Experimental and computational investigation of autoignition of jet fuels and surrogates in nonpremixed flows at elevated pressures

(1)

Proceedings of the Combustion Institute 37 (2019) 1605–1614

www.elsevier.com/locate/proci

Experimental

and

computational

investigation

of

autoignition

of

jet

fuels

and

surrogates

in

nonpremixed

flows

at

elevated

pressures

Gerald Mairinger

a

, Alessio Frassoldati

b

, Alberto Cuoci

b

,

Matteo Pelucchi

b

, Ernst Pucher

c

, Kalyanasundaram Seshadri

a ,∗ a_Department_of_Mechanical_and_Aerospace_Engineering,_University_of_California_at_San_Diego,_La_Jolla,_CA_92093-0411,

USA

b_Dipartimento_di_Chimica_Materiali,_e_Ingegneria_Chimica_“G._Natta”,_Politecnico_di_Milano,_Piazza_L._da_Vinci_32,

Milano20133,Italy

c_Institute_for_Powertrains_and_Automotive_Technology,_Vienna_University_of_Technology,_A-1040_Vienna,_{Gusshausstrasse}

27–29/315,Austria

Received1December2017;accepted27June2018 Availableonline5October2018

Abstract

Experimentalandcomputationalinvestigationsarecarriedouttoelucidatethefundamentalmechanisms of autoignition of surrogatesof jet-fuels atelevatedpressures upto6bar.The jet-fuelstestedareJP-8, Jet-A,andJP-5,andthesurrogatestestedaretheAachenSurrogatemadeupof80%n-decaneand20% 1,3,5-trimethylbenzenebymass,SurrogateCmadeupof 60%n-dodecane,20%methylcyclohexaneand 20 %o-xylenebyvolume, andthe 2ndgeneration Princeton Surrogate madeupof 40.4%n-dodecane, 29.5%2,2,4-trimethylpentane,7.3%1,3,5-trimethylbenzeneand22.8%n-propylbenzenebymole.Using thecounterflowconfiguration,anaxisymmetricflowofagaseousoxidizerstream,madeupof amixture of oxygenandnitrogen,isdirectedoverthesurfaceof anevaporating poolof aliquidfuel.The exper-imentsareconducted atafixed valueof massfraction of oxygen intheoxidizerstream andatafixed valueofthestrainrate.Thetemperatureoftheoxidizerstreamatautoignition,Tig,ismeasuredasa

func-tionof pressure,p.Experimentalresultsshow thatthecriticalconditions,of autoignition of the surro-gatesareclosetothatof thejet-fuels.OverallthecriticalconditionsofautoignitionofSurrogateCagree bestwiththoseofthejet-fuels.Computationswereperformedusingskeletalmechanismsconstructedfrom adetailedmechanism.Predictions of thecriticalconditionsof autoignitionof thesurrogatesarefound toagreewellwithmeasurements.Computationsshowthatlow-temperaturechemistryplaysasignificant roleinpromotingautoignitionforallsurrogates.Thelow-temperaturechemistry,of thecomponentofthe

∗_{Corresponding}_author.

E-mailaddress:seshadri@ucsd.edu (K.Seshadri). https://doi.org/10.1016/j.proci.2018.06.224

(2)

1606 G.Mairingeretal./ProceedingsoftheCombustionInstitute37(2019)1605–1614

surrogatewith thegreatestvolatility,wasfoundtohavethemostinfluence onthecriticalconditionsof autoignition.

ThisisanopenaccessarticleundertheCCBYlicense.(http://creativecommons.org/licenses/by/4.0/ )

Keywords: Autoignition;Surrogatefuels;Jetfuels;Lowtemperaturechemistry;Kineticmodeling

1. Introduction

Improved understanding of the combustion of jet-fuels at elevated pressures is essential for accurate predictions of chemical processes tak-ingplaceinair-breathingpropulsionsystems. Jet-fuelsaremixturesofnumerousaliphaticand aro-matic compounds. It has been established that auseful approach to obtainfundamental under-standingof combustionof jet-fuelsistofirst de-velopsurrogatesthatreproduceselectedaspectsof combustion of thesefuels [1] .Numerous experi-mental,computationalandanalyticalstudieshave addressed combustion of jet-fuels, surrogates of jet-fuels,andhigh-molecularweighthydrocarbon fuelsatelevatedpressures[2–8] .Thesestudies in-clude measurements and prediction of combus-tionprocessesinhomogeneoussystemsatnormal andelevated pressures[2–5] ,extinction of coun-terflow nonpremixed flames [7] , autoignition in nonpremixedflows[6,9–11] ,andlaminarpremixed flames[12] .Thereare,however,verylimited stud-iesonautoignitionofjet-fuelsinnonuniformflows andat elevated pressures. The present study ad-dressesthisdeficiency,andisfocusedonelucidating themechanisms of autoignition of surrogates of jet-fuelsatelevatedpressuresupto6bar. Autoigni-tionisanimportantpropertythathasasignificant influenceonperformanceofcompression-ignition engines. The present studies are carried out in nonpremixedsystemsbecauseitbestdescribes au-toignitionandcombustionincompression-ignition engines.

Experimentalandcomputationalinvestigations were carriedout recentlyto elucidatethe funda-mental mechanismof autoignition of n-heptane,

n-decane, and n-dodecane in nonpremixed flows atelevatedpressures upto 6bar [8] .Theresults showedthatn-dodecaneismosteasytoignite fol-lowed by n-decane and n-heptane. This was in agreement with previous experimental and com-putational studies at 1 atm where a similar or-der of reactivities for these fuels were observed at low strain rates [11,13] . A noteworthy find-ing was that low-temperature chemistry played a dominant role in promoting autoignition [8] . The influence of low-temperature chemistry was found to increase with increasing pressure [8] . Thepresentstudyextendsthistojet-fuelsandits surrogates.

2. Experiment

Experimental and computational studies are carriedoutemployingthecounterflow configura-tion whereina cylindricalduct is placed 12 mm aboveanevaporatingpoolofaliquidfuel.From thisductanaxisymmetricflow of agaseous oxi-dizerstream,madeupofamixtureofoxygen(O2)

andnitrogen(N2),isdirectedtowardthesurfaceof

theliquidfuel.Attheexitof theduct,the magni-tudeoftheinjectionvelocityisV2,thetemperature

T2,thedensityρ2,andthemassfractionof

oxy-genYO₂,2.Heresubscript2represents conditions

attheoxidizer-boundary.Thetemperatureatthe liquid-gas interfaceisTs,andthe massaveraged

velocityonthegassideoftheliquid-gasinterface, asaresultofevaporationoftheliquidfuel,isVs.

Heresubscriptsrepresentsconditionsonthegas sideoftheliquid-gasinterface.Astagnationplane isformedneartheliquid-gasinterface.Ithasbeen shownpreviously[11] thattheradialcomponentof theflowvelocityattheliquid-gasinterfaceissmall andcanbepresumedtobeequaltozero.Athin boundary-layerisestablishedattheliquid-gas in-terfacein theasymptoticlimitof largeReynolds number,calculatedusingV2,thekinematic

viscos-ityof theoxidizerstreamattheexitof theduct, andthedistance,L,between theexitof theduct andtheliquid-gasinterface.Theinviscidflow out-sidethisboundary-layerisrotational,andthelocal strainrate,a2,intheinviscidregionevaluatedatthe

stagnationplane,obtainedfromintegrationof Eu-ler’sequations,isa2=2V2/L[11,14] .

Allgaseousstreamsarecontrolledbycomputer regulated analog mass flow controllers. A Pt-Pt 13 % Rhthermocouple with a wire diameter of 0.127mmandajunctiondiameterof0.21mmis usedto measurethe temperatureof theoxidizer stream, T2. The measuredtemperatures are

cor-rectedtoaccountforheatlossesbyradiationfrom thetheromocouplewires[15] .Fuelissuppliedtoa fuelcupbyasyringepumpwithavolumetricflow accuracyof ± 0.01mL/min.The temperatureof thefuelenteringthecup,Tc,ismeasuredbya

ther-mocouple.Furtherdetailsof theburnerincluding theprocedureemployedtomaintainthelevelofthe liquidfuelinthecuparedescribedindetail else-where[7,8] . Experimentsareconducted with the counterflow burner placed inside theHigh Pres-sureCombustionExperimentalFacility(HPCEF)

(3)

Fig.1. Highspeedphotographofatypicalautoignitioneventatp₌4bar,YO2,2=0.15,a2=138s−1,andT2=1101K.

ThefuelisJet-A.

described elsewhere [7,16,17] . The reactive flow fieldischaracterizedbythevaluesofYO₂,2,T2,a2,

andpressure,p.Theexperimentswereconducted withYO2,2=0.15 anda2=138−1s. The

tempera-ture of the oxidizer stream at autoignition, Tig,

was measuredas afunction of p. Thevalues of

YO2,2anda2wereselectedtominimizeratesofsoot

formation.

Theprocedureformeasuringcriticalconditions ofautoignitionisasfollows.Firstthedesired cham-berpressureisestablishedbyintroducingnitrogen intothechamber.Watercoolingsystemsare acti-vated. The flowfieldisestablished attheselected valuesofYO2,anda2andatavalueofT2thatisless

thanTig.Liquidfuelisintroducedintothefuel-cup

andthesyringepumpisusedtocontrolitslevel. Thetemperatureof theoxidizerstreamis gradu-ally increasedinsmallincrements,allowing suffi-cienttimeforthesystemtoreachsteady-state,until autoignitiontakesplace.Theonsetofautoignition isrecordedusingahighspeedvideocamera oper-atingat1000framespersecond.Figure 1 showsthe progressionoftheautoignitioneventatsuccessive values of thetime,t,recordedbythehigh speed camera.Thefirstimageisthestartafterthecritical conditionshavebeenreached,thesecondimage,at 3.0ms,showsafaintglowwheretheflameisseen asasmallcirculardiscaroundtheaxisof symme-try.Inthesubsequentimagesthediameterof the flameincreases.Therepeatabilityofthe

measure-mentsofTigis± 5K.Theaccuracyisestimatedto

be15Kandisattributedtouncertaintiesin eval-uatingthecorrectionthatmustbe appliedtothe measuredtemperatureduetoheatlossesfrom ra-diationfromthethermocouple.Onlythose experi-mentaldatawhereautoignitionisobservedtotake placeclosetotheaxisof symmetryisincludedin thedataset.Thesemeasurementsareperformedfor differentvaluesofp.

3. Fuelstested

Thejet-fuels tested areJP-8,Jet-A,andJP-5, andthe surrogates tested arethe Aachen Surro-gate,SurrogateC,andthe2ndgeneration Prince-tonSurrogate.TheAachenSurrogateismadeup of80%n-decaneand20%1,3,5-trimethylbenzene bymass(H/C2.00).Thissurrogatehasbeenused instudiesdescribedinRef.[7,9,10,18–20] .It accu-ratelyreproducedcriticalconditions of autoigni-tion of JP-8 at nonpremixed conditions at at-mospheric pressure [9,10] , critical conditions of extinctionof JP-8 atpremixedandnonpremixed conditionsatatmosphericpressure[9,10,20] ,and sootformation atnonpremixed conditionsat at-mospheric pressure [9] . Kinetic modelling stud-ies have been carried using detailed chemical-kineticmechanisms[9,10,20] .SurrogateCismade upof 60%n-dodecane,20%methylcyclohexane

(4)

1608 G.Mairingeretal./ProceedingsoftheCombustionInstitute37(2019)1605–1614 and 20 % o-xylene by volume (H/C 1.92). This

surrogate has been used in studies described in [10,19–21] .Ithasbeenfoundtoaccurately repro-duce at atmospheric pressure, critical conditions of autoignition of JP-8 at nonpremixed condi-tions[10,21] ,andcriticalconditionsofextinction of JP-8atnonpremixedandpremixedconditions [10,20,21] .The2ndgenerationPrincetonSurrogate ismade upof 40.4% n-dodecane,29.5 % 2,2,4-trimethylpentane, 7.3 % 1,3,5-trimethylbenzene and22.8%n-propylbenzenebymole(H/C=1.96). It has been used in studies described in [2,3,7] . It has been found to accurately reproduce vari-ousaspectsofcombustionofjetfuelsthatinclude ignition delay times measured employing shock tubesandrapidcompressionmachines[2,3] , lami-narburningvelocitiesofpremixedflamesand crit-icalconditionsofextinctionof diffusionflameat atmosphericpressure[2] ,andcriticalconditionsof extinctionofJP-8atmoderatepressure[7] .

4. Formulation

Kineticmodellingdescribedhereiscarriedout using a skeletal mechanism made up of kinetic stepsselectedfromthePOLIMIdetailed chemical-kinetic mechanism [22] that describes pyrolysis, partialoxidationandcombustionofkeroseneand aviationfuels.Thisskeletalmechanismismadeup of 231speciesandhasbeenemployedpreviously tocharacterizecombustionofmanysurrogate mix-turesproposedforkerosene [22] .Reference com-poundsinkerosenesurrogatestypicallyincluden -alkanes,iso-alkanes,methylcyclohexane,aromatics (from toluene up to C9 aromatics), decalin and

tetralin.Thekineticstepsforthesecompoundsare includedinthekineticmechanism,andhavebeen tested[22] .Thekineticsubsetfor methylcyclohex-anewas updatedfromthatin[23] in viewof re-centexperimentalmeasurementsof ignitiondelay timesinrapidcompressionmachinesathigh pres-sureandlowtointermediatetemperatures[24] . De-tailsoftherevisionandtestsofthepredictionsof theupdatedkineticmechanismfor methylcyclohex-aneareprovidedasupplementarymaterial.

The computations were performed with the OpenSMOKE++code[25] .Thestructureof the reactiveflow-fieldisobtainedbysolvingthe con-servationequationsof mass,momentumand en-ergy,andthespeciesbalanceequations.Attheexit of the duct,thevalues of V2,andthemass flux

ofO2andN2arespecified,andtheradial

compo-nentoftheflowvelocityispresumedtobeequalto zero.Attheliquid-gasinterface,mixedboundary conditionsareappliedforthespeciesbalance equa-tions,andtheenergyconservationequation[8,13] . Thetemperatureattheliquid-gasinterfaceis ob-tainedusing Raoult’s Law[26] .Empirical coeffi-cientsforcalculatingthevaporpressure, andthe heat of vaporization, forthesefuels aregivenin

[27] .Simulationsareperformed,withYO2=0.15,

a2=138s−1,andforpbetween 3barand6bar,

using a computational grid with more than 400 grid points to ensure grid insensitive results. A convergednon-reactive(cold) solutionisfirst ob-tainedforT2=300K.Thetemperatureofthe

ox-idizerstreamisincreasedatarateof10K/suntil autoignitiontakesplaceandahotflameis estab-lished. Numericalanalyses showed that the pre-dictedTigisnotaffectedbytherateof

tempera-tureincreaseifitdoesnotexceed100K/s.Forthe given value of p, autoignition is defined to take placeatthevalueofT2=Tig,whereanabrupt

tran-sitiontakesplacefromaweaklyreactiveregionto aflame.

5. Results

Figure 2 showsthe measured temperature of theoxidizerstreamatautoignition,Tig,asa

func-tionof p,forJP-8, Jet-A,JP-5,theAachen Sur-rogate,SurrogateC,andthePrincetonSurrogate. Forallfuels,Tigdecreaseswithincreasingpressure.

ThevaluesofTigforSurrogateCandthe

Prince-tonSurrogateagreebestwiththoseforthejet-fuels atpressuresaround3barwhileatpressuresclose 6bar,the valuesof Tig forSurrogate Candthe

AachenSurrogateagreewellwiththoseforthe jet-fuels. Overallthe critical conditionsof autoigni-tionof SurrogateCagreebestwiththoseforthe jet-fuels.

Figures 3 ,and4 showthepredictedaxial tem-peraturedistribution intheflow-fieldasa conse-quenceof gradualincreaseinthevalueof T2.In

these figures results are shown at p=6bar, be-causeatthispressuretheroleoflow-temperature chemistry and the negative temperature regime aredisplayedbest.Moreover atthispressurethe values of Tig for the surrogates differ the most.

Figure 3 showsthat forT2 around700 K,there

is a small, but sudden, increase in the value of thetemperatureof theflow-fieldnearthesurface of theliquidpool,especiallyfortheAachen Sur-rogate (marked asLT in Fig. 3 ), indicatingthat low-temperatureignition hastaken place. Thisis followedbythenegative-temperaturecoefficient re-gion(NTC)and“hot”-ignition.

Figure 4 showsthetemperatureincrementT [K]andtemperatureattheliquid-gasinterface,Ts,

asafunctionofT2 forthesurrogates.The

quan-tityTisthedifferencebetweenthepeakvalueof thetemperatureinthemixinglayerandT2.Forthe

AachenSurrogate,asT2isincreased,aweakly

reac-tiveregionappearsinthevicinityof700K,thatis markedbyanincreaseinthevalueofTfollowed byadecrease.FurtherincreaseinT2leadstoarapid

riseinT.Thisbehaviorissimilartothatshownin Fig. 3 .Thetemperatureriseintheweaklyreacting regionisindicativeoflow-temperatureignition fol-lowedbytheNTCregionandfinally“hot”-ignition

(5)

Fig.2. Themeasuredtemperatureoftheoxidizerstreamatautoignition,Tigasafunctionofpressure,p,ata2=138s−1,

andYO2,2=0.15.Thesymbolsrepresentexperimentaldataandthelinesbestfittotheexperimentaldata.Theerrorbar

shownononeexperimentaldataappliestoall.

Fig.3. Predictedaxialtemperaturedistributionintheflow-fieldofthesurrogates,asaconsequenceofagradually increas-ingthetemperatureoftheoxidizerstream.Thecalculationsweremadeatp=6bar,YO2,2=0.15,anda2=138s−1.

markedbylargeincreaseof T.Thehigher reac-tivityoftheAachenSurrogatewhencomparedto theother surrogatestestedhere isaconsequence ofitslargeparaffiniccontentnamelyn-decanethat wasfoundinpreviousstudies[8] toexhibit signifi-cantlow-temperaturechemistry andNTC behav-ior. Figure 4 showsthatTs forallsurrogates

in-creaseswithincreasingT2anditsvalueislessthan

their respectiveboiling point temperature, about 30 Klowerattheonset of low-temperature igni-tionand20Kattheonsetof“hot” ignition.The increaseinthevalueofTswithincreasing

temper-ature of theoxidizerstream isaconsequenceof

thebalancebetweenthevapor-pressureandrates ofvaporization.Raisingthetemperatureofthe ox-idizerstreamenhancestheheatfluxtotheliquid poolthatleadstohighervaluesofTsandtherates

ofevaporation.TheboilingpointsfortheAachen Surrogate,SurrogateC,andthePrinceton Surro-gateatpressureof1baris446K,411K,and412K, respectivelyandatapressureof6bar,itis532K, 501K,and502K,respectively.Forthejetfuels JP-8andJet-Athevalueoftheboilingpointat1baris roughly500K.Thepresenceofcoolflames,NTC behaviorandtwo-stageignitionwasalsoobserved inpreviousexperimentalandcomputational

(6)

stud-1610 G.Mairingeretal./ProceedingsoftheCombustionInstitute37(2019)1605–1614

Fig.4. TemperatureincrementT[K],temperatureattheliquid-gasinterface,Ts,asafunctionofthetemperatureofthe

oxidizerstream,T2,forp=6bar,YO2,2=0.15,anda2=138s−1.

Fig.5. Mole-fractionofhydrogenperoxide(H2O2),XH2O2,andmole-fractionofketo-hydroperoxide(KET),XKET,as

functionsofthetemperatureoftheoxidizerstream,T2,forp=6bar,YO₂,2=0.15,anda2=138s−1.KETismarkedas

OQOOH.

iesofautoignitionof n-heptane,n-decane,and

n-dodecaneincounterflowconfigurationandisolated fueldroplets[8,28,29] .

Figure 5 showsthemole-fractionof hydrogen peroxide(H2O2),XH₂O₂,andmole-fractionof

keto-hydroperoxide (KET), XKET, asfunctions of T2.

Forallsurrogates,H2O2andKETareformedatthe

onset of low-temperatureignition andtheir con-centrationbecomesnegligibleattheonsetof“hot” ignition. Thus,for allsurrogates ‘hot’-ignitionis preceded by low-temperature ignition. For pur-posesofillustration,computationswerealsodone forSurrogate C,withkineticsteps characterizing low-temperaturechemistryformethylcyclohexane removed.Figure 5 showsthatasaconsequenceof thisremoval,theformationof H2O2 atlow

tem-peratureissuppressedandH2O2isformedinlower

amounts only prior to the “hot”-ignition. Thus low-temperature ignition of Surrogate C comes

primarilyfrommethylcyclohexaneinSurrogateC. Computations show that for the surrogates, the exothermicreactionzonethatleadstoautoignition isontheoxidizersideofthestagnationplane.The fueldiffusesthroughthestagnationplanetoazone where low-temperature reactions take place and formH2O2,peroxidesandketo-hydroperoxidesas

wellasheptenesandheterocyclecompounds.These species,especially H2O2 diffusefurther upstream

towards theoxidizer boundaryandfacilitate the “hot”-ignition.

Figure 6 compares predicted values of criti-cal conditions of autoignition forthe surrogates withexperimentaldata.Ingeneralthepredictions agree well with measurements. At a given value of p, Tig for the Aachen Surrogate is the

low-est followed by Surrogate C and the Princeton Surrogate.Therelativeorderof reactivitycanbe explained by consideration of the volatility and

(7)

Fig.6.Thetemperatureof theoxidizerstreamatautoignition,Tigasafunctionof pressure,p,ata2=138s−1,and

YO2,2=0.15.Thesymbolsrepresentexperimentaldataforthesurrogates(sameasthoseshowninFig. 2 )andthelines

arepredictions.ThefigurealsoshowspredictionsforSurrogateCobtainedafterremovingreactionsthatcharacterize low-temperaturechemistryofmethylcyclohexane(Markedascalc-NoLTMCYC6).

low-temperaturereactivity of thecomponentsof thesurrogates.Apreviousworkhasshownthatfor high-molecularweighthydrocarbonfuels autoigni-tionisinfluencedbylow-temperaturechemistryat lowstrainrateswhileathighstrainratesitis in-fluencedbymoleculartransport[13] ,andthe influ-enceof low-temperaturechemistry increaseswith increasingpressure[8] .Thepresentworkisat rela-tivelylowstrainratesandelevatedpressure, there-foreautoignitioncanbeexpectedtobeinfluenced bylow-temperaturechemistry.TheAachen Surro-gateiseasiesttoignitebecausethevalueof Ts is

highestamongthesurrogates tested.Asa conse-quencelargeamountsofn-decane,thatisknownto havesignificantlow-temperaturereactivity[8,13] , vaporizesfromthesurfaceleadingtoignition. Sur-rogateCandthePrincetonSurrogatehavesimilar volatility butthecomponentshavedifferent low-temperaturereactivities.Inparticular,thepresence of 2,2,4-trimethylpentaneinthePrinceton Surro-gate(insteadofmethylcyclohexaneasinSurrogate C) and of alarger amountof aromatics are re-sponsible forits lower reactivity.Figure 6 shows thatthereactivityofSurrogateCdecreasesif reac-tionsthatcharacterizelow-temperaturechemistry ofmethylcyclohexaneareremoved,andforagiven

pthedifferencebetweenthepredictedvaluesofTig

with and withoutlow-temperature chemistry for methylcyclohexaneincreaseswithincreasing pres-sure.Thisfurtherhighlightstheinfluenceof low-temperature reactivity of methylcyclohexane on autoignition.

Figures 7 –9 show the sensitivity coefficients for the Aachen Surrogate, Surrogate C, andthe Princeton Surrogate just before onset of “hot”-ignition.Theyshow thatautoignitionissensitive toreactionsthatcharacterizelowand intermedi-atetemperaturechemistry[30] ofthemorevolatile componentinthesurrogate.Forexample,the con-centrationof n-dodecane inthegasphaseabove theliquidpoolofSurrogateCisabouteighttimes smallerthanthatof methylcyclohexane,although

n-dodecaneisthelargestconstituentofthis surro-gate.Asaconsequence,autoignitionof Surrogate Cismainlyinfluencedbylow-temperature chem-istry of methylcyclohexane, andnot that of the morereactiven-dodecane.Theeffectof o-xylene, whichhasavaporpressurebetween thoseof the other twocomponents,is alsohighlighted inthe sensitivityanalysisshowninFig. 8 .Theinfluence ofHO2isalsoevidentfromthesensitivitiesofits

reactionsshowninFigs. 7 and8 .Itisformedin theNTCregion, andtheopposinginfluences of itsterminationstepwhichformsH2O2,andits

re-actionswiththebenzylictyperadicalsofo-xylene and1,3,5-trimethylbenzenethatconvertitintothe morereactiveOH,playmajorrolesindetermining ignitionpropensity[31] .Moreover,the decompo-sition of H2O2 is also a key reaction

control-ling the transition from NTC to high tempera-tureignition.Forthesamereasons,Fig. 7 shows thatautoignitionofAachenSurrogateisprimarily

(8)

1612 G.Mairingeretal./ProceedingsoftheCombustionInstitute37(2019)1605–1614

Fig.7. SensitivitycoefficientfortheAachenSurrogatejustbeforeonsetof“hot”-ignitionatp=6bar,YO2,2=0.15,and

a2=138s−1.

Fig.8.SensitivitycoefficientforSurrogateCjustbeforeonsetof“hot”-ignitionatp₌6bar,YO2,2=0.15,anda2=

(9)

Fig.9. SensitivitycoefficientforPrincetonSurrogatejustbeforeonsetof“hot”-ignitionatp₌6bar,YO2,2=0.15,and

a2=138s−1.

influenced by low-temperature reactivity of

n-decane.Besidebeingthemainconstituentof this surrogate,becauseitsvaporpressureisclosetothat of 1,3,5-trimethylbenzene,itsconcentrationis rel-ativelyhighinthegasphase.Figure 9 showsthat thecriticalconditionofautoignitionofthe Prince-ton Surrogate is primarily influenced by 2,2,4-trimethylpentaneasitisthemostvolatile compo-nentof thissurrogate,despitethelargerpresence ofthemorereactiven-dodecane.

6. Concludingremarks

Experimental and computational studies de-scribed here show that the Aachen Surrogate, Surrogate C, and the 2nd generation Princeton Surrogatereproducethecriticalconditionsof au-toignitionof thejetfuelsJP-8,andJet-A reason-ablywell.Thepredictionsofcriticalconditionsof autoignitionofthekineticmodelarereasonably ac-curate.Thecomputationalstudyshowsthe domi-nantroleplayedbylow-temperaturechemistryin promoting autoignition, and a noteworthy find-ing isthat autoignition is promotedby the low-temperature chemistry of the mostvolatile com-ponentinthesurrogatemixture.Theinvestigation describedherewasrestrictedtolowvaluesof the strainratesandlow valuesof YO₂,2.Itwouldbe

ofinteresttotesttheinfluenceoflow-temperature chemistryonautoignitionathighervaluesofstrain rates,oxygenmassfraction,andpressure.

Acknowledgments

The authors acknowledge Professor F. A. WilliamsandProfessorE.Ranzifortheir contri-butions andhelpful discussions. The research at UCSDissupportedbytheU.S.Army Research Office , Grant # W911NF-16-1-0054 (Program ManagerRalph A.Anthenien Jr). Politecnicodi Milano has receivedfundingfrom theEuropean Union’s Horizon 2020 research and innovation programmeundergrantagreementNo723525 .

Supplementarymaterial

Supplementarymaterialassociatedwiththis ar-ticlecanbefound,intheonlineversion,atdoi:10. 1016/j.proci.2018.06.224 .

References

[1] T. Edwards , L.Q. Maurice , J.Propul.Power17 (2001) 461–466 .

[2] S. Dooley , S.H. Won , J. Heyne , et al. , Combust.

Flame159 (2012) 1444–1466 .

[3] T. Malewicki , S. Gudiyella , K. Brezinsky , Combust. Flame160 (2013) 17–30 .

[4] K. Narayanaswamy , P. Pepiot , H. Pitsch , Combust. Flame161 (2014) 866–884 .

[5] Y. Zhu , S. Li , D.F. Davidson , R.K. Hanson , Proc.

Combust.Inst.35 (2015) 241–248 .

[6] A. Smallbone , W. Liu , C. Law , X. You , H. Wang ,

(10)

1614 G.Mairingeretal./ProceedingsoftheCombustionInstitute37(2019)1605–1614 [7] R. Gehmlich , A. Kuo , K. Seshadri , Proc.Combust.

Inst.35 (2014) 937–943 .

[8] G. Mairinger , A. Frassoldati , R. Gehmlich , et al. ,

Combust.TheoryModel.20 (2016) 995–1009 .

[9] S. Honnet , K. Seshadri , U. Niemann , N. Peters , Proc. Combust.Inst.32 (2008) 485–492 .

[10] S. Humer , R. Seiser , K. Seshadri , J.Propul.Power27 (2011) 847–855 .

[11] K. Seshadri , S. Humer , R. Seiser , Combust.Theory Model.12 (2008) 831–855 .

[12] G. Jomaas , X. Zheng , D. Zhu , C. Law , Proc. Com-bust.Inst.30 (2005) 193–200 .

[13] R. Grana , K. Seshadri , A. Cuoci , U. Niemann , T. Faravelli , E. Ranzi , Combust.Flame 159 (2012) 130–141 .

[14] K. Seshadri , F.A. Williams , Int.J.HeatMassTransf.

21 (2) (1978) 251–253 .

[15] R. Seiser , NonpremixedCombustionofLiquid

Hy-drocarbonFuels, Technical University of Graz, 2000

Ph.D thesis .

[16] U. Niemann , K. Seshadri , F.A. Williams , Proc. Com-bust.Inst.34 (2013) 881–886 .

[17] U. Niemann , K. Seshadri , F.A. Williams , Combust. Flame161 (2014) 138–146 .

[18] F. Carbone , A. Gomez , Proc. Combust. Inst. 35 (2015) 761–769 .

[19] L. Tosatto , F. Mella , M.B. Long , M.D. Smooke ,

Combust.Flame159 (2012) 3027–3039 .

[20] K. Seshadri , A. Frassoldati , A. Cuoci , et al. ,

Com-bust.TheoryModel.15 (2011) 569–583 .

[21]S. Humer , A. Frassoldati , S. Granata , et al. , Proc. Combust.Inst.31 (2007) 393–400 .

[22]E. Ranzi , A. Frassoldati , A. Stagni , M. Pelucchi , A. Cuoci , T. Faravelli , Int.J.Chem.Kinet.46 (2014) 512–542 .

[23]A. Agosta , N.P. Cernansky , D.L. Miller , T. Far- avelli , E. Ranzi , Exper.ThermalFluidSci. 28 (2004) 701–708 .

[24]B.W. Weber, W.J. Pitz, M. Mehl, E.J. Silke, A.C. Davis, C.-J. Sung, Combust. Flame 161 (8) (2014)1972–1983,doi:10.1016/j.combustflame.2014. 01.018 .

[25]A. Cuoci , A. Frassoldati , T. Faravelli , E. Ranzi ,

Com-put.Phys.Commun.192 (2015) 237–264 .

[26]J.M. Smith , H.C.V. Ness , M.M. Abbott , Introduction

toChemicalEngineeringThermodynamics, Fifth, Mc

Graw Hill, 1996 .

[27]D.W. Green , R.H. Perry , Perry’sChemicalEngineers’ Handbook, 8, McGraw Hill, 2008 .

[28]A. Cuoci , A. Frassoldati , T. Faravelli , E. Ranzi , Proc.

Combust.Inst.35 (2015) 1621–1627 .

[29]A. Frassoldati , G. D’Errico , T. Lucchini , et al. ,

Re-ducedkineticmechanismsofdieselfuelsurrogatefor

engineCFDsimulations,Combust.Flame 162 (2015)

3391–4007 .

[30]J. Zádor , C.A. Taatjes , R.X. Fernandes , Progress

En-ergyCombust.Sci.37 (2011) 371–421 .

[31]G. da Silva , J.W. Bozzelli , Proc.Combust.Inst. 32 (2009) .