0&111441 liliLlllllLJLllLilLJA 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 W “b (s) Figine 77. Fluorescence intensity as a function ofcquivalence rmio (4) for various crank-angle degree(intake process). FT I r VI T T ‘I T r j r 1— I I r 1 T I 1 r1 T/A 1m" (I «- . ’/ /q 1 ' ././ ,m’ . i I. ’ /V’ /“ a 0.75- ’/ /" ,e'M'; E P /’///aéé'és%=”’” .. . .1 .— ’ , .4 Qw— ,/ ’X‘ég _ I. ’/ I§:;' d b / ,és/ a «mans-immune : F’ , /¢ 0 «trimaran-am w” / s/ : massages: " : ”3% o 3-mMa-wa.osrba 3 Gt: 1 1 l r J 1 L1 L_l__l L L 1 L_1_1 L J 1 L 1 J .l‘ 0 0.2 0.4 0.6 0.8 1.0 12 Willie“) Figure 78. Fluorescence intensity as a fimction of equivalence ratio (0) for various crank-angle degree (compression stroke). 178 unnamed-mom (55) Here, a(0)andfl(0)areeaplicitfimctionsofthecrank-angledeuce(0), andareagain imphcitfimcfimsofuansiempresarewmpammixnnedensitymdmcdistance betwecneacitafionzoneandthempofthepistonhead.1hedashedlincsinFig. 77and 78mfitsoqu.(55)todredata.Onmebasisofdtemcmrrcdfluorescenceintensityand chvalenceratioathefoflowingparamewrai.e., a(0)andfl(0),wereobtaincdfornine representative crank-angledeueesandwcretabulatedin‘l‘able 7. Thosevalues inTable Gwereusedascah'brationfactorsinthevaporphueanalysis. Table 7. Tnmsient Temperature, Pressure, Mixture density, and Fluorescence Absorption Efl‘cct in die Emission Fluorescence Intensity. \ Crank-Awe Danae) 0(9) 6(9) GOCAD 0.794 1.126 900w 0.936 1.023 120 CAD 0.995 0.952 150 CAD 0.947 1.079 180 CAD 0.975 1.044 210 CAD 0.962 1.129 240 CAD 1254 0.811 270 CAD 1.505 0.773 300 CAD 1.523 0.967 179 5.2.1. Actual engine measurements and interpretation ThebottomviewsoftheZ-Dscctionofliquidphasefireldisuibutionwerc 0M0incdat600rpmenginespccdusingaChrysler2AL, lévalvedirectinjcction spark- ignition engine at wide open throttle (WOT). The isooctane fuel doped with exeiplex forming chemicals (e.g., S%DMA05%1,4,6—TMN090% isooctane), was directly injected inmfliecomhrstionchamber.1hemjecfimpresanemddmafimwassetm700psiand 1.5ms, respectivelyandthefuel injectionstartwasmadeatéOCADsflerTopDead CenerTDC).medieprelhnmaryexpcdmcnaitwufomdthnthefiqmdphasefinl could surviveuntil 150 CAD alter fuel injection. Therefore, it was necessary to observe the qualitative liquid phase fuel distribution within tint time interval. Figures 79(a) and (b)showthehqrudphasefireldismhrfimsbetwcen74md90CAD.Tho&imagesare avmgeovafiveluaexpoauesobtaincdwimmehodmnnlluersheapassmgflrrough thecrosssectionofthecombustionchambcrathmbelowthetopofthequartzcylinder. ‘I'hetwosequentialimagesatfixcdcrank-angledcueeinFigs. 79(a)and(b)represent thecycle-to-cyclevmiafionshthefireldistrbufionsThefirstimagehFigmc 79(a)was taken at 3.89 ms (74 CAD) after injection start, and revealed a relatively highly stratified chargedism’hrfiminthecentrdputofthespray.Asfimeproccededmesuafiflcd chngedispawdwidelyanddissipmcdmiswasmdcipatcdmnmthefastaominfion andvaporizationproccesintheaccompanyingairflowfleld. DuringthistimeJhchigh intensitypartofthefirel(e.g.,rcdcolor)inthecentnlpartofthesprayfluctuatedalong thehorizontallineofthecombustionchamber. I80 74(CAD) 76(CAD) 78(CAD) 86(CAD) 38(CAD) 90(CAD) Figure 79(a). [rs-cylinder liquid phase fuel distributions using a 5%DMA05%1,4,6— TW090°Ioisooctane exeiplex system (fire! iry’ccn’ou start : 60 ATDC andfirel injection duration : 1.5 ms). [81 74(CAD) 76(CAD) 78(CAD) 86(CAD) 88(CAD) 90(CAD) Figure 79(b). An example of cyclic variations in Fig.79(a). 182 A. Uquidphasefneldistributions Liqindphaseflrelimagewasobtainedbyonelampulseexcitafloni.e.,no avaagingwaantdrcdlnordermobsuvethemstameoirscycflcvriafimstwo conscanivehnageswueacqtdrcdataflxcdaank-mgledeuee,flom74m300CAD. Figmes80(a)md(b)showtumconsecuflvefiq\ddphasefinldismbufimimagesasa functionof CAD. At 3.89 aflerfuel injection start(74 CAD), highly localized fuel wasobservcdnnidethefirelspray,pmficularlymdreleflmdfightcmnas1hisisa directindicafimmnmmtcnsejetoffiqmdcmagesfiomflreleflmddghtsidcsofthe fuel injector tip. The consecutive image at this crank-angle shown in Fig. 80(b) represents quite difl‘erent features, namely, the existence of significant cyclic variations. Approximately8.3ms(90CAD) later,theliquidphasewasdistributedacrossmoaof thecyfindercross-sectiomandaroimd16.6ms(120CAD),thearcainsidethecylinder wasalmoflmifamlyocarpicdlntheculyinnkeprocasaflerthestmtofmjecfiom it seemedthagdthoughthefiqrddphaseinifiaflypeneflflcdmwudtheaxidfineofthe wrayinjcction, it quicklybrokeupintosmalldropletswiththeaccompanyingairflow andrapidlyvaporizcd.InthetimebetwcenZSmsOSOCAD)and4l.7ms(210CAD) aflerthestutofhjccfiombothatomizafimmdvapmiafimprocesseswereacuvemh thataveryanaflamormtoffiqrfidremaincdSomescaflercdumrpsofdropletswem fomdnearthewaflsofthecomhrsfionchambcr,mdonthelefiwaflofthecylindera drinhyerofthefiqmdfllmdevelopedThisisbehevcdmbedinmtheimpmgemcmof liquidontothewallofthecombustionchamber.At50ms(240CAD)aflerthestartof mjccdomhoweva,thefiqmdfirelwasenhmced.especiaflymthecennalputofthe combustion chamber. In Fig. 80(b), the same feature was found at 58.3 ms (270 CAD». 183 210(CAD) Figure 80(a). ln-cylinder liquid phase fuel distributions using a 5%DMA-5%l.4.6- TMNOW/oisooctane exeiplex system (fire! iry‘cction start : 60 awe and file! iry’ecrion duration : 1.5 ms). 240(CAD) 270(CAD) MCAD) Figure 800)). An example of cyclic variation of liquid phac fuel distribution in Fig. 80(a). 185 As time evolved m, the liquid firel disintegrated into small droplets but still remainedmthecenualpmtofmecombusfionchamberat300CAD.1heenhancemem oftheliquidphaseat240CAD(or270CADinFig. 80(b))isnot1mderstoodwellbut befievcdmbeduemairflowmofiminsidecombmfionchamber.Anotherpossrble reamfortheflumescenceenhmcememisfiqufacumofvapofiudfinlnmisspcciflc valueofCAD. Incrdermmakeaqumfitafiverepresentafionofflrefiqiudphascfieldismbmion. Fig. 80(a) was converted to a iso-intensity contour map, and the results of intensity- concentration (or mass) relation of the liquid phase calibration in Chap. 4 were employed. Figure 81 shows the mass distribution of the liquid phase fuel inside the combustionchamber.ThenmnberindicatesflreamoumOfmasspresentinimit of nanouams (107g). As shown in the image taken at 74 CAD, the fuel distribution is highlyasymmetric.Relativelydensefuelwaslocalizcdinthelefiandrightcomerofthe comhrsumchamber.Asdmeproceede¢themassofthehqtudwasnducedmdspread omEvarmafly,averysmauporfionoffiqiudfirelremainedatBouochadCenter (BTC)(0==l80°CAD)cxceptforathinlayerofliquidfilmonthelet'tsideofthe combustioncharnber.Atalatcrtimeafierthestartofinjcction(0=240°CAD),asmall amomtoffiqrfidwddcnlyappcucdmmostputsofthecombusfimchmber.Almough mmhhqiudwapmatcdaTopDcadCenmrapproachcdasnaflmassofmedroplemsfin remainedinthecentralpmtofthecombustionchamberat300CAD. I86 240(CAD) 270(CAD) 300(CAD) Figure 81. lrr-cylinder liquid phase firel distributions (contour-map) using a 5%DMA05%l.4,6- TMN090‘Yoisooctane exeiplex system (number indicates 10" g / pixel) (fire! injection start : 60 ATDC and fire! injection duration : I .5 ms). 187 B. Vapor phase fuel distribution Asinliqrddphaseimagingtwoconsccutiveimagesofthevaporphase distributionwereobtaincdatthesameCAD.Exceptfortheearliesttimeaflerinjcction (e.g.,74CAD),drevaporphasedisuihruonwasqidtedifl‘erent&omthat0fthefiqmd phmT‘hetwoconsecutiveimauesobtainedatthesamcCADinFigs.82(a)and(b) representthecycie-to-cycievariationinthevapordisn'ibution. lnitially.adensevapor pinseappcmedmdthenmucklydifliuedflrroughmnthecomhntionchamba.Asme cmnk-angleadvanccd,i.e.,&om90t0210 CAD,thediflirsionprocesswasdepletedthe cenualpmtofthechamba.Ontheconmy,Mgthecompressionprocessvapor phasefielwuenhmceducaflyandfmmcdstrafificdchargemoimdthefiwlmjcctor.m ordermmakeaquanfiufiverepresentafiomthevaporphasedism'bufioninFig 82(a)was wnvutcdmmiso-mtcnsitymmmapmdtherendtsofmtensity-equivflencemfios ofthe vaporphase calibration in Chap. 5 was employed. Since, at given CAD(0), the flumescencemmnsityisknownfiomthemcaanemethwoimknownparameters, suchasa(0)andfi(0)canbeobtainedfi'omTable6,thecorrespondingequivalenceratio (0) can be calculated firm the equation, 1,. = a(0)(l-exp[-B(0)¢]). Figure 83 represents thevapmphaseflrel'dim’bufioninthecombusuonchamberasafimctionofCAD.The indivimnlnimbasnotcdmthecmmmapmecqruvalencemfiosAscanbeseenhere duhgthemnkeprocessitseemedmmefikdyflrattheamomtofvapormsweuasits eqtdvdencerafiodhnimshedprohablymnmflrediflusimofvapormuingflremmke mCmvasely,Mgmecompressimprwesamoremtensevapmphaseformed asmecrmk-mgleadvmccdandhighlysuafifledvapmchmgemixnuewasdeveloped near the fuel injector at 300 CAD. 188 74(CAD) 90(CAD) 120(CAD) . ' ‘ ‘v': ‘ J .1, ;..5.~;.-m;§j§ \ 3 “1?; ’4 ' ‘1 24MCAD) 270(CAD) MCAD) Figure 82(a). ln-cylindcr vapor phase fuel distributions using a S%DMA05%I.4.6- TMN090°/oisooctane exeiplex system oilel infection start : 60 ATDC and fire! injection duration : 1.5 ms). 240(CAD) 270(CAD) 3MCAD) Figure 82(b). An example of cyclic variation of vapor phae fuel distribution in Fig. 82(a). 240(CAD) Figure 83. ln-cylinder vapor phase firel distributions (contour—map) using a 5%DMA05%1.4.6- TMNo90%isooctanc exeiplex system (number is the equivalence ratio (o) (fire! injection start : 60 ATDC and file! injection duration : [.5 MS). I91 AlthmrghmesUafificafionofvaporphasefirelasthepistonapproachesTopDcad CenerTDQisnotweflmdaaoodthismaybrmgbenefltsmmefirelcombusnon processbccmnenrafificdchrgemimnemmdmespmkplugcanpreventthemisfire whichobfiouslycmscsfirelwaste.llccenfly,mmysmdiesofthedirect injection shadfiedchzge(DISC)enginehaveappeared."‘“"”Withaleanmixnne,the strafificafionofchlgemotmdthespmkplugismemostimpmfacmrmthedircct injccfionenginemhnprovingfireleficiency‘”mdrcducingmwN0,emissions Chapter VI. Conclusions and Further Recommendations 6.1.Summaryandconcluions lnthisstudy,vinnlinfionofatwo-phnsespnywuperformcdwithlasersheet flluminanmmmopdcaflyaccemrblequuucylmderusingahscrinducedexciplex flmracencemchmque.Two-dimcnsiondflumceccnceimagesecfimsflomthefiquid mdvapmphacswaeobninedlheexciplexflumescmceprovidednmcolmemission flomthemommaandenciplexwhichuggedrcspecuvelythevaporandliquid components of the spray. The work evolved in four stages: (1) development of the exeiplex m um’ng NN-dimethylaniline (DMA) and trimethylnaphthalene (TMN) - isomers in a 2,2,4-trimethylpentane (isooctane) solvent; (2) calibration of the exeiplex fluorescence for quantitative interpretation ofliquid and vapor phase fuel concentrations; (3)exciplexvisualizationofahoflowconesprayinastatictestcylinder;and(4)exciplcx vianlizafimofthefiseldisnihaimmmmtunalcombusnmchambermdermotorcd mgimmndiflmouavafiommdconclusiomfiomthefomsecfimsmnpresentedm the following. 6.1.1. Development of new exeiplex system Newexciplexsystemsformcdbetweendimethyi-ordiethyl- substinrtcdaniline andnimethylarbsnnnenaphthalcneinisooctanefirelwcredevelopedThe 5%DMA-S%l,4,6—TMN in 90% isooaane exeiplex system had the best quannim yield andiargefispectalseparafionbetwcenthemonomermdexciplexemissimA 192 I93 flmrescaiceexpefimennconduaedattempemnuesaromdzwc,mvcaledthatmis systemwassufliciendystabletopermitusefiuapplicationsinmisregime. 6.1.1Cfibrationofexelpleaflnorcscence A.Arnbientcondition Usingtheopfimizcdexciplexflmrescencemtensityvscmcennafimcah‘hadon expaimcntswuepafmmedlhereafltsshouedthatasmaflcmcennafimswoththe fiqiudmdvapmphasefluoreacencemtensityaerelineuafiflitheconcennafion(m mass) of the fluorescing molecules. Although the molar absorption coeficient (a) of the liquid droplet employed in the liquid phase calibration was optically thick (8 2 3.0), the mcaancdflmreecencemtensityvmdroplamassshowedafineunhnmship.This cmbeexphmcdbythesannafimefl'caofmeflmrescencemmnsitycmmledwimavery highexcitafimflomthemtenseexcimerhserptflse.Underthesecondifionsthe flmreecencesignalismdependentofthesomceinadimce,flumesccnceqummm eficiency, and the molar absorption coeficient c0), because the rates of stimulated absupnmmdemissionuebalmcchherebtheflmrcscencemmnsityMydepends mmewncmuafimofsolme.mmevapmphasecah'brafiomtwodifl’acntapproaches, dircapreeunemeannememmdabsorpfionmeanuemcnnwcrepuformedmcalauate themoluconcenflafimoftheDMABasedontheClmsius-Clapeyrmcmmmthe direapresanemeaanemcmwasmoreacanateflhepresanemcaanementsflomthc directmethodgavemuchhighcrvaporpreemuesameflvefimedthmmoscof absupdonmmuementsltappcmedmnthemosthkelycauseofthediscrepmcywas anincorrcctvalueofthemolarabsorptioncoeflicientofDMA,perhapsduetothefact [94 matmevalmusedwasmeMAmasolufiomwhcrcasthedirectpresanemeasnement msinthevaporphase.Morcover,sincediemolarabsorpnoncoeficiem(s)isa charactu'isficofflresolmemddependsmthewavelengthoftheexcitafionfightsolvent, andelcabempdonmethodismmesensifivemurorforvapor-phase concentrationswhicharedependentontempcranne. B. Motorcdenglneconditioe In a motored engine condition. liquid phme fluorescence intensity-concentration (ormass)cah’brafionreafltsweredhecflyusedbyasstmingthattheefl‘ectsof tempuahnemdpresunemtheflmrescencehtensitywemweakhthevaporphasea pre-mixedDMAvaporchargewassupplicdandimiformitywasprovidedinthe combustion chamber. A measurement of the fluorescence intensity versus equivalence ratio (e) was performed. The calibration results showed that the fluorescence intensity maeascdafithrcspectmcqtuvalencerafioasthemkmglemcreaschomover,at fixcdcqmvalencemuo,flmrescencemtensitymaeasednmlinculyasafimcnonof crmkmgle.Thiscmbeexphmedmainlybythefactmmecomprcssimincrasesthe density ofthe DMA molecules. However, negative contribution‘could be expected from increased collisional fluorescence quenching, re-absorption, and additional Mutation. 6.1.3.Reealtsofthefnelsprayinastatieqnartzeylindcr Rearltsofthefiselspreyobtaincdbylascrilluminationshowcdthatithada hoflowconedrape.“ehiMcmcentafimsofthevapmwereassociatedwith¢e highcfimncenflafimsofthefimfidphaseimplyingthathevapmfirelwasevapmanng 195 flomindividual liquiddroplets.Thespraygcnerallyshowcdsymmen-yabomthespray axisbmmhighconccnuafionmuymmenywasobsavedThefiquidphaseflumescence wastypicafly70-80dmeslargammthmfithevaporphasemmostpartsofmespray mTomedasmdthedroplavapminnmpmcessmmespraymsimpledroplet modelwasemploycd. chutsshowcdthandlflingtheinjccfionprocessverysmall droplets(dia.$0.l um)playanimportmrtroleinthevaporintionproceesandthemass dimision-cmuoflcdvapodmfimflnhammhcadiflusim-cmnofledfisthedominam efi'ect. 6.1.4.Rcsnltsinarnotorcdengine Two-dimensional image sections were obtained from an optically accessible combustion chamber equipped with a 2.4L, l6-valves Direct Injection Spark-ignition (DISDengineasafimcnmofthecrmkangledeuce.Thein-cylindermeaanements revealed that both the liquid and vapor phases showed cycle-to-cycle variations. However,theliquidphaseshowcdmuchmorecyclicvariation.Thelifespanofmostof theliquidphasewasrelativelyshort,i.e.,dm'ingtheintakeproccssmostoftheliquidwas vapofizedThisisbefievcdtobeduemfastammizanmmdevaporadonreadfingfiom thennbtdauahflow.lntheearlystageofthecompression,thefiqmdphascwas enhancedahrupdy.Neartheendofacycle,mostofdieliquidplnsefi1elwas Wandavcrysnaflmassoffiqmdsfiflmmainedinthecennalpmtofthe combufimchamba.1hisispmbablymnmm—cyfinderairflow.Theficldismhrfion ofthevaporphaedifl'aedflommefiquidphase.Asfimeproceedc¢mmevaporphase developed in the central part of the cylinder and eventually a very intense snatified [96 chargedevelopcdncarmeinjeaor.1hereasonwhysuauficadmofmevaporphase masssbepistonapproachesitsleisnotweflmdastoodHoweverJtmaybfing benefitsmmefiielcothmprowsssmceasnafifiedchargemtmdthesparkplug cnnpreventmisfircswhichwutcsfirel. . 6.2.Fartherrecenunendations Although the application of the DMAel,4,6-TMN exeiplex system in a direct mjecumspukigninmmlsnenginemacnnlly“see”metanpordmdspafialbehavim oftheinjcctcdfinlsprayhasprovenflidtfimitrcquiresmorestudyforbeuer understanding. Firstly, the exeiplex fluorescence performance deteriorates at high mmpaanmAswmpaumemmflmrescencemtensitydecrcasesmdimpeak emission intensity wavelength is blue-shiflcd. Therefore, the cross-talk between the monomer and exeiplex emission increases, and the monomer fluorescence emission contains mavoidabie contributions flour the liquid phase fluorescence, and vice versa. Moreimpmumly,flumescenceemissimflommostorganicmolectdeamchasDMA, aremnnchedbyoxymmaeforeamnogencnvuemnentwasnccessuymmoaofom engineexpahnenmlhis-eusthstitwasnotpossrblemscethevapormdfiqiud characterisficsoffinlmanalcomhnflmenvummentthatrcqiureeoxygeanthe DMAel,4,6-TMN exeiplex systems, the normal boiling point of the fuel solvent, isooctane,ileO°Candthevaporphasemarker,DMA.isl94°C.1hedifl‘erent volafifiducmnediflamtwapmafimMsomncmlymmeenginecyclemeuseof DMAmaycauseanimdercstimateoffirelvaporconcentration. Also, any fuel distribution studies should be accompanied by an in-cylinder air flow study, using e.g., [97 MTV, PR! or LDV, for better understanding of fuel-air mixing processes: droplet atominfimvapodnfiommddiflirsimofthevapmphase.80mememsmaddressthese problemsmercmdrcdfmthebemermdersmdmgoftheairfinlmifingmus disur‘hrdomandevenmallyrcalcombustionbehaviorinanengine. APPENDIX Theappendixcontainstwomajorparts.0neisthetechnical information(A.l& A2)massisttheusamthe0perafionoftheeq1upmemmthissenrpandismeantasa mpplemcntmdreeqmpmentmmrnlsTheotheristhedetaflcdmfmmafionabomhow qimtitativeimageswerecalciuatcdinamotoredcnginecondition(A.3). A.l Equipment Connection for Laser Induced Fluorescence Imaging Theconnections fortheICCDdetcctorsystemandexcimerlaserforthelaser mdmcdflmrescencepmfionoftheexperunentmeshowninFig.84.1heComecfions showninFig.84areacombinationofCaselandCaseZinsection63.l,ICCDGating ExperimentsExamples, ofthePrincetonInstrmnentsmanuals.m Thetiminzdiasfamfor thisconfigwafimisshowanig.85.Theexcunerhseruiggastheprusegenemmr whichsendsahighvoltagcgatepulsetotheintensiflerintheICCDdetcctor.Thedelay bemecnmehsaniggermdgmemdseisadjmblesothatthedetcctmcanbegated aflerthelaseremissionocctna'l'he‘enable in’to‘notse.’ connectionpreventsthe arrayfi'ombeingexposcdduringreadoutbyblockingthegatepulses. The‘triggerout’ m‘extsym.’conmcumimfimesreadomwhendresystemisbdngextcrnauymggered andthecontoflerisbehgopuatedmtheexternalsynchrmizanonmode.1hedme Mgwhichdmisacqlurcdisdaaminedbythe‘expomnefime’setinthewfiware. Thenumberofexpoauesacamulatedbymedetcctmmyperreadomisdetammedby I98 199 theexposrnefimemdthelaserrepedfionrate.Foroneexponneperreadom,the expoaneflmecmbesetmmormelascrrepenflmmecanheadjustcdaccordingly. A.2 Operational Information for the Laser Induced Fluorescence Imaging SM ThegainoftheICCDdetcaormdthepulsewidthofthegateptdsegenaatorarc connoflcdbyten-nnnpmenfiomamhnmenmfineuwiththepomfiomemrsetfing. Expaimenmweredonemdeterminemegamandpdsewidthversuspotenflometer setfingTheICCDdetcctorgainisvariahlefi'omltolOOtimes.Table8smnmarizesthe average intensity gain versus potartiometer setting for the detector. Potentiometer setfingsarercadflom0m1000.ThegamversuspotennometersetfingfortheICCD detectorisshowninFig.86. ‘I‘hepulsewidthoftheFG-lOOpulsegeneratorisvariablefl'om18t0650nsin rangelofthevariablewidthmode.ATektronixllBOZoscilloscopewasuscdtomeasure theyflsefidfi.Table9mmuizcsthepmsewidmvmispmenfiometersetfing.Thc pdcnfiometercoflnwiththesetfingmdicamrmmkmnmusthennnedclockwisewhen seuingtheposifimThaeisaanaflmomtofphyinthecoflarwhichwfllafi‘ectme pinsewidthbyasmlrchasSOnaIfthepotentiometerisnnnedgentlytoi,000,thcpulse widthwiflbeGOOnaIffiglumrqmisappficdmmeknob,mepidsewidflrwiflmcrease m650nsAflotherposinonswcrenpcambleThepidseMdthverwspomnnometer settingfortheFG-lOOpulsegeneratorisshowninl-‘ig.87. Anothaexperimentwasconductedtodeterminefliegatepidsedeiaysetung required to initiate detector gating immediately after the laser laser emission pulse, yet 200 avoidgafingbefmelasercmissimhaddecnychhiswasnwessarymenmmatthe imageintensifiesmthecah’brafimwaedlnomymflmrescenceemissimmdnotduca hsaemissimltwmcmfiedmubyfomsingthedetcctmonapieceofduminumwhich scauucddrelasunflsewhileemitfingnegliubleflumwcencematwmfldobscmcme decayofthehsaemissimpube.1helasuptflsewidthisapproximamlyl8m.The shmteudaeamgmewbewidmdmofapproximamlyl8nswasuschhemaximum mtensityoftheuuageswasthenmonimrcdwhflevuyingthedehysetflnglhepeak imageinMsifiesocauredatadehypotennometersetfingof624.lnacasingor dcaeasingthedehyflom624readtcdmadecrcasemthemtensifies.1haef0tethe lasaemissimprusemdmedetcctorgatepuhewaewmpletelyoverlappedaéu. Theabsolutcdelaytimecouldnotbemeasured. However,thechangeindelay fimeforagivenchmgempotenfiomaasetfingcmddbemeaauedwimmeosdfloscope. methisitwasddamincdthatthedehyfimeincreaseslfinspamitchangeofthe potenflometersetfingDchyhgdiegatepulseanaddiflmalZOnsshouldplacethe leadingedgcofflnuteprflseZmaflamemilmgcdgeofmehsaemissimprdse.This n'anslatcdintoadelaysettingof637. Atthissflfinutheimageintensitiesindicatedthat meaningedgeofmeofmehsaptflseswbemgceughtbythegmeprdse.Thiswas probablyduetotriggcrjitterintheexcimerlaser. Adelaysettingof640clearcdthisup mdwasumdfmthemh‘brafimaangingfliegaepulsewidthdocsnmafl'eathedehy setfingrcquiredlheuaflingedgeofthegateprdsemoveswhenthepidsewidmu charged. The leadingedge remaim fixed relativeto the trigger. Table 10 smumarizes the gatepuhedelayfimevermpotenflometersetfingandtheFigme88showstheir relationship. 20l arp.drn2 386-PC Dell Connect 50100 :15” Pulschncrator sync“ 13 I4 extsync. morn STIBOCmOller m crflrlein denatu- unpavmidale m is: Figure 84. Connections for laser induced exeiplex fluorescence experiment)” 202 1.. .. 51 J 1 1 J T14 ml : .200v «Harms m Figine 85. Timing diagram for ICCD detector system configuration)” 203 TABLE 8: Average Intensity Gain versus Potentiometer Setting woman Setting IntensityGain 1000 100.0 990 86.86 980 76.35 970 66.84 960 55.91 950 4722 940 40.82 930 34.83 920 29.61 910 25.47 900 20.97 850 8.864 800 3.577 750 1.683 700 ~ 1.039 650 . 0.860 600 0.816 550 0.810 120 100 20 0 204 U'ITUIUIIlIIWTIIIIIII‘rrfi )- IrTTj—Ij‘rolIrIrI—rrTfTIU'II'rT—ITT'TW ALJALAJILI .1 j JJLJJJJIILLLLILLJLIIILIlLLL+JJllj 40050060070080090010001100 Weaning Figure 86. A plot ofgain versus potentiometer setting forthe ICCD detector. 205 TABLE 9: Gate Pulse width versus Potentiometer Setting PotentiometerSetting Pulsewidth(us) 1000 600-660 998 522 997 450 996 421 995 392 993 355 990 305 986 254 980 203 970 149 945 92 900 61.3 0 18.7 206 PuleewldtMne) § § § § IIWIIII‘WTTT—ITIII'IIIfiITIIIIITI 100 WLLLl—I—L—LLH- 0'" o L114an-n1L1L11114114+L44111L1, 880 900 920 940 960 98010001020 Pohntiometereetting Figure87. AplotofpuhewidthvamspmurfiometasetfingfortheFG-loowsegaicruor. 207 TABLE [0: Delay Time Versus Potentiometer Setting Ponnuunneuniksfing: [lass Thnc(hs) 0 76.1 50 164 100 237 150 329 200 402 250 493 300 566 350 657 400 730 450 821 500 894 550 986 600 1059 650 1150 700 1223 750 1314 800 1387 208 TrTIIIIUIIIUTIIUUV Figure“. Aplotofgnemdehyfimcvmmfimsafing. 209 A.3. Cellhntlon Procedure: I- ll-Cyllntler Measurements Thisappeodixdeem'beednedehiledinfmmafionabmflhowamfitafiveimagee mwmi.e.,eqtfivdeneenfiow.flmceinmnsityfmmevaporphaeemd mofthefiqfiddmpletmflmceinMsityforthefiqfidplmeinamamed (A) Liquid Phase Foam-dons «Wadhmhwflmqummmamm (1)1dmfialgwmeuyismedbetmnmedroplaeah‘bnfimexpahnentmdmem- cylhdafinlspnymmmeommdnegleaedthepreeenceofmequartzcyfinder. (2) Fluorescing volume is given by the thickness of the laser beam (800 pm) multiplied bytheinmrsectedarenoftheimidecylinder. (3)1heflmteecenceateaisasannedmbethemoftheindividualhodzonml linethat ismnnedtohavethesameeficiency. (4)Dropletsizeisdifi‘erentandindepeodent. (5)111: number density of the fluoreecing molecules in the liquid droplet is assumed to besmallwithintheintemectedareewiththelaserbeem. Thenthe in-cylindetliquidphase fuel dish-113mm: esa function ofcnnIt-angledegree (CAD)wasqmntifiedusingthedropletmass(M)vs. 0W intensitymteletion obtainedinChep.4aeehownbe1ow. l (eon-ts) 8656 (mum x M (11;) (56) Where 1 (counts) is the liquid (koplet ameenee intensity md M (pg) is the droplet musedintheeolibntion. 210 mqmdfiadonprxeluesasfdbw: (a) In-cylinder liquid phase fuel distribution as a function of crank-angle degree (CAD) wasobtainedinfigsqa). (b) Figue80(a)waseonvertedtoaniso-inmosityeonmmlnap. (c)Theiso-intensityeonm1umapwastheocmvmedmamassmmapusing5q. 56mdtherea1ltwssshowninFi381 ssafinctionofCAD. (B) Vapor Phase Feel Dlstrlhutioas Thein-eylindavapotphasefinldimhldonmafimetionofmk-mgledegtee (CAD) was quantified using the equivalence ratio (9) vs. fluorescence intensity (I) relationobtainedinChap. Sasshownbelow. I 9: «(Wham—MOW) (55) Wherelistheflmceintensity,¢isdleeqlfivalencerafio,mda(0)mfl(0)mthe fitting constant at given CAD (0) nequnttfledanpmdamasfoaow: (a) The ill-cylinder vapor plme fuel distribution as a function ofcnnk-angle degree (CAD) WIS obtained in Fig. 82(a). (b) Figure 82(a) was oonvertedto an iso-intensityoontom map. (c) Theiso-inmusitycontommapwuthenconvatedmmiso-eqldvdeocendocomom mapusingtheEq.(55)(withtheknownvaluesof intensity“):ndthoseo.(6)and 3(0) in Table 6 obtained as a filnction ofCAD). 211 (d) Ibequmfimfivereadtoftheiso-eqrdvalencemfiooonmmapofvaporphasefinl distribution is shown in Fig. 83. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 24. REFERENCES T.Johnen.M.PhI5“SpraymeadmObsuvafimmdleFflmDevelopmauMeunemamm thelntakeofaSparklgnition Engine.”SAE Paper950511.1995. M. Tabata. M. Kataoh. M. Fujimoto. Y. Noll. “in-Cylinder Fuel Distribution. Flow Field. and Combustion Characteristics of a Mixture Injected 81 Engine.” SAE Paper950104, 1995. .1. U. Kim, B. Golding. H. I. Schock. D. G. Nocera. P. Keller. “Exciplex Fluoreacmce VmaliznimSyuemsfwhe-ComhndeiaposisofmAmomofiwGasolmeEngine.”SAE Paper960826. 1996. J.U.Kim.W.KHmlLI.Schock.B.Golding.D.G.Nocerm"fheElecumDonor~Aeceptor Exciplex of N.N-dimethylmliline and 1,4,6-trimethylnaphthalene.” Chemical Physics Letters. 267. 323 (1997). LA. Melton,“UseofF1uorescence MethodsforMeasurementot‘DropletandV/apor'l‘emperamres.” unpublished paper, Chemistry Dep., University omeDallas. 1993. M. R. Wells. L. A. Melton. ASME Journal of Heat Transfer, 112. 1008 (1990). p T. R. Hanlon. L. A. Melton. “Exciplex Fluorescence Thermometry of Falling Hexadecane Droplets.” ASMB Journal of Heat Transfer. 114, 450 (1992). J.Zhng.L.A.Mehm.“P«enfialSyflemaficEnorsthpletTempaahneaObtahedby Fluorescence Methods.” ASME Journal of Heat Transfer. 115. 325 (1993). M. Winter. “MW of the temperature Field Inside a Falling Droplet.” “SS-AMERICAS. 4" annual Conference. Extended Abstracts. Hartford. C11, (1990). J. Zhang. L. A. Melton, “Ntnnerical Simulations and Restorations of Laser Droplet—Slicing Images.” Appl. Opt. 33, 192 (1994). T.Kadota. Y. Taniguchi. K. Kadowski.“ExciplexMethodforRemoteProbingofFuel Droplet Temperature.” JSME Int. 1., Ser. 11. 34, 242 (1991). ‘1'. Kadota. K. Miyoshi, M. Tsue, “Exciplex Method for Remote Probing of Fuel Droplet Temperature. Sheets of Ambient Conditions on Fluorescence.” JSME int. 1., Ser. 8.. 36, 371 (1993). G.0.Rubel,M.Seaver,“The‘1‘emperanlreofan EvaporatingWatermopletwithaMonolayer: A comparison ofExperiment and Theory,” Atmos. Ken. 31, 59 (1994). T.Yoshinki.1'. F1ma11ashi.l(.Nishida.l-i. Hiroyam. ’2—DMeasurementsot‘theLiquidPhase Temperature in Fuel Sprays.” SAE Paper 950461. 1995. '1'.Q.Ni,L.A. Melton. ”TemperaturelmpingingofLiquidAlkanesonHotSteel Surface.” unpublidled data (1993). KChilma. H. Baht‘fiuenchingofdml FluorescencesonyreneVaporbyl-lidl- Preenrre Oxygen or Nitric Oxide.” Chem. Phys” 25, pp. 299.306, 1977. R.A.Covelesltie, D. A. Dolson.C.S. Paramenter,“ADireetVie\voflntramolecularVibrational Redistribution in Slp-difluorobenune.” J. Chem. Phys. 72. pp. 5774-5, 1980. S. Hirayama. D. Phillips. “Vapor-Phase Exciplex Formed between 9.10-Dicyanoanthracene and Nonamine Additives.” J. Phys. Chem. 85, pp. 643-7, 1982. T. .1. Whitaker, B. A. Bmhaw, “Laser-Induced Fluoreacence Analysis of Vapor-Phase Pyrene.” I. PbyaCbem., 85.171). 2180-2. 1981. F. ZhaoetalnfiemporaldeyclicFlucmationofl-‘uel VaporConcenuation inaSparklprition Engine.” SAE Paper912346. 1991. M. EA. Bardsleyetm‘Q-DV'mnlizationofLiqtlidandVaporPuel mm [.C. Engine.” SAE Paper 880521. 1988. F. V. Bracco. MModelhgofEngineSprays.”SAEquer850394 1985. .SC.Johnston.C. W. Robinson.W.S. Rorke..LR.Smith.P..O Witne.“AwlicationofLaser Dragnosncstoan 1n;eetedEngine."SAEPaper790092. 1979. ‘1‘. M. Dyer, “New Experimental Teclmiquea for [n-cylinder Engine Studies," SAE Paper 850396. 1985. . w. r. Lyn. E. Valdmanis. “The Application of High Speed Photography to Diesel Engine Research.” I. Photographic Science. 10, 1962. 212 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 213 J. F. Alcoelt. W. M. Scott. “Some More Liflrt on Diesel Combustion.” 1. Mech. E. (Automobile Div.),pp. 179-191. 1962. M.Parks.C.Polonski.R.Toye.“PenetrationofDiesel FuelSpraysinGases.”SAEPaper660747. 1966. R. D.Reitz.F.V.Brscco,“Ulua1-lighSpeedFilmingofAtomizingRoldeets.” Phys. Fluids. 22. pp. 1054-1064.1979. K.-.LW11.C.C. SmRLSW.D.A.vaicca.F.V.Bracco.“MeaslaememoftheSpray Angle ofAtomiz'mg Jets.” 1. ofFluids Engineering. 105, pp. 4106-4111983. filfiroymT.Kadaa.M.Arfi.“leSprayChmtainfimmDiae1Engina”Symp.m CombtnfionModelinngeciprocafingEfllinfl-GmeralMaoraDeuoieMl. 1978. T.S1fiozaki.T.Sua1ki.M.Shhnoda.“ObservafionofCombmtionProceumD.L Enginesvial-lidl SpeedDirectsndSchlierenPbotogaphy.”SA£Pqer8m025, Diesel CombtntionandEmissions. Proceedings P-86, 1980. MMEMAMM.“FIBISprayMode1mgfmSwidChmbaDiadEngim”Plpet 1312,1CLASS-85. Imperial College. Londm. U.l(...luly1985. M.J.‘1"mda1.T.I.Williams.A.T.ilmcombeTlselSprayFormationinanExperimentalSwirl Chamber,” SAE Papa870451, 1987. P. Walkman W. P.Cmellieri.“Fuel lnjeetiondeombustion Hlenomenainal-ligh Speed D.l. Diesel EngineObservedbyMeansot‘Endoscopic HighSpeedeotognphy,”SAE Paper870097, 1987. K.-J.Wu.A.Cogbe.D.A.Santavicca,l-'.V.Bracco.“LDVMeas1nementsofDropVelocityin Diesel-type Sprays.” AIAA 1., 22. 1263-1270. 1984. K. Wandres. l-l. Weigand. J. Muller. [-1. Steinbichler. “High Speed Photography and Pulsed Laser Holography for Dispostic mmgafion of Miatln'e Formations and Vibration in Reciprocating Engines.” International Conference on Combufion in Engineering. 1, l. Mach. PM C79I83, 1983. A.O.zurLoye.D.A.Santavicca.F.V.Bracco.“PreliminaryStudyofFlameStrucmreinan1C Engine Using 2-D Flow Visualization.” lntnl. Symp. on Diagnostics and Modeling of Combustion in Reciprocating Engines (COMODlA 85). Tokyo. Illa-n- 1985. T.A.Baritaud.R. M.Green.“A2-DFlameVisualizstionTechniqueAppliedtothe LC. Engine.” SAEPaper860025, 1986. A.O.arrLoye, F. V. M'rmDimmfidemnlinfimofPremixed-Chugeflamem in a [.C. Engine.”SAE Paper870454, 1987. P. G. Felton. .1. Manama. M. E. A. Bardsley. F. V. Bracco. “2-D Visualization of Liquid Fuel Injection in an Internal Combustion Engine.” SAE Paper872074, 1987. G.M.Bmwn.J.C.l(ent.“FllmrescentLightSectionTeclmiqueforFuel Sprabearacterization.” ‘Flow Visualization 111’, Yang, WJ. (ed). Proceedings of the Third lntematiornl Symposium on FlorUVisualintion.1983. X.Li.R.S.kain.“AstudyofColddeombustionFlowArotmdBlufi’BodyCombustors.” Combustion Science and Technology. 52. pp. 173-206, 1987. S.C.lo1m:ton. “Ant-Experimental InvestiptiallntotheApplicationofSpontaneousRaman ScatteringtoSprayMeasurementsinanEngine.”Proc.ASMB. Fluids EngineeringConf. Boulder. CO., ASME. NY, pp. 10077-118. June 1981. S. C. Johnston. “Precombustion Fuel/Air Distribmion in a Stratified-Cup Engine Using Laser Raman Spectroscopy,” SAE Paper 790433, 1979. A. R.Chraplyvy,“NminmfiveMeaanemenuoquorConcenuafims1nsideSprays.” Appl. 096“. 20. pp. 2620-4, 1981. J.M.‘1'ishkofl‘.D.C.Jr.Hammond.A. R.Chrap1yvy,“DiagnosticMeas1nementsofFueISpray Dispersion.” I. of Fluids Eng, pp. 313.7, 1982. D.A.Hmsen.E.K.C.Lee.“RadiativeandNomadiativeTransitionsmtheFust ExcitedSingletState of Simple Linear Aldehydes.” J. Chem. Phys. 63, 3272. 1975. F. Beretta. V. Cincotti. A. D’Alessio. P. Menna. “Ultraviolet and Visible Fluorescence in the Fuel Pyrolsis Regions ofGaseous Difl'usion F1ames.”Comblnt. Flame 61. 211. 1985. A. LmB.Yip.R.K.l-ianson.“Acetone:atracerforconcentration meastn'ementsingaseous flows by planar laser-induced fluorescmce.” Experiments in Fluids 13. 369. 1992. 50. $1. 52. 53. 54. $5. 56. S7. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 70. 71. 72. 214 P. [-1. Paul. 1. Van Cruyninpn. R.K. Hanson. G. Kychfiofl. “Hifll Resolution Digital Flow field 1magingot’1ets.”ExperimentsinFluids9.241. 1990. P.Andresen.G.Meier.1LSchlma,lLVogeaA.Koch.W.HemscheLW.0ppamamE.Rothe. “Humucencelmag'nghddenlntuulCommEngineUsinngableExcimaLasa.” Applied Optics 29. 2392,1990. A.Arnold.l-1.Becker,R.Stmtz.P.Mor*house.J.Wolfitnn.R.Maly,W.Pfister.“Flame&ont imagingmmhnunflmbnsdmmgiubyhsa-Mmedfluamofmldehyde.”0pdcs Letters15,831. 1990. W.Lawrenz.I.Koh|er.F.Meier,W.Stolz.R.Wirth.W.H.Bloss.R.R.Maly,E.Waper.M.Zahn. WZDLlFMeunemenuofAidFulRmMgmelmfieSuokemaTmspmentSI Engine.” SAEPaper922320, 1992. TAWTAWWWMwWMPflFhaSlEnM’SAE Paper9223$5.1992. AMAWB.CmM.Decka.F.VmbeLV.SichLWolfim“Simm imagingofFuelandllydmaylRadicalsinmln-lineFom-CylinflrSlEngine.”SAEPapa932696. 1993. l-LNeij,B.Job-rsson.M.Alden.“DevelopmemandDema1strationoFZD-LIF forStudiesof MiaturePrepcationinSi Engines.”Combtnt. Flame 99, 449, 1994. J.Reboux.D. Puechberty,F. Dionnet.“ANewApproachoflearLaser1nducedFluorsscence AppliedtoFueUAirRafioMeaanememmtheCompressionSuokeofanOpdcalSL Engine.”SAE Paper941988. 1994. E.W‘mk1hofer.1-LPhiflipp.G.Fraid1.1£Fuchs.“FlnlmdFlamelmaginngIEngine.”SA£Paper 930871.1993. LAMeMWWfiWEmisfimsforDiaelleDrwleudeapor." W-Ofianwom“. 1983- L. A. Melton. J. F. Verdieclr. “Vapor/Liquid Visualization in Fuel Sprays.” Twentieth Symp. (1ntemational)onCombustion.‘l'beCombustionlnstiulte.pp. 1283-1290, 1984. 1.. A. Melton. J. F. Verdiwk. “VaporfLiquid Vindication for Fuel Sprays.” Combust. Sci. and Tech. 42. pp. 217-222. 1985. A.A.Rntunno.L.A.Mehmfi¢WC¢hafimproce¢mforExciplea~baedVapodLimfid Visualiution of Fuel Spray,” Combust. Sci. and Tech, 71, 247, 1990. M.E.A.Bards1ey,P.G.Fehon.F.V.Bracco.“2-DVisualiaationofaHollow-Cone8prayinaCup- in-Head. Ported. [.C. Engine.” SAE Paper890315, 1989. P.G.Felton.F.V.Bracco.M.E.A.Bardsley,“OntheQuantituiveApplicationofEaciplex Fluorescenceto Enm'neSpays.”SA1-2 Paper930870. 1993. LAMeMerUmofExciplex-hdeqlodLimdqumlindeynems(AUsus MmmlL”FmflRepmtAmyRaeuchOficeCmactDML03—86-K-0082.1988.(N115wder8 ADA191649). LB.Berlman.“FluorescenceSpectraofAromatic Molecules,”2ndedition.AcademicPtess.New York. 1971. 1.. A. Melton. “Exciplex-Based VapodLiquid Visualization Systems Appropriate for Automotive GaolmumUsastubedeepaeTeaasmghaEducafimCocdimdngBoudEnergy ResearchApplieationsPromConuactln‘l, I993. RShimiNQaLWofAh-FulmmmawmefinfineunnngfiF Teclmique,”SAEPaper922356.1992. .D.H.Shack.W.C.Reynolrh.“AppflcafionofPuficleTrackingVelocimeuymtheCychc VuiabilityofthePre-CombustionFlowFieldinaMotoredAxisymmen-ic Engine.”SAEPaper 910475.1991. D. L.Retm.R.J.Adrin.C.C.Lllheth.D.T.French.T. D. Fansler. “InstantaneotnPlanar MementofVeloeityndhrngcdeVadchymdSuianmmEngineUsmgPardcle imageVelocimetry." SAEPaper890618, 1989. “Molecular‘l‘agging VelocimetryMl'V),” inthewwwsite.l997. M.M. Koochesfahani. D..G Nocera. “Molecular Tagging Veloeimetry: A review of Techniques. ChemicalConceptnandApplication.” Expenmenthlulds. 1997 invitedpaper. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82 83 84 85 88 91 92. 93. 94. 95. 96. 97. 215 M.M. Koochesfahani. R.l(. Cohn. C.P. Gendrich. D.G. Nocera. “Molecular Tagging Diagnostics for meSMyofKimmadamdMgmLiqmdPhufim”DevdopmmmusaTxmuquesm Fluid Mechanics. Springer-Vang: Berlin. 1997, in press. REFflcoJlG.Nm“anfitmiveMulfipohnMeunememdemnliudmofDmseSolid- LiquidFlowsbyUsingLaserlnducedeotoehemicalAnemomeu'yalPA).” Butterurorth- Heinemarm: Boston.l992;Chapter3. p59. C.P. Gendrich. M.M. Koochesfahmi. 0.6. Nocera. “Meleequ Tawng Velocimetry and other NovelApplicationsofsNewaospborucentSmolectdea.” ExperimentinFluids.1997. inpress. C.P. Genrh'ich. M.M. Koochesfahani. 0.6. Nocera. “Analysis of Molecular Tagging Velocimetry [mapsforMSimmMulfipothdoeityVecmrszulLAmPhysSoe 1994.39. 1980 B.Sda.NLM.chhafithD.G.NocagFU.Schoch“MdsulcTaggingVebcimeUymGas PhaseFlow.”Bull.Am.Phys.Soc.1995,40.2000. B.Sda.M.MchbeafithD.G.NomFU.Schxh“VdodtmemaSwadyflow Model ofm 1C Model using MolecularTagging Veloeinntry,” Bull. Am. Phys. Soc. 1997. 41.1804. RobertA.Albelty.. “PhysicalChemistry.”6thedition,.lohn Wileyt Sons.pp.69$-696.1983. Lakowic¢.J.R., “Prmciplesot'Fluorescence Speetroscopy.”P1emnn: New York. 1983;Chapter9. W. M. Vaufim. G. Weber, Biochem. 9. 13464-473, 1970. C. N. Banwell, “Flmdamentals of Molecular Spectroscopy,” Berkshire. England: McGraw-l-lill.. 1966. K1.Cattolica.S.R.Vosen.'Two-Dimensional Mmentsoftheml'flinaConstantVolume Combustion Chamber.” Twentieth Symposium (Internmional) on Combustion. Pittsburgh: The Combustion Institute pp. 1273-1282. 1984. KP.Gross.R.L.McKenzie. P.Lom“MeasurementsofTemperatme.Density.Prm1re.and Their Fluctuations in Supersonic Tmbulence Using Laser induced Fluorescence.” Exp. Fluitb. Volume 5. pp. 372-380, 1987. B.E.Banleel.M.Seiumm.RK.Hmsm.“QmfinfiveP1mLua-hdwedfluaescence lmagingofRadical SpeciesinHighPressureFlames.”AIAAPaper94—0229.32“Aerospace Sciences Meeting. Reno. NV.. 1994. Kobayamietal.,“ApplicationofUVandVUVExcimerLasersinCombustion Measurements Using Enhanced Raman Scattering.” Laser Diagnostics and Modeling ofCombustiat. Stringer-Veda; pp. 133-140. 1987. WKohse—HomghauaneLaser-lndwedflmcezmehcent Developments in Combustion Diagnostics.” Appl. Phys. B 50, 455-461. 1990. P.Andresen.A.Bath.W.GrOgereta1..“Laser-lnduced Fluorescencewithumableexcimerlasersas a possrhle method for inmraneous temperature field measurements at high presnne: Checks with an atmospheric flame.” Applied 0P5“- Vol. 27, No. 2. 1988. P.Andresen.GerardMeijer. HaraldSchliheretal., “Fluorescence Imaging lnsidean internal Combustion Engine Using Tlmable Escimer Lasers." Applied Optics. Vol. 29, No. 16. 1990. A.C. Eckbreth.,“1.aserDiagnostieaforCombustion TemperatureandSpecies.”Abacus Press. TImbridge Wells. UK. 1988. R.l(. Hanson. “Twenty-First lntl's Symposium on Combustion.” The Combustion Institute. pp. 1677,1986. C. Km.“Recmtadvmcesmthopletvapmiztionmdcomhmion.”ngEnagyComm Sci. 8. 171. 1982. W.A.Sirignano.“Fueldrop1etvaporindonmdspmycombtndontheory.”ng. EnergyCombust. Sci. 9, 291. 1983. G. M. Faeth. “Evaporuion and combustion ofsprays.” Pro; Energy Combust. Sci. 9. 1. 1983. 6.1.W6.Mmefije“hvesfipfimmtomeVapmiafimofhdividmlSolmePuficlesm Flame Spectrometry.” Analytical Chemistry. Vol. 46. No. 7. 901. 1974. C. Th. .1. Alkemade. “Flame Emission and Atomic Absorption Spectrometry. Vol. l-Theory.” J A. Dean.T. C. Rains. Ed..Marcel Deldrer. New York. N. Y..p101.1969. F. A. Williams. “Combustion Theory.” Addison-Wesley. Reading. MA.. 1965. 216 98. R. B. Bird. W.RSMEMLimWrmsponfllenomene”PanIohaney.NewYmk. 1960. 99. L. LeeWComhasfimmdhopulfimCdbquim’AGARD.PagmnmPress.Newahp 451-497. 1958. 100.F. T. Gamer. Trans. Instr. Chem. Eng. Land. 28. 88-96. 1950. 101.8. L.WakiLM.M.Priem.R.I.Brflrowski.1LJ.MeyaaO.AUyehma.“Experimenmlmd caladatedtempaahnemdm.ldstoriesofvaporizingfiseldmp.”NACATN3490, Wahington. DC. 1956. 102.6. M. Faathh-acteriniesofblendedmonnpmpdlamrhuplets.”AIAA Paper70-77. 1970. 103.6. M. Faeth. R. S. Lazar. AIAAJ. 9. 2165-2171, 1971. 104 IN. Dunn. “Excited State Lifetime Meunement.” Academic Press. pp. 62. 1983. 105 A.M. Murray. LA. Melton. “Flumwence Methods for Determination of Temperature in Fuel Sprays.” Applied 011613. Vol. 24, No. 17. pp. 2783-2786. 1985. 106 M. Gordon. W.R. Ware, 'lhe Exciplex.” pp. 1-21, Academic Press. New York. 1975. 107.SJ. I-Iale. L.A. Melton” “Absolute Quantum Yiela for Exciplex Fluorescence.” Appl. Spectra. 44. 101. 1990. 108.L. A. Melton. “Exciplex-Based VqlodLiquid Visualization Systems Amroprime for Automotive Gasolinea.” Appl. Spect.. 47, 782. 1993. 109.“!1andhookot'Dataon0rgmicCompotmds.”3rdedition. Boea Raton. Fl: CRC press.c1994. 110.Y.-g.K.Shin.V.M.Miskowski.D.6.Nocera.InorgChem..29.2308.1990. 111.N. I. Turro. Modern molecularphotochemistry(Benjamin Cummings). Menlo Park. CA. 1978. 112.S.L.Murov.I.Carmichael.6. LI-Iug.eds. HandbookofphotochemistryMarcel Dekker. New York. 1993) p. 176. 113.V. Balzani. L. Maggi. M. F. Manfiin. F. Bolletta. Coord. Chem. Rev. 15. 321. 1975. 114.8. L. Murov. I. Carmichael. 6. 1.. Bug. Handbookofphotochemistry. 2nd edition. Marcel Dekker. New York. 1993. Ch.1. 115.1. E. Bowman. K. A. Bergllmd. D. G. Nocera. Rev. Sci. Instrum. 64(1993) 338. 116.P. A. Manic. R. C. Only. I. L. R. Williams. S. Farid. “Effect of Polymeric Matrices and Temperatures on Exciplex Formation.” Polymer Sci. (Polymer Lett.) 15. 295 (1977). 117.A. Weller. Z. Phys. Chem. 133 (1982) 93. 118.B. Stevens. Adv. Pbotochem. 8(1971)161. 119.14. Helsen. L. Viaene. M. Van DerAuweraer. F. C. Deschryver. .1. Phys. Chem. 98 (1994) 1532. 120.6. S. Bahn. “Role of Vaporization Rate in Combustion of Liquid Fuels.” p. 104. in the Literattn'e of The Combtstion of Petroleum AC8. America Chemical Society, 1958. 121.W. D. Bachalo. M. I. Houser. “Phase Bowler Spray Analyzer for Simultaneous Measurements of Drop Size md Velocity Distributions.” Opt. Eng, 23. 583 (1984). 122.P. Massoli. F. Beretta. A. D’Alessio. M. Lamro. “Temperature and Size ofSingle Transparent Droplets by Light Scattering the Forward and Rainbow Regions." Appl. Opt. 32. 3295 (1993). 123.Zhang. “Fluorescence Methods for Determination of Temperature in Aerosol Particles.” Ph.D dissertatirm. Mech. Eng. Dept.. University of Nebraska-Lincoln. 1991. 124.]. [-1. Stuffiebemn . “Exciplex Fluorescence Thermometry of Liquid Fuel." Appl. Spectrosc. 43. 274 (1989)- 125.0.Arce.“l'emperanneandSieeMeamrementsofFree FallingDecaneDroplets.”M.S.tbesis. University ofTertas at Dallas. (1993) 126K. F.Schrum.A. F. Williams.S.A.I-Iaether. D.Amotz,“MolecularFluorescenceThermomeny.” Anal. Chem. 66. 2788. (1994). 127.“Mm1uals." Princeton Instrument. Inc. 1991. 128.“User’s manual." CSMA software version 2.4.. Princeton Instrunents. Inc. 1994. 129.V. Wylen. R. Sonntag. C. Borgiakke.“F1mrhmentalsofC1asm'cal Thermodynamics" 4‘ed.. pp 49. 1993. 130.R. C. Weast. “CRC Handbook of Chemistry and Physics”. 64‘ ed.. CRC press. 1983. 131.M. E. A. Bardsley. “Two Dimensional Visualization ofFuel Sprays in an Internal Combustion Engine.”M.S.Tbesis.Mech. EngDepLPrincetonUniversity.1990. 217 l32.“Catalog Handbook of Fine Chemical.” Aldrich Chemical Compuly. Inc. 1994-1995. 133.J.D. Ingle. S.R. Crouch. “Spectrochemical Analysis.” Prentice-Hall. Englewood Clifi's. New Jersey. 1988. I34.T. D. Fmsler. D. T. French. M. C. Die. “F131 Distrl'btaions in a Firing Direct-Injection Spark- Ignition Engine Using Laser-Induced Fluorescence Imaging.” SAE Paper 950110, 1995. 135.Private discusion with T. D. Fmsler md D. T. French 136.R.M.Frank.J. B.Heywood.'TheImpatanceofInjectionSynemChmactefisdcsonHy¢ocarbon Emissions liom a Direct-Injection Stratified-Clings Engine.” SAE Paper900609.1990. 137.N. MiyalnotmflomT.Shud0.F.T¢eyama. “Combusfiondemism‘onsinaNewConceptDl Stratified Chn'ge Engine with Two-Stage Fuel Injection.” SAE Paper 940675. 1994. 138.P.I-I.Cunpbell.l(. M.Sinko.B.Chehroudi.“LiquidandVaporPhueDisnib1nionsinaPiloted Diesel F131 Spray,” SAE Paper 950445. 1995.