Rh promoted ceria based Co on catalysts: An XPS study Oxidation active catalytic states of centers in ethanol steam reformingreaction Journal of Molecular Catalysis A: Chemical

ContentslistsavailableatScienceDirect

Journal of Molecular Catalysis A: Chemical

j o ur na l h o me p a g e :w w w . e l s e v i e r . c o m / l o c a t e / m o l c a t a

Oxidation states of active catalytic centers in ethanol steam reforming reaction on ceria based Rh promoted Co catalysts: An XPS study

Erika Varga, Zsuzsa Ferencz, Albert Oszkó, András Erd ˝ohelyi, János Kiss

DepartmentofPhysicalChemistryandMaterialsScienceoftheUniversityofSzeged,Aradivértanúktere1,H-6720Szeged,Hungary

a r t i c l e i n f o

Articlehistory:

Received1August2014 Receivedinrevisedform 26September2014 Accepted11November2014 Availableonline15November2014

Keywords:

Steamreformingofethanol Hydrogenproduction Cobalt–ceriacatalyst Rhodiumpromoter XPS

a b s t r a c t

X-rayphotoelectronspectroscopic(XPS)investigationswerecarriedouttostudytheoxidationstates ofCeO2andCeO2supportedCo–Rhcatalystsduringthetemperatureprogrammedstreamreforming ofethanolreaction(SRE).Gaschromatographyalsowasusedtoanalyzetheproductcomposition.An initialre-oxidationofthepre-reducedcatalystswasobservedbywaterreactantandourresultsrevealed atendencyoftheoxidizedmonometalliccatalyststopromotealdolcondensation-typereactions.Itwas foundthatRhenhancesthereductionofCoduringthepretreatment,andthehighestH2selectivitywas obtainedwiththebimetalliccatalystinSREreaction.Moreover,acetoneformationwasnegligibleonthis sample.EnhancedC Cbondscissionandhydrogenproductionweredetectedfrom650K.Incontrast topureethanoldecomposition,duringtheEtOH+H2Oreactionminorbutimportantchangescouldbe detectedontheCe3dspectra.Itwasconcludedthattheaccumulationofstronglybondedcarbidespecies inthecaseofCo/CeO2catalystcancontributetothedecreasingactivity.Thistypeofcarbonwasabsent inthepresenceofatraceamountsofRh,thereforethecatalystwasmorestable.

©2014PublishedbyElsevierB.V.

1. Introduction

Sustainabledevelopmentrequiresnewalternativecheapand renewable sources of energy. Great efforts are currently made toproducehydrogen,e.g.,forfuelcellapplicationsbyheteroge- neouslycatalyzedprocesses.Thisdemandinspiredstudiesofthe dehydrogenationofoxygenatedhydrocarbons[1–3].Inparticular, thelightalcohol ethanolis animportantcandidateasa chemi- calhydrogencarrier.Noblemetals,especiallyRh,areprovedtobe excellentcatalystsforthedehydrogenationreaction[4],buttheir pricesareprohibitivelyhigh.Asanalternative,thelessexpensive transitionmetalCoisconsideredtobeapromisingcatalystforthe steamreformingofethanol(SRE)[5–10].Inaddition,amixtureof carbondioxideandmethanecanserveasafeedforthecatalytic productionofhydrogenbydryreformingofmethane(DRM)[11,12]

wherecobaltcontainingcatalystsmayhaveanimportantrole.Dur- ingSRE,acidicsupportslikeAl₂O₃favordehydrationandthereby increasethetendencyforcokeformationduetothepolymerization ofethylene[13–15].However,onceria(CeO2),whichisconsidered tobeabasicsupport,dehydrationislimitedanditsredoxprop- ertieshindercokeformation[5,16].Theoxygenexchangecapacity

∗Correspondingauthor.Tel.:+3662544803;fax:+3662544106.

E-mailaddress:jkiss@chem.u-szeged.hu(J.Kiss).

ofceriumoxideisassociatedwithitsabilitytoreversiblychange theceriumoxidation statesbetweenCe⁴⁺and Ce³⁺[17–19].All theseobservationsledtotheoutstandingattentiontothecatalytic propertiesofCo/ceriasysteminSRE.

Naturally,thesurfacepropertiesofthemetalandoftheoxide support, and also themetal/oxide interface determinethe for- mationandstabilityoftheintermediatespresentintheethanol transformationprocesses.Itisgenerallyacceptedthattheprimary stepinalcoholactivationistheformationofalkoxide[20].Depend- ingontheparticularmetal,dehydrogenationandC Cbondscission leadtotheformationofalkoxide,oxametallacycle,aldehyde,acyl andcokeonthesurfaceandmostlyH2,CH4,CO,CO2andaldehyde inthegasphase[21–30].RecentstudiessuggestedthatCo²⁺sites aretheactivecentersinSRE,andCo⁰sitesareresponsibleforcoke formation[29,30],whileotherauthorsconsideredmetalliccobalt toplaythekeyroleinSRE[32].HighpressureX-rayphotoelec- tronspectroscopicstudies(HPXPS)demonstratedthatataconstant ethanol(withoutwater)pressureof0.1mbarthereductionofCe⁴⁺

toCe³⁺increasedsignificantly between320and 600K duetoa highermobility ofoxygenor Ce³⁺centersatelevatedtempera- tures.Nocokeformationwasobservedupto600KonCeO₂.During thereactionofethanolwiththeCo/CeO2(111)modelcatalystthe amountofCo²⁺decreaseddrasticallywithincreasingtemperature, and at600Kthemajority ofCowasmetallic;this process was accompaniedbyaseverereductionoftheceria[33].

http://dx.doi.org/10.1016/j.molcata.2014.11.010 1381-1169/©2014PublishedbyElsevierB.V.

Veryrecentlywehavefoundthattraceamountsofrhodium promoter(0.1%)dramaticallyalteredthereactionpathwaysofSRE onCo/ceria catalysts.In contrast toCo/ceria, on rhodiumcon- tainingCo/ceriacatalystsacetonewasnotobserved.Additionof asmallamountofRhaspromotertotheCo/CeO2catalyst,how- ever,resultedinasignificantincreaseinthehydrogenselectivity [34].Theseveryimportantfindingsmotivatedustoestablishthe oxidationstateofthecatalyticallyactivesitesbeforeandafterthe catalyticreactionofethanol+water(SRE)onpureceria,Co/ceria andRh+Co/ceriacatalystsby X-rayphotoelectronspectroscopy (XPS).

2. Experimentalprocedure

Thecatalystspreparationandcharacterizationmethodswere detailedelsewhere[34].TheceriasupportedCocatalystswerepre- paredbyimpregnatingthesupportCeO2(AlfaAesar,43m²/g)with theaqueoussolutionofCo(NO₃)₂toyieldanominalmetalcon- tentof2wt%(0.056mol%).Theimpregnatedpowdersweredriedat 383K,calcinedat973Kandpressedtopellets.TheRh–Cobimetallic sampleswerepreparedbysequentialimpregnation(impregnation withCofirst,thenthesameprocedureafterimpregnationwith 0.1wt%Rh−0.0017mol%).Aftercalcinedat973KtheBETsurface areasoftheCeO2support,the2%Co/CeO2andofthe0.1%Rh+2%

Co/CeO₂ catalystswere21.5m²/g,7.4m²/gand7.6m²/g,respec- tively,whiletheaverageporesizeswerebetween14.4and12.3nm, whichisconsistentwithamesoporousmaterial[9,34].Beforethe measurements,fragmentsofcatalystpelletswereoxidizedat673K inflowing O₂ for20minandreducedat773KinflowingH₂ for 60mininthecatalyticreactor.

Catalyticreactionswerecarriedoutinafixed-bedcontinuous- flow reactor (200mm long with 8mm i.d.), which was heated externally.Thedeadvolumeofthereactorwasfilledwithquartz beads.Theoperatingtemperaturewascontrolledbyathermocou- pleplacedinsidetheovenclosetothereactorwall,toassureprecise temperaturemeasurement.Forcatalyticstudiessmallfragments (about1mm)ofslightlycompressedpelletswereused.Typically, thereactorfillingcontained50mgofcatalyst.Inthereactinggas mixturetheethanol:watermolarratiowas1:3.Theethanol–water mixturewasintroducedintoanevaporatorwiththehelpofanHPLC pump(Younglin;flowrate:0.007mlliquid/min);theevaporator wasflushedwithArflow(60ml/min).Argonwasusedasacar- riergas(60ml/min).Thereactinggasmixture-containingArflow enteredthereactorthroughanexternallyheatedtubeinorderto avoidcondensation.Thespacevelocitywas72,000h⁻¹.Thesam- pleswereheatedinthegasmixturefrom373to773Katarateof 3K/min.

Theanalysisoftheproductsandreactantswasperformedwith anAgilent6890NgaschromatographusingHP-PLOTQcolumn.The gasesweredetectedsimultaneouslybythermalconductivity(TC) andflameionization(FI)detectors.ToincreasethesensitivityofCO andCO₂detectionamethanizerwasappliedbeforethedetectors.

Theamountandthereactivityofsurfacecarbonformedinthe catalyticreactionsweredeterminedbytemperature-programmed hydrogenation.Afterperformingthereactionsofethanol–water mixtureat823Kfor120minthereactorwasflushedwithArat thereactiontemperature;thenthesamplewascooledto373K, theAr flowwaschangedtoH₂,andthesamplewasheatedup to1173Kwitha10K/minheatingrate.Theformedhydrocarbons weredeterminedbygaschromatography.

ForXPSstudies,thepowdersampleswerepressedintopellets withca. 1cmdiameter anda fewtenthofmmthickness.Sam- pletreatmentswerecarriedoutinahigh-pressurecell(catalytic chamber)connectedtotheanalysischamberviaagatevalve.The sampleswerepre-treatedinthesamewayasdescribedabove.After

930 920 910 900 890 880 870

B

reaction at 773 K reaction

at 773 K

u''' u''

u' u

u₀ v₀ v v' v'' v'''

u₀ v u

u''' v''' v''

reduced

A

930 920 910 900 890 880 870 reduced

v0

u'' u' v'

Binding ene rgy (eV)

Fig.1.Ce3dspectrabeforeandafterSREreactiononreducedceria(A)andon0.1%

Rh+2%Co/ceriacatalysts.

thepre-treatment,theywerecooleddowntoroomtemperaturein flowingnitrogen.Then,thehigh-pressurecellwasevacuated;the samplewastransferredtotheanalysischamberinhighvacuum (i.e.,withoutcontacttoair),wheretheXPspectrawererecorded.

Asthenextstep,thesamplewasmovedbackintothecatalytic chamber,whereitwastreatedwiththereactinggasmixtureatthe reactiontemperatureunderthesameexperimentalconditionsas usedforthecatalyticreaction.XPspectraweretakenwithaSPECS instrumentequippedwithaPHOIBOS150MCD9hemispherical electronenergyanalyzer,usingAlK_␣radiation(h=1486.6eV).The X-raygunwasoperatedat210W(14kV,15mA).Theanalyzerwas operatedintheFATmode,withthepassenergysetto20eV.The takeoffangleofelectronswas20^◦withrespecttosurfacenormal.

Typicallyfivescansweresummedtogetasinglespectrum.For dataacquisitionandevaluationbothmanufacturer’s(SpecsLab2) andcommercial(CasaXPS,Origin)softwarewereused.Thebind- ingenergyscalewascorrectedbyfixingtheCe3du peak(see below)to916.6eV.

3. Resultsanddiscussion

Basedonformerstudies[34,35],itcanbeconcludedthatthe oxidized and reduced ceria is not fully inactive either in the decomposition of ethanol or in the SRE reaction. On the CeO2

support(withoutcobaltandrhodium),initiallyonlyacetaldehyde wasformed(at3–5%ethanolconversion),butbetween650and 800K (where the ethanol conversion was ∼25–30%) the main product was ethylene besidesthe less amount of acetone and CO₂.

TheCe3dspectraofreducedceriabeforeandafterSREreac- tionat773KareshowninFig.1A.Generally,theCe3dregionof CeO₂israthercomplex,i.e.,itiscomposedofthreedoublets,(u, v),(u,v)and(u,v)correspondingtotheemissionsfromthe spin-orbitsplit3d_3/2and3d_5/2corelevelsofCe⁴⁺.Thethreedou- bletsareassigned todifferentfinalstates:u (916.6eV)andv (898.4eV)areduetoaCe⁰3d⁹4f⁰O2p⁶finalstate,u(907.7eV) andv(889.0eV)toaCe3d⁹4f¹O2p⁵finalstate,andu(900.9eV) andv(882.5eV)toaCe3d⁹4f²O2p⁴finalstate[36,37].Aminor reduction ofCe⁴⁺ toCe³⁺ isbestdetectableas thesmallinten- sityincreaseoftheu(903.9eV)andv(885.3eV)peaksandalso theweakeru₀(899.3eV)andv₀(880.2eV)components,whichare characteristicofCe³⁺.Interestingly,thisspectralfeatureofceriadid notchangeafterthereactionwithethanol–watermixtureat773K (Fig.1A).ToquantifytheamountofCe³⁺,theratiooftheintegrated peakareasofCe³⁺spectralcontributionstothetotalCe3dspec- trum,i.e.,Ce³⁺/(Ce³⁺+Ce⁴⁺)wasused.TheCe³⁺contentwas11%

duringthewholeprocess.Thisisincontrastwiththeethanol–ceria interactionwithoutwater,wherethereductionofCe⁴⁺toCe³⁺was morepronouncedbetween320and600Kduetoahighermobility ofoxygenorCe³⁺centersatelevatedtemperatures[33].Itseems that thedissociation of water in SREcould re-oxidize theCe³⁺

centers.

On2%Co/CeO₂theconversionoftheethanolandtheproduct distributioninSREreactionaredisplayedinFig.2A.Atlowconver- sionupto500K,acetaldehydeandacetoneweredetectedinthegas phase.From500to700K,theacetaldehydeselectivityattenuated, whiletheselectivitiesofH2,ethyleneandCO2increasedmoder- atelyandthemaincarbon-containingproductstillwasacetone.

Above700KthedominantproductswereH₂,acetaldehyde,eth- ylene,CO₂,CO,andmethane.ThemaineffectofCoascompared tothepureCeO2casewasmanifestedinthemediumtemperature range(650–750K)astheincreasedconversion,accompaniedby enhancedselectivitiesforacetoneandH₂attheexpenseofethyl- ene(Fig.2A).Themainreactionrouteofacetaldehydeformationis thedehydrogenationofethoxyspecies[34]:

C2H5O_(ads)→ CH3CHO_(ads)+H_(ads) (1)

Acetaldehydedesorbseitherasaproductorimmediatelyoxi- dizestosurfaceacetatespeciesbylatticeoxygenorbyOHgroups.

The other reaction path is the formation of acetone (CH₃COCH_3(g)), which is the dominant product at medium temperaturesinourcase.Accordingtotheliteraturedata,acetone canbeproducedthroughaldolcondensationofacetate(Reaction (2)) or via the reaction of acetyl groups (CH₃CO) withmethyl species(Reactions(3)–(5))[5,34,38]:

2CH₃COO_(ads)→ CH₃COCH_3(g)+CO₂+O_(ads) (2)

CH3CHO_(ads)→CH3CO_(ads)+H_(ads) (3)

CH₃CO_(ads)→ CH_3(ads)+CO (4)

CH3CO_(ads)+CH_3(ads)→CH3COCH_3(g) (5)

At∼800K, theconversionand the H₂ selectivitytransiently dropped,whichwasalsoseenasanincreaseintheacetaldehyde selectivity.Apossiblereasonistheacetaldehydedesorptionand recombination/reductionwithhydrogenformingethanol.

CH3CHO_(a)+2H_(a)→ CH3CH2OH_(g) (6) Inordertogetclosertotheclarificationofsurfacemechanism, weidentifiedtheoxidationstateofcobaltandceriabeforeandafter reactionatdifferenttemperatures.InFig.4AwedisplaysomeCo 2p_3/2photoemissionsfromthe2%Co/ceriasample.Afteroxidation at673KthesignalfromCo²⁺appearedat780.4eVwiththechar- acteristicshake-upsatelliteat786.2eV.Afterreductionat773K, thepeakpositionspracticallydidnotchange,butasmallerreduced statewasdeveloped(777.8eV),itmeansthatCoishardlyreducible atthistemperature.Afterthereductionprocedure,someintensity decreasewasobserved.Thischangemaybeattributedtothesin- teringandsomeencapsulationofcobaltclustersbysupport.After reactionsat473and 623KtheCowasmainlyinoxidizedstate.

Thedeconvolutedpeaksobtainedafterreductionandafterreac- tionat473KaredisplayedinFig.6A.Thesatellitepeakat786.2eV remainedalsodetectablewhichalsosupportsthepresenceofoxi- dizedCo.ItisworthmentioningthatCo²⁺gainedintensityafterthe reactionat473and623Kwhichcouldbeexplainedbysomedis- ruptionofCoclustersduetostronginteractionwiththereactants.

Followingthe773Kpost-reactionofCo/ceriatheCo2p_3/2 spec- trumshowedaweakshoulderat777.8eVduetoaslightreduction ofCo²⁺(Fig.4Aand6A).Thechangeinitsintensityafterthereac- tionat773Kmayreflectaslightencapsulationbyceriasupport orcoveringbycarbonspeciesformedinthecatalyticprocess(see Fig.7).Sinteringordiffusionofcobaltintothebulkalsocannotbe

excluded.AfteracarefulanalysisoftheCe3dspectraofCo/ceria catalyst(notshown)wemayconcludethattheSREreaction(in contrasttothedecompositionreactionofethanolwithoutwater) causedonlyaminoradditionalreductioninceria.Itshouldbemen- tionedthatusingpureethanolintheinteractionwithCo/ceria,the ceriabecamemoreandmorereducedwithincreasingreactiontem- perature[33].ItisveryprobablethatwaterduringtheSREreaction mayservesufficientOHgroupstore-oxidizethereducedcenterof ceria.

Theacetaldehydeformationbelow600Kandhighacetonepro- ductionbetween500and750Ksuggestsapropensityoftheoxide phasesforaldolcondensation-typereactionssinceourcatalystcon- tainsasignificantnumberofCo²⁺sitesandceriaisstilloxidized inthistemperaturerangeinthesteamreforming(SRE)reaction.

Inharmonywiththeliteraturedata[5]andourrecentfindings theacetoneformationmaybeattributedtotheunreduced,nearly stoichiometricceriasupport[34].

BeforeturningtotheRhpromotedCo/ceriasystem,wesumma- rizetheproductdistributionandtheconversionofethanolobtained onceriasupportedsmallamountofRhwithoutCo(Fig.2B).On 0.1% Rh/CeO2 initially acetaldehyde and small amounts of CO andmethanewereformed,butbetween650and800Kthemain productswerehydrogen,acetone,andCO₂.AstheRhsurfacecon- centrationwasverylowtheethanolconversionwasalsolow,at 770Kitwasnotmorethan58–60%.TheRhXPSsignalsobtained beforeandafterSREat473,623Kand773KaredisplayedinFig.4B.

AfteroxidationtheRh3d_3/2 appearedat309.2eV,while theRh 3d_1/2wasdetectedat314.0eV.Afterreductionat773K,theRh3d_3/2 movedto307.4eV.Thisvalueissomewhathigher(by0.3eV)as wasobservedonbulkphasemetallicrhodium.Thisbindingenergy differencecanbeattributedtothesmallparticlesizeofRhclus- tersonceria.Thepeak positiondidnotalter afterthereaction (Fig.4B).

The presence of a small amount of Rh in Co/ceria catalyst increasedtheethanolconversionandbasicallyalteredtheproduct distributionofSREreaction,significantlyincreasedthehydrogen selectivity(Fig.2C).Themostcatalyticandspectroscopicmeasure- mentswerecarriedouton0.1%Rh+2%Co/CeO2catalyst.TheCo/Rh atomicration is33.0in thiscase.On theRh-promotedCo/CeO₂ catalyst,below 550Ktheproductswereacetaldehyde,methane andCO.Between600and750Khydrogen,methane,CO,CO2,and acetaldehydeweredetected,andaround800K(wheretheethanol conversionreaches90–95%),hydrogenandCO₂ weredominant, butCH4andCOaswellassmallamountsofethyleneandacetalde- hyde were also formed (Fig. 2C). It is worth emphasizing that acetonewasnotdetectedatanytemperatureontheRh-promoted Co/CeO2catalyst,inspiteofthefactthataround700Kitwasthe majorhydrocarbonproductadsorbedon2%Co/CeO₂andon0.1%

Rh/CeO₂,and itwaswelldetectableeven onCeO₂ atthesame temperature.

Toobtainadditionalinformation, time-dependentisothermal measurementswerecarriedoutat723K.Fig.3AandBdisplaythe ethanolconversionandhydrogenselectivityasafunctionofreac- tiontimeondifferentceriabasedcatalysts.Itisclearlyseenthat themostefficientwastheRhpromotedCo/ceriacatalyst.Inorder topointouttheefficiencyofthe0.1%Rhpromoter,weperformed someexperimentswithaRh-free10%Co/CeO2catalyst.Fig.3Aand Bclearlydemonstratesthattheeffectof0.1%Rhon2%Co/CeO₂is moresignificantthantheincreaseinColoadingintermsofboth conversionandselectivity.

OneofthemostimportantXPSobservationsisthatrhodium promotedthereductionofcobaltat773K.Fig.5Aand6Bshow thatinthereductionprocesstheCo3d_3/2issplitup,anewcompo- nentbelongingtometallicCodevelopedasashoulderat777.8eV.

Theeffectcanbeexplainedbythehydrogenspilloverphenomenon [34,39].The promotingeffectof noble metals onthereduction

400 500 600 700 800 900 1000 0

20 40 60 80 100

%

400 500 600 700 800 900 1000 0

20 40 60 80 100

%

400 500 600 700 800 900 1000 0

20 40 60 80 100

%

hydrogen carbon monoxide methane carbon dioxide ethylene ethane acetaldehyde acetone diethyl ether ethyl acetate Conversion of ethanol

Temperature (K) C A

B

Fig.2. Selectivitiesasafunctionoftemperatureintheethanol–watersteamreformingreaction(1:3ratio)performedwithlinearheating(3K/min)from373to800K.(A) 2%Co/ceria,(B)0.1%Rh/ceriaand(C)0.1%Rh+2%Co/ceriacatalyst.

ofcobaltwasobservedearlieronaluminasupportedCocatalyst [12,39,40].Itwasassumedthathydrogenreducedthenoblemetal first,wasactivatedonit,andspilloverledtotheincreaseofthe reducibilityofCo[39].Anotherexplanationcanbethatinthepres- enceofnoble metalchanges ofthecrystal sizeresulted higher reducibility.Zhangetal.[41]foundbymeansofXPSandXRDthat aminoramountofRhcanpreservethedispersionofCoonalu- minaandthiswayalsohindersthedeactivationofthecatalystin methanedryreformingreaction.

Thesecondimportantresultoftheelectronspectroscopicmea- surementsisthattraceamountsofrhodiumsignificantlyaltered

Fig.3.Conversionofethanol(A)andselectivityforhydrogen(B)asafunctionof reactiontimeat723Kon2%Co/CeO2(䊉),10%Co/CeO2(),0.1%Rh+2%Co/CeO2

(),0.1%Rh/CeO2()andCeO2()catalysts.

theoxidationstatesofcobaltandceriaduringtheethanol–water (SRE)reaction(Figs.1B,5Aand6B).Interestingly,afterSREreaction at473–623K,whereacetaldehyde,methaneandcarbonmonox- ide were the dominant products, the cobalt was re-oxidized (Figs.5Aand6B).ThemostintensivepeakwasdetectedforCo²⁺, andthesatellitestrengthenedat786.2eV.The existenceofthis satelliteindicatesthatthecobaltoxidationstateismainly+2at thistemperaturerangeandonlyamuchsmallercomponentcould

795 790 785 780 775 770 316 312 308 304

777.8

200 cps 786.2 780.4

773 K

623 K 473 K reduced

oxidized

Binding energy(eV)

1000 cps

314.0

309.2 312.2

Rh3d

B _307.4

A ^Co2p3/2

Fig.4.Co2pspectraafterethanol–watersteamreformingreactionatdifferenttem- peratureson2%Co/ceriacatalyst(A)andRh3dspectraon0.1%Rh/ceriacatalyst.

Thespectraafteroxidationandreductionpretreatmentarealsoshown.

795 790 785 780 775 770 316 312 308 304

Binding energy (eV) 786.7 777.8

773 K

623 K 473 K reduced

780.4

oxidized

1000 cps Co 2p

3/2 Rh 3d

314.0

309.2 312.2

307.4

200 cps

B A

Fig.5.Co2pspectraafterethanol–watersteamreformingreactionatdifferenttem- peratureson0.1%Rh+2%Co/ceriacatalyst(A)andRh3dspectraon0.1%Rh+2%

Co/ceriacatalyst.Thespectraafteroxidationandreductionpretreatmentarealso shown.

bedetectedformetalliccobalt(777.8eV).ThesignatureofCo³⁺

isaCo2p_3/2peakat780–781eVwithnosatellite[29,31].These observationevidencedthatthisre-oxidationoccursthroughisthe formationofwater-inducedCooxidesunderreactioncondition.

VeryrecentlyLinandco-workers[42]foundsuchkindofprocess onaverysimilarsystem(supportedcobalt/ceria–zirconiacatalysts) underethanolsteamreformingconditions.Whenthereactiontem- peraturewasincreasedto773K,wherethehydrogenproduction isdominated,asignificantfractionofmetallicCoappearedagain (Fig.5AandFig.6B).Similarre-oxidationandreductionstepswere detectedbyXRDonCo/ceriaduringESR[9].AsitwasexpectedRh remainedinreducedstateonbimetalliccatalystatanyreaction temperature(Fig.5B).It isworthmentioningthatthereduction degreeofceriaisincreasedupslightlyupto773Kduringreaction.

TheCe³⁺concentrationafter773Kreactionwas17%(Fig.1B).

795 790 785 780 775 770 795 790 785 780 775 770

Binding energ y (e V)

500 cps 500 cps

1000 cps 500 cps

500 cps

786.7 780.4 777.8 786.2 780.4

777.8

Co 2p_3/2

500 cps

A B

reduced reaction at 473 K reaction at 773 K

Fig.6.PeakfittingforselectedCo2p3/2ofFigs.4and5.Co2p3/2afterreductionand afterreactionatdifferenttemperatures;2%Co/ceria(A)and0.1%Rh+2%Co/ceria (B).

FromourobservationsweconcludethatthepromoterRhhasat leasttwodifferentrolesinthiscatalyticsystem.First,thereduction ofCo(andCeO₂)inH₂wasmuchmoreefficientinthepresence ofRhduetothehydrogenspilloverphenomena[34].SinceH₂is alsopresentasaproduct,Rhmayalsohelptokeepthecobaltin themetallicstate.Ontheotherhand,RhpromotestheC Cbond scissionreactionofethanol,producingadsorbedCH₃[3,43].This isinagreementwiththefactthatonourRh-containingsamples theselectivityofmethaneishigherthanontheCeO₂andCo/CeO₂ systems.Takingintoaccounttheseobservations,weproposethat Reactions(3)and(4)arethemainreactionstepson0.1%Rh+2%

Co/CeO2catalyst,followedbyReactions(5)–(7):

CH_3(ads)→ C_(ads)+H_(ads)+H_2(g) (7)

CH_3(ads)+H_(ads)→ CH_4(g) (8)

2H_(ads)→H_2(g) (9)

Thefactthatthebimetalliccatalyst wasthemostactiveand selectiveinhydrogenproductionandatthesametimeitcontained thelargestfractionofCoinmetallicstateindicatesthatmetallic cobaltsitesareactiveintheSREreaction.Thebimetalliccatalyst withethanol–watermixturerepresentsaninterestingredoxsys- tem.ThereareaRhassistedcobaltreductionandareoxidationstep withwater.Mostlikely,bothCo²⁺andmetallicCoplayrolesindif- ferentstepsoftheSREreaction.Supposedly,Co²⁺isactiveinthe dehydrogenationofethanolatlowtemperatures(aldehydeforma- tion),whilemetallicsitesareparticularlyactiveabove700KinC C bondruptureanddecarbonylation.ThepromotingeffectofRhwas mainlyrationalizedbyanincreasedefficiencyinC Cbondscission andhydrogenformationonbothRhandmetallicCosites.

The amountand type of carbon formed in catalytic ethanol steamreformingisanimportantissue.Surfacecarbonisaknow reactionproductinthedecompositionofethanol[21].Inagreement withearlierfindings[44],carbondepositswereformedcovering bothsupportandcobaltparticles,regardlessofthetypeofsup- portused.Theextentofcokeformationandprobablyitssurface structuredependedonthesupport.Inthecaseofpureceria,in harmonywithearlierfindings[33,34]theformationofcarbona- ceousspecies is almostnegligible. After120min of reactionat 823K,theamountofdepositedcarbononCo/CeO₂was344␮mol/g.

Interestingly, although the Rh-promoted Co/CeO2 showed the highestandmoststablehydrogenselectivity,theamountofsurface carbon, 1135␮mol/g, was higher than that on Co/CeO₂. Previ- ousstudiessuggestedthatcarbonbuild-updoesnotnecessarily leadtodeactivation[5,45,46].Inorder togetmoreinformation aboutthetypeofcarbonaceousspecies, wemonitoredtheC 1s region before and after reaction at 473 and 773K on Co/ceria andRhpromotedCo/ceriacatalysts.OnCo/ceriawherethecon- version wasratherlow, (473K),themainC 1scomponent was detectedat284.5eV(Fig.7A).Theintensityofthisphotoemission increased with increasing reaction temperature (773K). Earlier electronmicroscopicresultsobtainedonceria-typesupportedCo catalystrevealedtwotypesofcarbonspecies:eithercarbonaceous layeronthesurfaceofthegrains,orfilaments[46].Presumably, thesetypesofcarboncanbedetectedat284.5eVbindingenergy.

VerylikelythesetypescannotberesolvedinourXPSapparatus.

Ashoulderatthehigherbindingenergysideat286.3eVcanbe attributedtochemisorbedethanol/ethoxide.Inharmonywiththe DRIFTSresultsthesespeciesarepresentupto773K[34].Thesignals oftheothercarboncontainingintermediates(aldehyde,acetate) overlapwiththebroadCe4sphotoemissiondetectedat288.8eV.

Interestingly,athighreactiontemperature,atwhichtheconversion ishigh(∼90–95%),anewC1speakdevelopedat282.3eV(Fig.7A).

Weattributethisphotoemissiontocarbide-likespecies.Carbidic carbonatlowbindingenergywasdetectedafterpotassiuminduced COdissociationonCofoil[47],andcarbidiccarbonformationwas

296 294 292 290 288 286284282280 296294292290288 286 284 282 280 282.3

286.3 288.8 Ce 4s

reaction at 473 K reaction at 773 K

284.5

1000 cps

B

1000 cps

288.8 Ce 4s

284.5

286.1

500 cps

reduced

500 cps

500 cps

A

C 1s

400 cps

Binding energ y (e V)

Fig.7.C1sspectrawithCe4scontributionafterethanol–watersteamreforming reactionatdifferenttemperatureson2%Co/ceria(A)andon0.1%Rh+2%Co/ceria catalyst(B).C1sregionafterhydrogenpretreatmentisalsoshowninbothcases.

observedwithC1speakmaximumat282.3eVonSi(100)during thedecompositionofCo₂(CO)₈[48].

Asimilar C 1sseriesis displayed inFig.7Bfor0.1% Rh+2%

Co/CeO2catalyst.Thecarbonaceouslayerandfilamentsstructure appearedat284.5eV.ThespeciesduetoC Obond(286.3eV)was detectedwithlessintensity,indicatingthatethoxidestability is limitedonthissurface[34].Itisveryremarkablethatcarbide-like speciesdoesnotformonthiscatalystduringethanolsteamreform- ing.Itwaspointedoutearlierthatthedecreasingactivityisrelated totheformationofcarbonfilaments[46].Inthelightofourresults wemayconcludethattheaccumulationofstronglybondedcar- bidespeciesinthecaseofCo/CeO₂catalystcancontributetothe decreasingactivity.Thistypeofcarbonisabsentinthepresenceof traceamountsofRh,thereforethecatalystwasmorestable.

4. Conclusions

Combined X-ray photoelectron spectroscopic and gas chro- matographicexperimentswerecarriedouttofindarelationship between the efficient H2 production from EtOH+H2O mixture (SRE)andtheoxidationstateofceriasupportedCo–Rhcatalysts.

Accordingtotheresults,CeO₂alonealsohadaslightactivityin the reaction, but on mono- and bimetallic Co–Rh samples the conversionandH2selectivityweremuchhigher.Becauseofthere- oxidationofCe³⁺centerbywater,wecouldnotfollowthereaction mechanismbymeansofCe3dspectraonCo/ceriacatalyst.Cobalt sinteredduringthepre-reaction,butunderreactionitdisrupted, andre-oxidized.Inthisstageacetoneformationwasdominant,soit canbeconcludedthatoxidizedcentersareneededforaldolconden- sation.For0.1%Rh+2%Co/ceriarhodiumremainedinmetallicstate afterreduction,anditenhancedthereductionofCoandslightly inducedthereductionofceria,too.OnheatinginEtOH+H₂Omix- ture,CowasbeingpartiallyreducedandincreasingH2production couldbedetected.Atthesametime,inthepresenceofrhodiumin Co/ceriacatalyst,theacetoneformationwashindered.Thepromot- ingeffectofRhwasmainlyrationalizedbyanincreasedefficiency inC CbondruptureonbothRhandmetallicCosites.Finallywe canconcludethattheRhpromotedCo/CeO₂catalystisthemost appropriateforhydrogenproductionabove700Kinethanolstream

reformingduetoitshighmetalCocontent,butCo²⁺isalsonec- essaryforthelowtemperaturereaction,probablybecauseofits dehydrogenationactivityinwhichthealdehydeisthemainreac- tionproduct.Inthelightofourresultswemaysuggestthatthe accumulationof stronglybonded carbide speciesin thecase of Co/CeO₂catalystcancontributetothedecreasingactivity.Thistype ofcarbonisabsentinthepresenceoftraceamountsofRh,therefore thecatalystwasmorestable.

Acknowledgements

ThefinancialsupportbytheAlexandervonHumboldtFounda- tionwithintheResearchGroupLinkageProgram,byCOSTAction CM1301, by TÁMOP-4.2.2.A-11/1KONV-2012-0047 is acknowl- edged. The authors wish to thank Mrs. Kornélia Baán for the preparationandcharacterizationofthecatalysts.

References

[1]L.F.Brown,Int.J.HydrogenEnergy26(2001)381–397.

[2]A. Haryanto,S.Fernando, N.Murali,S.Adhikari,EnergyFuels19(2001) 2098–2106.

[3]P.R.delaPiscina,N.Homs,Chem.Soc.Rev.37(2008)2459–2467.

[4]A.Yee,S.J.Morrison,H.Idriss,Catal.Today63(2000)327–335.

[5]L.V.Mattos,B.H.Jacobs,F.B.Noronha,Chem.Rev.112(2012)4094–4123,and referencestherein.

[6]J.Llorca,N.Homs,P.R.delaPiscina,J.Catal.227(2004)556–560.

[7]S.S.-Y.Lin,D.H.Kim,S.Y.Ha,Catal.Lett.122(2008)295–301.

[8]H.Song,U.S.Ozkan,J.Mol.Catal.A318(2010)21–29.

[9]B.Bayram,I.I.Soykal,D.Deak,J.T.Miller,U.S.Ozkan,J.Catal.284(2011)77–89.

[10]I.I.Soykal,H.Sohn,U.S.Ozkan,ACSCatal.2(2012)2335–2348.

[11]S.S.Itkulova,G.D.Zakumbaeva,A.A.Mukazhanova,Y.Y.Nurmakov,Cent.Eur.J.

Chem.12(2014)1255–1261.

[12]Zs.Ferencz,K.Baán,A.Oszkó,Z.Kónya,T.Kecskés,A.Erd ˝ohelyi,Catal.Today 228(2014)123–130.

[13]M.Badlani,I.E.Wachs,Catal.Lett.75(2001)137–149.

[14]A.Erd ˝ohelyi,J.Raskó,T.Kecskés,M.Tóth,M.Dömök,Catal.Today116(2006) 367–376.

[15]M.Dömök,M.Tóth,J.Raskó,A.Erd ˝ohelyi,Appl.Catal.B:Environ.69(2006) 262–272.

[16]A.Yee,S.J.Morrison,H.Idriss,J.Catal.186(1999)279–295.

[17]J.Kaspar,P.Fornasiero,M.Graziani,Catal.Today50(1999)285–298.

[18]A.Trovarelli,CatalysisbyCeriaandRelatedMaterials,ImperialCollegePress, London,2002,pp.1–528.

[19]A.M.daSilva,K.R.Souza,L.V.Mattos,G.Jacobs,B.H.Davis,F.B.Noronha,Catal.

Today164(2011)234–239.

[20]M.Happel,J.Myslivecek,V.Johánek,F.Dvorak,O.Stetsovych,Y.Lykhach,V.

Matolín,J.Libuda,J.Catal.289(2012)118–126.

[21]M.Mavrikakis,M.Barteau,J.Mol.Catal.131(1998)135–147.

[22]C.Diagne,H.Idriss,A.Kiennemann,Catal.Commun.3(2002)565–571.

[23]P.-Y.Sheng,A.Yee,G.Bowmaker,H.Idriss,J.Catal.208(2002)393–403.

[24]M.Tóth,M.Dömök,J.Raskó,A.Hancz,A.Erd ˝ohelyi,Chem.Eng.Trans.4(2004) 229–234.

[25]J.Raskó,M.Dömök,K.Baán,A.Erd ˝ohelyi,Appl.Catal.A299(2006)202–211.

[26]J.Raskó,J.Kiss,Appl.Catal.A287(2005)252–260.

[27]C.J.Weststrate,H.J.Gericke,M.W.G.M.Verhoeven,I.M.Ciobica,A.M.Saib,J.W.

Niemantsverdriet,J.Phys.Chem.Lett.1(2011)1767–1770.

[28]D.R. Mullins, S.D. Senanayake, T.-L. Chen, J. Phys. Chem. C 114 (2010) 17112–17119.

[29]E.Martono,J.M.Vohs,J.Catal.291(2012)79–86.

[30]E. Martono,M.P.Hyman,J.M. Vohs, Phys.Chem.Chem.Phys. 13(2011) 9880–9886.

[31]M-.P.Hyman,J.M.Vohs,Surf.Sci.605(2011)383–389.

[32]M.S.Batista,R.K.S.Santos,E.M.Assaf,J.M.Assaf,E.A.Ticinalli,J.PowerSources 124(2003)99–103.

[33]L.Óvári,S.KrickCalderon,Y.Lykhach,J.Libuda,A.Erd ˝ohelyi,C.Papp,J.Kiss, H.-P.Steinrück,J.Catal.307(2013)132–139.

[34]Zs.Ferencz,A.Erd ˝ohelyi,K.Baán,A.Oszkó,L.Óvári,Z.Kónya,C.Papp,H.-P.

Steinrück,J.Kiss,ACSCatal.4(2014)1205–1218.

[35]A.Gazsi,A.Koós,T.Bánsági,F.Solymosi,Catal.Today160(2011)70–80.

[36]P.Burroughs,A.Hamnett,A.F.Orchard,G.Thornton,Chem.Soc.DaltonTrans.

17(1976)1686–1698.

[37]D.R.Mullins,S.H.Overbury,D.R.Huntley,Surf.Sci.409(1998)307–319.

[38]H.Idriss,C.Diagne,J.P.Hinderman,A.Kiennemann,M.A.Barteau,J.Catal.155 (1995)219–237.

[39]K.M.Cook,S.Poudyal,J.T.Miller,C.H.Bartholomew,Appl.Catal.A:Gen.449 (2012)69–80.

[40]G.Jacobs,T.K.Das,Y.Zhang,J.Li,G.Racoille,B.H.Davis,Appl.Catal.A:Gen.233 (2002)263–281.

[41]Y.Zhang,L.Chen,G.Bai,Y.Li,X.Yan,J.Catal.236(2005)176–180.

[42]S.S.-Y.Lin,D.H.Kim,M.H.Engelhard,S.-Y.Ha,J.Catal.273(2010)229–235.

[43]J.Zhang,Z.Zhong,X.-M.Cao,P.Hu,M.B.Sullivan,L.Chen,ACSCatal.4(2014) 448–456.

[44]B.G.Johnson,C.H.Bartholomew,D.W.Goodman,J.Catal.128(1991)231–247.

[45]J.M.Guil,N.Homs,J.Llorca,P.R.delaPiscina,J.Phys.Chem.B109(2005) 10813–10819.

[46]J.C.Vargas, S.Libs, A.C. Roger,A. Kiennemann,Catal. Today107 (2005) 417–425.

[47]D.A.Wesner,G.Linden,H.P.Bonzel,Appl.Surf.Sci.26(1986)335–356.

[48]D.-X.Ye,S.Pimanpang,C.Jezewski,F.Tang,J.J.Senkevich,G.-C.Wang,T.-M.Lu, ThinSolidFilms485(2005)95–100.