An experimental and kinetic modelling study of n-C4C6aldehydes oxidation in a jet-stirred reactor

Proceedings of the Combustion Institute 37 (2019) 389–397

www.elsevier.com/locate/proci

An

experimental

and

kinetic

modelling

study

of

n

-C

₄

–C

₆

aldehydes

oxidation

in

a

jet-stirred

reactor

Matteo

Pelucchi

,

Sylvain

Namysl

,

Eliseo

Ranzi

,

Alessio

Frassoldati

,

Olivier

Herbinet

,

Frédérique

Battin-Leclerc

,

Tiziano

Faravelli

a_{CRECKModelingLab,DepartmentofChemistry,MaterialsandChemicalEngineering“G.Natta”,Politecnicodi}

Milano,P.zzaLeonardodaVinci32,20133Milano,Italy

b_{LaboratoireRéactionsetGéniedesProcédés,CNRS,Université deLorraine,ENSIC,NancyCedex,France}

Received28November2017;accepted22July2018 Availableonline8August2018

Abstract

Inrecentyearsafewexperimentalandkineticmodellingstudieshavebeendevotedtotheunderstanding of theoxidationchemistryofaldehydes,becauseoftheirimportanceasintermediateandproductspecies inalkaneandbiofueloxidation.Inthiswork,newjet-stirredreactorexperimentaldataarepresentedfor

n-butanalandn-pentanal,extendingtheavailabilityoftargetsforkineticmodelvalidation.Consistentlywith previousdetailedmeasurementsonn-hexanaloxidation,experimentshavebeencarriedoutforbothfuelsover thetemperaturerange475–1100K,ataresidencetimeof2s,pressureof106.7kPa,inletfuelmolefraction of0.005andatthreeequivalenceratios(φ =0.5,1and2).ArecentlypublishedliteraturemodelbyPelucchi etal.wasusedtointerprettheseexperiments.TheassumptionaccordingtowhichmostoftheCnaldehyde

reactivityiscontrolledbythelow-temperaturebranchingpathwaysof theCn-1 alkylradical,allowsgood

agreementbetweenexperimentsandmodelintermsoffuelconversionandformostofthedetectedspecies. ThesystematicandcomparativeanalysisherepresentedforC4–C6 linearaldehydesfurtherconstrainsthe

generalraterules,applicabletothedescriptionofhighermolecularweightaldehydes,whichcanbeproduced fromheavieralcohols(n-pentanol,n-hexanoletc.)andfossilfueloxidation.

© 2018TheAuthor(s).PublishedbyElsevierInc.onbehalfofTheCombustionInstitute.

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

Keywords: Aldehydes;Low-temperaturekinetics;Jetstirredreactor;Kineticmodeling

1. Introduction

Pursuing a sustainable energy scenario for transportationrequirestheblendingof renewable

∗_{Correspondingauthor.}

E-mailaddress:matteo.pelucchi@polimi.it (M.Pelucchi).

oxygenated fuels (e.g. alcohols) into commercial hydrocarbonfuels. Alcoholcombustionis univo-cally associated with anincrease inthe emission of toxic compounds such as aldehydes and ke-tones [1,2]. The synergistic development of new fuels andengine technologies [3],as wellas the optimization of pyrolysis and gasification pro-cesses of biomass derived bio-oils, requires the assessmentoftheinfluenceofdifferentfunctional

https://doi.org/10.1016/j.proci.2018.07.087

1540-7489© 2018TheAuthor(s).PublishedbyElsevierInc.onbehalfofTheCombustionInstitute.Thisisanopen accessarticleundertheCCBYlicense.(http://creativecommons.org/licenses/by/4.0/ )

groupsonthereactivityof agivencompound[4]. This motivatesrecentresearchefforts devotedto a better understanding of long chain aldehydes combustion [5–10]. Veloo et al. [5,6] presented laminarflamespeedsandjet-stirredreactor(JSR) measurementsforC3–C4 aldehydes,togetherwith

akineticmodel.Pelucchietal.[7,8]developedand validatedakinetic mechanismcoveringpyrolysis, highandlow-temperatureoxidationconditionsfor different aldehydes(Rn-(C=O)-H)[8].Propanal,

n-butanal and n-pentanal pyrolysis and high

temperatureoxidation wereinvestigatedin shock tubes [7]. The kinetic model was then extended to low temperature conditions for propanal and

n-butanal. The dominant H-abstraction channel at the aldehydic site forms a carbonyl radical (Rn-CO),which rapidly decomposes to an alkyl

radical (Rn) and CO. The low-temperature

oxi-dationof thegenericCn aldehydethenfallsinto

thelow-temperaturepathwaystypicalofCn-1alkyl

radicals.Afurtherconfirmationtothisassumption wasrecentlygivenbySerinyeletal.[10],analyzing theoxidationof twobranched C5 aldehydes (i.e.

2- and3-methylbutanal)in a JSRat500-1200K andp=10 atm. Indeed,sec–butyl and iso–butyl radicals were found to rule the low-temperature reactivityof2-and3-methylbutanal,respectively.

RecentlyRodriguezetal.[9]studiedthe oxida-tionofn-hexanalinanatmosphericpressureJSR at475-1100K,measuringreactionproductsusing aGCandcontinuouswavecavityring-down spec-troscopy(CRDS).Theseadvancedanalytical tech-niquesallowedidentifyingspecificlow-temperature oxidationproductssuchascyclicetherswithaC5

skeletonandmulti-oxygenatedC6 species.The

ki-netic model used,despitecorrectlycapturingthe overallreactivity,wasnot abletoexplainthe for-mation of some minor species such as acetone, propanalandacrolein.

Thisstudyaimsatpresentingnew experimen-tal data of n-butanal andn-pentanal oxidation,

complemented by previous data on n-hexanal

oxidation[9],andatrefiningthepreviouskinetic models[7–9]inordertobetterdescribealdehydes low-temperatureoxidation.

2. Experimental

Experiments were carried out in a heated

isothermalquartzjet-stirredreactor[11,12]. Exper-imentswereperformedunder steadystate condi-tions, at 1.05 atm,at a residence time of 2s, at temperaturesrangingfrom500to1100K,andat threeequivalentratiosof 0.5,1and2withinitial fuelmolefractionof0.005.Thereactivegases, di-lutedinhelium,enteredthesphericalJSRthrough fournozzles,positionedinthecentreofthevessel, anddesignedinordertoensureaperfectgas mix-ingthroughturbulentjets.Thermalgradientsinside thereactorwerereducedbyusingaquartzannular

pre-heatingzoneprioritsentrance.Theresidence timeinthiszonewasnegligiblecomparedtothat inthereactor. Heliumandoxygenwereprovided byMesser(puritiesof99.99%and99.999%, respec-tively)with flows controlledusing flowrate con-trollers.Theliquidflow(aldehydeswereprovided bySigmaAldrich,withaminimalpurityof99.5% forbutanaland97%forpentanal)wascontrolled usingaliquid-Coriolis-flow-controller,mixedwith heliumandpassedthroughanevaporatorbefore beingmixedwithoxygen.Therelativeuncertainty ingasflowrateswasassumedtobearound5%.

Usingheated(at423K)on-lineconnectionsto avoidcondensation,thegasesleavingthereactors were quantified using three gas chromatographs (GCs):

• O2,CO, CO2 andCH4 wereanalyzedbya

GCequippedwithathermalconductivity de-tector,aflame-ionizationdetector,anda Car-bospherepackedcolumn,

• Molecules containing up to five carbon atomswereanalyzedbyaGCequippedwith a flame-ionization detector preceded by a methanizerandaPlotQcapillarycolumn, • Heavier molecules were analyzed by aGC

equipped with a flame-ionization detector andaHP-5capillarycolumn.

Identificationwasmadeusingafourthoff-line

GCfitted with a Plot Q or aHP-5MS column,

andcoupledwith amass spectrometer.Response factors were determined by injecting calibration mixtures or using the effective carbon number method. Relative uncertainties in mole fractions were estimated tobe± 5% forspecies calibrated using standards (the typical uncertainty on GC measurements).Forspeciesconcentrations calcu-lated with the effective carbon number method, uncertaintiesareestimatedtobearound10%.

3. Kineticmodelling

The detailed model of C3–C4 aldehyde

oxi-dation[7,8]was lumpedaccordingto procedures already discussed and adopted for many fuels

[13,14].Comparisonsbetweentheoriginaldetailed mechanismandthelumpedonearereportedinthe SupplementaryMaterial.Thederivedmodel,was systematicallyextendedtodescriben-pentanaland n-hexanaloxidation.Figure1showsthe effective-nessofthelumpedapproachintermsofreduction ofnumberofspecies.Theassumptionof thefast decarbonylationofcarbonylradicals[8]already al-lowsafirstreduction.Figure1schematicallyshows thelumpedlow-temperatureoxidationmechanism ofn-hexanal.Although,namesoflumpedisomers andsamples of molecular structures are therein reported,SupplementaryMaterialbetter summa-rizesnamesandmolecularstructures.Horizontal lumpingof isomerspermitstodescribealdehydes

Fig.1.Lumpedkineticmechanismofn-hexanal oxida-tion.NamesandstructuresofsomelumpedC6isomers.

mechanism with only 8 species, whereas the de-tailed mechanism of n-hexanal would need∼80 species.ThisreductionwithintheglobalCRECK model is of particular interest as this model is specificallyconceivedtodescribeoxidationofreal fuels,uptodieselfuels,largemethyl esters(C16),

biomass components[14], andalso heavy PAHs (C20)precursorsofsootparticles.

Assuming the fast decarbonylation reaction of Rn-CO, a single radical (RALDnX)

lump-ing all of the Cn aldehyde radicals carrying the

unpaired electron on the alkyl moiety has been accounted for. Again, one lumped unsaturated aldehyde (CnH2n-2O) coming from

dehydrogena-tion reactions (RALDnX=H+CnH2n-2O) or

fromintermediatetemperaturepathwayshasbeen considered. Atlow temperatures, successive oxy-genadditionandisomerizationreactionsproduce carbonyl peroxy radicals (RALDnOOX), alkyl

carbonyl hydroperoxide radicals (QAnX), peroxy

carbonyl-hydroperoxide radicals (ZAnX) and

di-carbonylhydroperoxides (KEAnX).

Interme-diate low-temperature decomposition pathways of QAnX radicals can lead to the formation of

smaller aldehydes together with smaller olefins. Moreover,QAnXcandecomposetoformlactones

(ETALDnX)andOH.

The CRECK kinetic model, obtained by

ex-tending the lumped version of C3–C4 aldehydes

oxidation[7,8]ton-pentanalandn-hexanal, con-sists of 416 species and11,500 reactions. Inthe attempttopursueandencourageacomplete uni-ficationof core mechanisms, theCRECKmodel wasrecentlyupdatedandimplementsaC0–C3core

mechanism obtained by couplingtheH2/O2 and

C1/C2 fromMetcalfeetal.[15],C3fromBurkeet

al.[16],andheavierfuelsfromRanzietal.[14,17].

The thermochemical properties were adopted,

when available, from [18]. The thermodynamic properties of additional fuelsspecific speciesare taken from [5,6,9]. The mechanism is provided in the Supplementary Material, together with transportandthermodynamicproperties.

Allsimulationshavebeenperformedusingthe OpenSMOKE++software[19].

4. Resultsanddiscussion

Figure2comparesexperimentalandpredicted molefractionprofilesofthethreealdehydes,at dif-ferentstoichiometries.Additionalcomparisonsare reportedintheSupplementaryMaterial(Fig.

S1-S10). CRECK mechanism predicts experimental

conversionsquite satisfactorily.Inparticular,the relativereactivity bothathigh andlow tempera-turesiswellcaptured:n-hexanal>n-pentanal> n-butanal.InadditiontothehigherO2concentration

atfixedfuelcontent,thelow-temperaturereactivity ishigherforheavieraldehydesalsobecauseof en-hanced RO2=QOOH isomerization possibilities.

Asanexample,atφ =0.5 experimentalfuel con-versionsatT=625Kare∼32%,∼70%and∼75% forn-butanal, n-pentanaland n-hexanal, respec-tively.Themodel slightly over-predictsn-butanal

Fig.2.Comparisonofexperimental(symbols)andpredicted(lines)molefractionprofilesof0.5%n-butanal,n-pentanal andn-hexanal[9] oxidationinanisothermalJSR,ϕ =0.5,1.0,2.0,p=1.05atm,τ =2.0s.Dottedline:verticallumping, n-pentanal=50%/50%(n-butanal/n-hexanal).Reportederrorbars:5%.

Fig.3. SensitivityanalysisofaldehydeconversiontorateconstantsatT=600K,ϕ =1.0,p=1atm.

conversion (∼40%vs∼33%) andaccurately cap-turesn-hexanallow-temperaturepeakconversion. Conversely, the model slightly under-predicts n-pentanalconversion(∼61%vs.∼68%).Although commercial n-pentanal was delivered with a de-clared97%minimalpurity,aGCanalysisrevealed a 98.8% purity. The fuel contained 0.8% of n-butanalandotherminorimpurities.Model simu-lationsprovedthenegligible effectof evenlarger n-butanalimpurities.

The equivalence ratioeffect onthe global re-activityisalsocorrectlyreproducedbythemodel, predictingdecreasinglow-temperatureextentsfor decreasingoxygenconcentrations.Aimingata fur-therreductionof thenumberof species,the dot-tedlineinφ =1.0 caseof Fig.2showspentanal fractions as obtained through thevertical lump-ingapproach[14].n-pentanalhasbeensimply as-sumedasacombinationof n-butanal (50%) and n-hexanal(50%).Theminordiscrepanciesbetween thesetwopredictionsencouragetheapplicationof vertical lumping, when modeling heavy homolo-gousspecies[20].

Results from sensitivity analyses (T=600K,

φ =1.0)ofaldehydeconversiontorateparameters are reported in Fig. 3 highlighting interesting features. Conditions where the low temperature reactivityofthethreefuelsisenoughpronounced toallowforsignificantinsightswere selected.As alreadydiscussedin[8],low-temperaturereactions ofalkylradicalscontrolaldehydelow-temperature reactivity. However, H-abstraction reactions by OH and HO2 radicals, from both the carbonyl

andthe alkylmoieties increaseoverallreactivity.

The higher impact of H-abstractions by HO2

radicalforn-butanal isexplainedonthebasisof a less pronounced low-temperature reactivity, as discussed above. When increasing carbon chain length, the importance of aldehyde specific low-temperature pathways increases,as it is thecase for n-hexanal. Namely, internal isomerization

(RALD6XOO=QA6X) and peroxy-carbonyl

hydroperoxide radical (ZA6X) decomposition to

di-carbonyl-hydroperoxides significantly impact

n-hexanal conversion. Most of the remaining

reactionsfallwithinpropaneandpentanesubsets, respectivelyforn-butanalandn-hexanal.

Tobetterassessmodelpredictions,Fig.4shows theeffectsofperturbationsofn-propylperoxy rad-icalsisomerizationrateconstant(nC3H7O2=nC3

-QOOH) on n-butanal conversion at φ =1.0, for

p=1 atm and 10 atm. The model over-predicts fuel conversion atatmospheric pressure,whereas under-predictionsareobservedathigherpressure

[6].Animprovedagreementcouldbeobtainedonly byincreasingandreducingtherateconstantathigh andlow pressures,respectively.Asexpectedfrom thelimitedpressuredependenceof the isomeriza-tionreactionsinthetemperaturerange550-750K, this observation seems to indicate the presence of some inconsistencies within the two sets of

Fig.4.Stoichiometricn-butanaloxidationinisothermal JSR. Experimental(symbols) [6] and predicted (lines) conversion.Red: 0.5%n-butanal/O2/He, p=1.05 atm, τ =2.0s. Black: 0.15% n-butanal/O2/N2, p₌10 atm, τ =0.7s.(Forinterpretationofthereferencestocolorin thisfigurelegend,thereaderisreferredtothewebversion ofthisarticle.)

Fig.5.n-butanal,n-pentanalandn-hexanal[9] oxidationinJSRat_{ϕ =}1.0,p₌1atmand_{τ =}2.0s.Comparisonbetween experimental(symbols)andpredicted(lines)molefractionprofilesofmajorspecies.

measurements.Therefore,itisreasonabletoaccept asimilarextentofover-andunder-predictionsfor bothdatasets.

Figure 5 shows comparisons between experi-mental data and model predictions for selected species from the stoichiometric oxidation of the threefuels.

Themodelcorrectlypredictsoxygen consump-tion, with largest deviations forn-pentanal case. Major species formation (CO, CO2, H2O, CH4,

C2H4,C3H6)isalsoaccuratelyreproduced.

Devia-tionsofuptoafactorof∼2areobservedforCH2O

atT>750K,andmainlyinn-butanalcase.Inthese

conditions, ∼80% of formaldehyde comes from

the decomposition of propyl hydroperoxide rad-ical:nC3-QOOH=>OH+C2H4+CH2O,whose

kineticparametershavebeenalreadyassessedfor propane oxidation[22]. The samereaction is re-sponsiblefor∼50%offormaldehydeformationat 10atm(Figs.S3–S6).Betteragreementisobtained forn-pentanalandn-hexanal.Whenasinglemodel isadoptedfordifferent fuels,asitisthecasefor theCRECKkinetic model,ad hoc modifications of rateparametersaccordingtoaspecificdataset stronglyreducesthephysicalmeaningofthemodel itself.Forthisreason,parametersbelongingtoa different kinetic subsethave not beensubject to modificationwithinthisstudy.

Fig.6. Rateofproductionanalysisofthelumpedlow-temperaturemechanismofn-butanal(black)andn-hexanal(blue)at T=600K(bold)andT=700K(italics).Speciesinbracketsarenotexplicitlyaccountedforinthekineticmodel.Examples ofspecificpathwaysarereportedintheSupplementaryMaterial.(Forinterpretationofthereferencestocolorinthisfigure legend,thereaderisreferredtothewebversionofthisarticle.)

Acetaldehydeformationiscorrectlypredictedin thethreecases,mainlyatlowtemperatureswhere its formation is controlled by ketohydroperoxide decompositions.

Athightemperatures,thesamepathways con-trolmethaneformation(CH3+HO2=O2+CH4)

and consumption (CH4+OH=CH3+H2O) for

thethreefuels.ContributionsfromH-abstractions byCH3accountonlyfor∼5-10%.Goodagreement

isfoundforn-butanalandn-hexanal,whereasthe modeloverpredictsmethaneyieldsinn-pentanal case.

Oxirane formation is also over-predicted at T>800K,inparticularforn-butanal.Onceagain the controlling reaction is not aldehyde specific. Formationof oxirane(C2H4O1-2)mostlyoccurs

viaC2H4+HO2=OH+C2H4O1-2,thereforeitis

relatedtoethyleneyieldsgenerallyaccurately pre-dicted. Consideringtheuncertainties inthe mea-surementsof H2O2 [9,21,22] the model provides

relativelygoodpredictions,mainlyatlow temper-atures.Afactorof ∼10deviationisobservedfor richcases(Fig.S10).Similardeviationswerealso discussedbyRodriguezetal.[9].

Cyclicethersandolefins,mostlycomingfrom low-temperatureoxidationpathwaysofalkyl radi-calsarewellcapturedforthethreefuels. Concern-ingcyclicethers,somedeviationsareobservedfor methyl-oxirane,properlypredictedinpropane oxi-dation[21].

Methanolisunder-predictedby∼2.5times,and iscontrolledbyH-abstractionsbymetoxyradical

athightemperatures.Atlowtemperatures, contri-butionscomefromrecombination disproportiona-tionof aldehydeperoxyradicals.Theimportance of suchreactions also foralkanes oxidation has beenpreviouslydiscussed[21,22].Relativelylarge deviations(factorof∼2)areobservedfor cyclopen-teneformationinn-hexanaloxidation.Thisspecies ismostlyformedbythedecompositionof pentyl-hydroperoxyradical(NC5-QOOH)toHO2and

n-pentene(nC5H10)thatisfurtheroxidizedthrough

typical low temperature pathways whose refine-mentisoutsidethescopeofthisstudy.

Figure6showsrateofproductionanalysis car-ried out forn-butanal andn-hexanal at φ =1.0, p=1atm,T=600and700K.Fluxanalysisfor n-pentanalisnotreportedas,fromthehierarchical natureofthisstudy,thesamereactionclasses ex-plainitsoxidation.

Selectivity of H-abstraction reactions on the carbonylmoietydecreasesfrommorethan70%for n-butanal toless than 60%forn-hexanal. These percentagesalsoreflecttherelativeweightof the Cn-1 alkyl radical low-temperature oxidation. At

T=600K,thealkylmoietyiscompletelyoxidized by addition to O2, while for increasing

temper-ature a limited contribution also comes from the isomerization to directly form the carbonyl radical (Rn-1+CO). Carbonyl peroxy radicals

(RALDnOOX) isomerize to different extents

to form QAnX. At low temperature, 23% of

RALDnOOX isomerizesto QAnXforn-hexanal,

Fig.7.DecompositionpathwaysofQAnXradicalswitharadicalpositionatthecarbonylmoiety.

Fig.8. Cn-1aldehydesformationfromCnaldehyde

oxi-dation.Symbols:Experimentaldatafromthisstudyand in[9] .Lines:simulationresults.

low-temperature branching in n-butanal case. This difference depends on molecule size and

on importance of RALDnOOX decomposition

pathways.

TheisomerizationproducesQOOHspecies car-ryingtheradicalatthecarbonylposition,these rad-icalsrapidlyproduceRn-1andCO.Dependingon

therelativepositionof the–OOHgroupwith re-specttothecarbonylmoiety,thisstepproducesthe QOOHradicaloftheCn-1alkaneoraCn-1aldehyde

andOH.Thetwopossiblechannelsinn-hexanal oxidationmechanismaredepictedinFig.7.This pathwayaccountsfor∼11-16%inbutanal,and ∼6-11%inn-hexanaloxidation.

Figure8showstherelativeimportanceof the secondchannelleadingtotheformationoftheCn-1

aldehyde,i.e.propanalandpentanalforn-butanal and n-hexanal, respectively. The higher

concen-tration of RALDnOOX in n-hexanal oxidation

sustainsthe productionof pentanal,whose yield does not exhibit the double maximum observed forpropanalfromn-butanal.Becauseofthe pres-enceofn-butanalasanimpurityinn-pentanalfeed (Section 3), experimental data are not reported, whereasn-butanalsimulationprofileinFig.8refers toa100%purityn-pentanalfeed.

Decomposition channels of QAnX are of

limited impact under the conditions

inves-tigated in Fig. 6. The analogous cyclization reaction producing cyclic ethers in alkanes low-temperatureoxidation,explains theformationof lactones (ETALDnX) in the case of aldehydes.

Figure9shows thepathways responsibleforthe formationof 6-methyl-tetrahydropyranoneand 5-ethyl-dihydrofuranoneasobservedin[9].Because of the lumped approach, the comparisons with modelpredictionsrefertothesumofthesespecies. Thepeakof lactonesconcentrationisreproduced reasonably,despitethescarcesensitivityof model predictionstostoichiometricratios.

A significant formation of ketones is ob-served for n-pentanal and n-hexanal oxidation.

Figure 10 shows acetone profiles at three dif-ferent equivalence ratios for n-hexanal. Acetone is mainly formed via the Korcek mechanism of keto-hydroperoxide decomposition, as shown in

Fig. 10. This pathway was not accounted forin Rodriguez et al. [9]. The largest deviations are observed in the intermediate temperature range (T=700-800K)and mostly for the stoichiomet-ricandtheleanmixtures.Therelativeimportance

Fig.9.n-hexanaloxidation.Caprolactoneformationpathwaysandcomparisonbetweenmodelpredictions(lines)and experimentalmeasurements(symbols)[9] .

Fig. 10.n-hexanal oxidation.Acetoneformation path-waysandcomparisonbetweenmodel(lines)and exper-imentalmeasurements(symbols)[9] .

of keto-hydroperoxidedecomposition via O–OH

bond breaking and via the Korcek mechanism

already discussed by Ranzi et al. [22] (∼10:1 at 750K)directly impacts acetone yields.Clearly,a modificationofthisratiotowardahigher impor-tance of the Korcek mechanism would strongly impactthelowtemperaturebranchingthus wors-eningthe agreementforfuel conversion(Fig.2). Itisworthnotingthatonlyn-pentanaland heav-ieraldehydesareexpected toproduce highyields of acetone as the formation of a primary alkyl radical(e.g.pent–1-yl) abletoeffectively isomer-ize to a secondary (e.g. pent–2-yl) is necessary. Moreover,apronouncedlowtemperature reactiv-ity(i.e.higherketohydroperoxideyields)favorsits formationthroughtheKorcekmechanism[22]. Fi-nally,recombination/disproportionationreactions ofperoxyradicalsfromCn-1alkylradicalscan

ex-plainbutanoneformation(Fig.S11of Supplemen-taryMaterial).

5. Conclusions

n-butanalandn-pentanaloxidationwas exper-imentallyinvestigatedinaJSR.Coupledwiththe previousmeasurementsofn-hexanaloxidation[9], thisstudyextendsandcompletestheexistinggaps inexperimentaltargetsavailableforthevalidation of aldehydes kinetic models,of relevanceforthe combustionof alternativefuelssuch asalcohols. Basedonapreviousstudy[8],andontherecent up-dateofthecoremechanismoftheCRECKkinetic framework, thedetailed model of aldehyde low-temperatureoxidation was lumpedandextended fromn-butanal,ton-pentanalandn-hexanal, pro-vidingoverallgoodagreementwithexperimental data.Aldehydesspecificreactionclasseswerefound toberesponsibleforsomeofthedetectedspecies (i.e.Cnlactones,ketones,Cn-1smalleraldehydesand

olefinsetc.).Thisworkfurtherconstrainspathways

relevanttoaldehydeoxidation,andhighlightsthe influenceof thecarbonylmoiety onintermediate speciesformation.

SupplementaryMaterial“SMM” SMMcontains:

- Kinetic mechanism (kinetics.CHEMKIN,

textformat)

- Thermodynamics properties

(thermo.CHEMKIN,textformat)

- Supplementary_Material.doc: additional

comparisons of model and experiments,

additionalexplanationsofkineticpathways,

lumped low temperature mechanism and

species nomenclature, comparison between detailedandlumpedmechanism.

Acknowledgments

The authors acknowledge the financial

sup-portofIMPROOFproject

(H2020-IND-CE-2016-17/H2020-SPIRE-S016) European Union’s

Hori-zon2020researchandinnovationprogram(grant

agreement no. 723706). Support from COST

CM1404SMARTCATsActionthroughtheShort

TermScientificMissionisalsoacknowledged.

Supplementarymaterials

Supplementarymaterialassociatedwiththis ar-ticlecanbefound,intheonlineversion,atdoi:10. 1016/j.proci.2018.07.087.

References

[1]S.M. Sarathy , P. Oßwald , N. Hansen , K. Kohse-Höinghaus , Prog. Energy Combust. Sci.44 (2014) 40–102 .

[2]E. Sadeghinezhad , S. Kazi , F. Sadeghinejad , A. Badarudin , M. Mehrali , R. Sadri , M.R. Safaei ,

Renew.Sustain.EnergyRev.30 (2014) 29–44 . [3]M.D. Boot , M. Tian , E.J. Hensen , S.M. Sarathy ,

Prog.EnergyCombust.Sci.60 (2017) 1–25 . [4]M. Pelucchi , C. Cavallotti , E. Ranzi , A. Frassoldati ,

T. Faravelli , EnergyFuels30 (2016) 8665–8679 . [5]P.S. Veloo , P. Dagaut , C. Togbe , G. Dayma ,

S.M. Sarathy , C.K. Westbrook , F.N. Egolfopoulos ,

Proc.Comb.Inst.34 (2013) 599–606 .

[6]P.S. Veloo , P. Dagaut , C. Togbé, G. Dayma , S.M. Sarathy , C.K. Westbrook , F.N. Egolfopoulos ,

Combust.Flame160 (2013) 1609–1626 .

[7]M. Pelucchi , K.P. Somers , K. Yasunaga , U. Burke , A. Frassoldati , E. Ranzi , H.J. Curran , T. Faravelli ,

Combust.Flame162 (2015) 265–286 .

[8]M. Pelucchi , E. Ranzi , A. Frassoldati , T. Faravelli ,

Proc.Comb.Inst.36 (2017) 393–401 .

[9]A. Rodriguez , O. Herbinet , F. Battin-Leclerc , Proc. Comb.Inst.36 (2017) 365–372 .

[10]Z. Serinyel , C. Togbé, G. Dayma , P. Dagaut , Energy Fuels31 (2017) 3206–3218 .

[11]O. Herbinet , F. Battin-Leclerc , Int.J.Chem.Kinet.46 (2014) 619–639 .

[12] O. Herbinet , G. Dayma , F. Battin-Leclerc , J. Simmie , E. Blurock , Cleaner Combustion, Springer, 2013, pp. 183–210 .

[13] E. Ranzi , M. Dente , A. Goldaniga , G. Bozzano , T. Faravelli , Prog.EnergyCombust.Sci. 27 (2001) 99–139 .

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

[15] W.K. Metcalfe , S.M. Burke , S.S. Ahmed , H.J. Cur- ran , Int.J.Chem.Kinet.45 (2013) 638–675 . [16] S.M. Burke , U. Burke , R. Mc Donagh , O. Math-

ieu , I. Osorio , C. Keesee , A. Morones , E.L. Petersen , W. Wang , T.A. DeVerter , Combust.Flame162 (2015) 296–314 .

[17] E. Ranzi , A. Frassoldati , R. Grana , A. Cuoci , T. Far- avelli , A. Kelley , C. Law , Prog.EnergyCombust.Sci.

38 (2012) 468–501 .

[18] B. Ruscic , Int. J. Quantum Chem. 114 (2014) 1097–1101 .

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

Comput. Phys. Commun. 192 (2015) 237– 264 .

[20] E. Ranzi , A. Frassoldati , S. Granata , T. Faravelli ,

Ind.Eng.Chem.Res.44 (2005) 5170–5183 . [21] M. Cord , B. Husson , J.C. Lizardo Huerta , O. Her-

binet , P.-A. Glaude , R. Fournet , B. Sirjean , F. Bat- tin-Leclerc , M. Ruiz-Lopez , Z. Wang , J.Phys.Chem. A116 (2012) 12214–12228 .

[22] E. Ranzi , C. Cavallotti , A. Cuoci , A. Frassoldati , M. Pelucchi , T. Faravelli , Combust.Flame162 (2015) 1679–1691 .