intrinsicallyadsence

IntroductionNanometer-sizedgoldparticlesonoxidesupportsareefficientcatalystsfortheselectivecatalyticoxidation(SCO)ofcarbonmonoxideunderconditionscompatiblewiththeoperationofPEMfuelcells.
NanosizedAu/-Al2O3catalystsareabletooxidizeCObetween20and70°CinanatmosphereofhydrogencombininghighCOconversionandsatisfactoryselectivitytoCO2(1).
Itisgenerallyagreedthatthecatalyticactivityofgolddependsonthesizeofthegoldparticles,butthenatureofsupportmaterial,thepreparationmethod,theactivationprocedurehavealsobeensuggestedtoplayakeyrole(2).
SitesatthegoldsupportinterfacehavealsobeenclaimedtoberesponsiblefortheactivityinCOoxidation(3).
StrainintheAuparticlesduetothemismatchofthelatticesattheinterfacewiththesupport(4)andtheeffectoflow-coordinatedsitesandroughness(4)havealsobeensuggestedasimportantfactorsforhighactivity.
Themostimportanteffect,concerningthecatalyticactivityofgoldnanoparticlesforlowtemperatureoxidationofCOisrelatedtotheavailabilityofmanylow-coordinatedgoldatomsonthesmallparticles(5).
Effectsrelatedtotheinteractionwiththesupportmayalsocontribute,buttoaconsiderablysmallerextent(5).
COoxidationovergroupVIIInoblemetalsisthemoststudiedcatalyticreaction.
However,selectiveoxidationofCOinhydrogen-richconditionsisnotaswellstudied(6,7).
CO(reactant)andCO2(product)sorptionprocessesarefundamentalelementarystepsforSCOandmoreover,theeffectofhydrogenontheirsorptionoversupportedAucatalystsremainsasubjectofsignificantinterest.
AmongthevarioussupportedAucatalysts,Au/Al2O3isperhapsonethathasshownthewidestvariationforCOoxidation,rangingfrombeingveryinactivetopracticallyasactiveasAu/TiO2(8).
ComparedtoAu/-Al2O3,multicomponentgold-basedcatalystsalsosupportedon-Al2O3andcombinedwithcommonmetaloxides,suchas:MnOx,MgO,FeOx(9,10)aswellasgoldcatalystssupportedonothermaterialssuchasAu/TiO2(11)andAu/FeOx(11,12)havebeenfoundtoexhibitevenmorepromisingcommercialperformancefortheSCOreaction.
However,basedonitssimplicity,awell-studiedAu/-Al2O3catalyst(1,9)hasbeenutilizedinthepresentworkasmodelsystem,inordertogainfundamentalinformationonthecatalyticbehaviourofgoldnanoparticlesandidentifytheactivesitesonAu/-Al2O3.
Inthispaper,newfindingswhichofferfurtherinformationconcerningthemechanismofCOpreferentialoxidationovernanosizedAuarepresented.
ThefirstfindingconcernsCO2formationintheabsenceofoxygenandhydrogenbothinbare-Al2O3aswellasinAu/-Al2O3(calledasCOdecomposition).
Athightemperatures(>200°C)inexcessofH2,overtheAu/-Al2O3catalyst,reversedwatergasshift(RWGS)reactionresultsinCO2consumptiontowardsCOandH2Oformation.
Finally,thekineticmeasurementsofthepresentworkindicatethathydrogenstronglyinfluencestheinteractionofCOonAu/-Al2O3,byweakeningCOadsorption.
OntheMechanismofSelectiveCOOxidationonNanosizedAu/-Al2O3CatalystsDGavrila,b,*,AGeorgakab,VLoukopoulosb,GKaraiskakisb,BENieuwenhuysaaLeidenInstituteofChemistry,LeidenUniversity,2300RALeiden,TheNetherlandsbPhysicalChemistryLaboratory,DepartmentofChemistry,UniversityofPatras,26504,Patras,GreeceAbstractNewfindingsgivefurtherinformationonthemechanismofcarbonmonoxideselectiveoxidationover-aluminasupportednanoparticlesizedgoldcatalysts.
a)CO2formation,increasingwithrisingtemperature,isobservedintheabsenceofhydrogenandoxygenpointingtoamodelofactivesitesconsistingofanensembleofmetallicAuatomsandacationicAuwithahydroxylgroup,b)Athightemperatures(>200°C)inexcessofH2,reversedwatergasshift(RWGS)reactionresultsinCO2consumptiontowardsCOandH2Oformation.
c)HydrogenstronglyinfluencestheinteractionofCOonAu/-Al2O3,byweakeningtheCOadsorption.
ThepresenceofhydrogenplaysanimportantrolebothdecreasingthestrengthofCObondingandinthepreventionofdeactivationandregeneration.
KeywordsInverseGasChromatography,RateConstants,Aucatalysts,-Al2O3supportedcatalysts,PEMFuel-cell,SelectiveCOoxidation,Hydroxylgroups,CationicAu.
*CorrespondingauthorE-mail:D.
Gavril@upatras.
grCurrentaddress:PhysicalChemistryLaboratory,DepartmentofChemistry,UniversityofPatras,26504,Patras,GreeceGoldBulletin200639/4192ExperimentalmethodsPreparationofnanometersizeAucatalystsThenanometersizedAu/-Al2O3catalystwithanintended5%wt.
compositionwaspreparedbyhomogeneousdepositionprecipitationwithurea.
Thismethodleadstothesmallestaveragegoldsize(3-5nm).
Allgoldcatalystswerefiltered,washedtoremoveCl-anddriedinairat80°Cforatleast16handthencalcinedinaflowofoxygenupto300°C(heatingrate5°C-min-1).
Thesampleswerekeptat300°Cfor2h,andthencooledtoroomtemperature.
ThemethodofpreparationandthesurfacecharacterizationofthecatalystusingAAS,XRD,HRTEM,XPSandMES,havebeenpresentedpreviously(1,9,10).
ThegoldloadingdeterminedbyAASwas5.
1±0.
3%,whilethemeandiameterofthegoldparticleswasmeasured4.
2nm(XRD)and3.
6±1.
4nm(HRTEM),respectively.
Thegoldcrystalliteswerefoundtosinterinaflowofoxygenatatemperatureof400°Candhigherinagreementwithliterature(13).
AlthoughXRD,HRTEMandXPSmeasurementsclearlypointtothepresenceofmetallicgold,thepresenceoftracesofoxidicgold(Au2OorAu2O3)cannotbeexcluded.
Activityexperimentsundercontinuous-flowTheflowsystemusedforthestudyofCOinteractionwiththestudiedcatalystisshowninFigure1.
Theactivityexperimentswerecarriedoutinalab-scaleflowreactor.
Typically0.
20gcatalystwasusedforeachreaction.
Priortomeasurementthecatalystswerereducedin-situinaflowof4vol%H2/Heat300°C(heatingrate10°C-min-1)andkeptatthistemperaturefor30min.
Areductivepre-treatmentineitherH2orCOisnecessaryinordertoobtainoptimalperformanceforCOoxidation,whichpointstoanintrinsicallyhigheractivityofthereducedgoldspecies.
Aftercoolingtoroomtemperaturethereactorwaspurgedwithheliumtoremoveallofthehydrogen.
Subsequently,thereactantflowwasintroduced.
Differentactivitytestswerecarriedoutinthefollowingseries:i)1.
2%vol.
COin~70vol%H2+~29vol%He,ii)4%vol.
COinHe,iii)2.
0%vol.
CO+2.
0%vol.
O2inHeandiv)0.
6%vol.
CO+0.
6%vol.
O2in~70vol%H2+~29vol%He.
Thetotalreactantmixtureflowratewas40ml-min-1.
Thereactantgaseswerefedseparatelytothesystemwithmassflowcontrollers(Bronkhorst),andledthroughasmallchambertoensurepropermixingbeforeenteringthereactor:Boththermocoupleandovenwereconnectedtoatemperaturecontroller(ShimadenSR50-2AE),allowingthereactiontobetemperature-programmed.
Theeffluentgaseswereanalyzedon-linebyagaschromatographequippedwithathermalconductivitydetector.
Theexperimentswerecarriedoutunderatmosphericpressure.
XRDmeasurementsofthefreshandspentgoldcatalystrevealednosinteringofthegoldparticles(dAu,fresh=4.
2nm)aftertheaboveactivitytests.
Activity/kineticmeasurementsundernoncontinuous-flowKineticmeasurementsandcomplementaryactivitytestswereperformedbyusingthenovelgaschromatographictechniqueofreversedflowgaschromatography.
Theexperimentalset-upofRF-GCispresentedinFigure2.
InRF-GC,anothercolumn(diffusioncolumn)isplacedperpendicularlyinthecenteroftheusualchromatographiccolumn(samplingcolumn).
ThecarriergasflowsGoldBulletin200639/4193MFC=massflowcontrollerFM=flowmeterGC=gaschromatographMFCHeCOO2H25/3bar5/3bar5/3bar5/3bar3barPressurevalve(mixingchamber)catalystbedvalveWastethermocoupleovenFMGCFigure1Schemeofthecontinuousfeedingactivityexperimentsset-upcontinuouslythroughthesamplingcolumn,whileitisstagnantinsidethediffusioncolumn.
IncontrastwithconventionalGC,wherethemobilephaseisthecentreofinterest,inRF-GCthesolidorliquidsubstanceplacedintothediffusioncolumnisunderinvestigation,asinthecaseofinversegaschromatography.
Thestationaryphase(catalystbedunderstudy)isplacedatthelowerclosedendofthediffusioncolumnandthedisplacementoftheinjectedsolute(CO)intothediffusioncolumndependsonitsinteractionwiththestationaryphase(e.
g.
adsorption,desorptionandsurfacereaction)aswellasitsdiffusionintothestagnantcarriergas(mobilephase),whileitisindependentofthecarriergasflow-rate.
ThesefeaturesmakeRF-GCanidealmethodnotonlyforactivitystudiesbutmainlyforkineticmeasurements.
AnotherfeatureofRF-GCisthesamplingprocedureofthephysicochemicalphenomenon,whichtakesplaceinthediffusioncolumn.
Thesamplingprocedureiscarriedoutbyusingafour-portvalve,enablingreversalsoftheflowofthecarriergasforashorttime,aswellastherestoringofitsflowinitsoriginaldirection.
Theabove-mentionedflowreversalsprocedureresultsinabriefenrichmentofthesolutequantityintothecarriergasandthus,extrachromatographicpeaksarecreatedonthecontinuousconcentration-timecurve(chromatogram).
Theextrapeaksaresymmetricalandtheirheightorareaisproportionaltotheconcentrationofthesoluteinthejunctionofthediffusionandsamplingcolumns,givingtoRF-GCahighersensitivityandaccuracy,succeedingaccuratedatacollectionandlimitingtheneedforcomputerdatareduction.
Theestimationofthevariouskineticparametersisdonefromplotsoftheheightsortheareasoftheextrachromatographicpeaksagainstthetimefromsolute'sinjection(whichareso-calledasDiffusionBands)andfromgeometricalcharacteristicsofthediffusioncolumn(suchasitslengthanditsvolume).
Thematerials,apparatusandcalculationsusedforthestudyofcarbonmonoxidesorptionprocessesundervarioushydrogenamountsinthegaseousfeed,byRF-GC,aswellastheadvantagesofthemethodhavebeenpresentedindetailinrecentpublications(14-17).
ResultsanddiscussionCriticalelementaryreactionstepssuchasCOandCO2adsorption,desorptionandsurfacebreaking/bondingoverAucatalystscanprovidesubstantialinsightsintocatalysisbygoldnanoparticles.
Intheframeworkofaforthcomingwork,thesorptionprocessesofcarbonmonoxide,oxygen(SCOreactants)andcarbondioxide(SCOproduct)overnanosizedAusupportedon-Al2O3havebeenkineticallystudied.
Apreliminarystudy,inwhichtheeffectofthegasfeedcompositionontheinteractionofCOwiththestudiedAu/-Al2O3catalyst,wascarriedoutbythecontinuousfeedingset-upofFigure1(c.
f.
ExperimentalMethodssection).
ThemostimportantfindingofthisstudywasthattheadsorptionofCOoverAu/-Al2O3,intheabsenceofoxygenfromthefeedingstream,resultsinCO2formation(COdecomposition).
TheexperimentaldataobtainedbyRF-GC(undernoncontinuousfeedingconditions)alsoascertainedthattheadsorptionofcarbonmonoxideonthestudiedAu/-Al2O3catalyst,intheabsenceofhydrogenandoxygen(carriergas100%He)resultsinCO2formation.
IdentificationofthetwosamplepeaksGoldBulletin200639/4194SeparationcolumnH2/Hecarrier-gasinletTCDectectorfour-portvalvesamplingcolumndiffusioncolumnD2D1z=Ly=0=0=l=l'+lz=0LL'll'zy=L2catalystbedCOinjectorFigure2Experimentalsetupusedforthekineticinvestigationoftheeffectofhydrogenincarbonmonoxidesorptionoverthestudied-Al2O3supportedAucatalyst,fromreversed-flowgaschromatographyshowsthatthefirstpeakbelongstotheadsorbateCO,whilethesecondonetoCO2.
TheCOtoCO2conversion,x,waschromatographicallyestimatedbyutilizingtheexperimentallymeasuredareasofthereactantACOanditsproductACO2(1)where:R=0.
966,istherelativemolarresponseofthethermalconductivitydetectorforCO2tothatforCOundertheusedexperimentalconditions.
TherespectivefindingsarepresentedinFigures3a(undercontinuousfeedingconditions)and3b(undernoncontinuousfeedingconditions).
TheinteractionofpureCOwiththestudiedcatalyst,undercontinuousfeedingconditions,wasinvestigatedbyfeedingamixtureof4%vol.
COinhelium.
Foreachstudy,afterstabilizationatroomtemperaturefor30mintworeactioncycles,eachconsistingofoneheatingandonecoolingbranch,wererecordedtomonitorhysteresis(ifpresent)andcatalyst(de)activation.
Afterthelastcoolingstage,thecatalystwaskeptatroomtemperatureforanextendedperiodoftime(~4h)toevaluatethechangeinactivityduringtimeonstream.
Thesecondheatingcurvewasfoundtobemorerepresentativeforthe"steadystate"activityandwasthususedforcomparisonofthecatalysts.
TheexperimentalfindingsofthisstudyaresummarizedinFigure3a.
TheconversionofCOtoCO2increasesfrom7%atambienttemperaturesto22%at280°C.
InthecaseoftheCOconversionexperiments,ahysteresiswasobserved,inwhichtheCOconversioninthecoolingbranchexceededthatoftheheatingbranchathightemperatures(>200°C).
Theoppositebehaviourwasobservedattemperatureslowerthan200°C,atwhichCOconversionintheheatingbranchexceededthatofthecoolingbranch.
Atambientconditionsaconstantactivityvaryingfrom7to11%wasobserved(c.
f.
Figure3a).
AfirstobviousquestionarisingfromtheexperimentallyobservedCO2formationintheabsenceofO2iswhetherthisactivityisduetogoldorthesupport.
Forthisreason,similarexperimentswerecarriedoutonthebare-aluminasupport.
Figure3bsummarizesthecomparativeactivitystudyofCO2formationoverbothAu/-Al2O3catalystandbare-Al2O3.
ThisstudywascarriedoutbyRF-GC.
AsmallamountofCO(1mlunderatmosphericpressure)wasinjectedattheclosedendofthediffusioncolumn(c.
f.
Figure2).
Sinceeachkinetictestiscarriedoutataconstanttemperature,theusedmethodologyofRF-GCpermitsbetterequilibrationofthecatalyticactivity.
Theexperimentaldataobtainedinthecaseof-Al2O3showthattheinteractionofcarbonmonoxideintheabsenceofoxygen(carriergas100%He)doesnotresultinCO2formationattemperatureslowerthan150°C.
However,CO2formationisdetectedabove150°CoverbareAl2O3,significantlyincreasingwithrisingtemperature,whereasoverAu/-Al2O3,carbondioxideformationisalreadyobservedinthetemperaturerange50-150°C.
ThisisthefirsttimethatCOdecompositiononAunanoparticlesisobserved.
However,forNOdecompositiononAusurfaceswasreportedrecently.
IthasbeendemonstratedbeforethatcertainsteppedgoldsurfacesareabletoformN2OfromadsorbedNO(18):2NON2O+Oads(2)whereasthedenselypackedAusurfacescannotdecomposeNO.
ThemainquestionisrelatedtothemechanismresponsiblefortheobservedCO2formation.
AfirstpossibilitycouldbeCOdissociation,asinthecaseovergroupVIIInobleGoldBulletin200639/419528COdecompositionintheadsenceofH2overAu/-Al2O3undercontinuousfeeding2ndheatingbranch%COtoCO2conversionT/°C2ndcoolingbranch262422201816141210850100150200250300Figure3aa)HeatingandcoolingbranchesoftheconversionofCOtoCO2intheabsenceofoxygen,overthestudiedAu/-Al2O3catalystversusreactiontemperature,measuredatcontinuousfeedingconditionsT/°C%COtoCO2conversionCOdecompositionintheabsenceofH2studiedbyRF-GCAu/-Al2O3pure-Al2O3252015105050100150200250300Figure3bb)ComparativestudyofCOdecompositionoverAu/-Al2O3catalyst()andthesupport-Al2O3(),measured,byreversed-flowgaschromatographyRACO2xRACO1+ACOmetals,suchas:Pt,Rh,Pdetc(17).
However,thereisstrongevidencebothfromexperimentalaswellastheoreticalstudiesthattheadsorptionofCO,overevennanosizedanddefectiveAuparticles,istooweakforCOdissociation(3-5,19,20).
Moreover,CO2formation,alsoobservedoverbarealumina,attemperatureshigherthan150°C,doesnotindicatethenecessityofgoldfortheobserveddecompositionofCO.
Alternatively,CO2couldbeformedastheresultofCO-COcoupling,forexamplebyformationofdicarbonylsonthesurface.
Bearinginmindthereductivepre-treatmentofthecatalystandbare-Al2O3withH2,thereactionofCOwithsurfacehydroxyl,asinastepinwatergas-shiftreactionmightseemmuchmorelikely.
Costelloetal.
(8)studyingthepotentialroleofhydroxylgroupsinCOoxidationoverAu/Al2O3suggestedamodeloftheactivesiteinvolvinganensembleofmetallicAuatomsandAu+-OH-.
WhilethereisspectroscopicevidenceofthepresenceofAucationsinanactivecatalyst,althoughtheirroleintheCOoxidationreactionisnotacceptedunequivocally(10),thereisonlyinferentialevidenceofthepresenceandparticularlytheparticipation,ofhydroxylgroupsinthereaction(8).
WhilethedecompositionofCOathightemperatures,overbare-Al2O3isprobablyrelatedtoAl-hydroxyls,thelowtemperatureactivityofAu/-Al2O3forCOdecompositionmuchmorelikelyindicatesthatAu-hydroxylsareinvolvedintheformationofCO2.
However,aAu-OHgroupaloneisnotsufficientforCOdecomposition.
AcorrespondingreactionmechanismforthelowtemperatureactivityofAu/-Al2O3forCOdecompositionmightinvolvetheinsertionofCOintoAu+-OH-toformAu-hydroxycarbonyl(8),whichisthenoxidizedtoabicarbonyl,mostprobablyinteractingwithOfromneighbouringAl-hydroxyl.
DecompositionofthebicarbonateremovesahydroxylfromAlandproducesCO2andAu-hydroxylcompletingthereactioncycle.
Thepresenceofsurfacehydroxylgroupsisalsosupportedfromtheexperimentalobservationthatwaterhasabeneficialeffect,enhancingtherateofCOoxidationoverthestudiedAu/-Al2O3catalyst(1,9,10),furtherindicatingareactionpathwayinwhichsurfaceOHgroupscanprovideOneededforCOoxidation,inagreementwithHaruta'sobservationthatwateronthecatalystismoreimportantthanwaterinthegasphase(21).
Similarly,thepresenceofhydrogeninthefeedhasbeenfoundtocausechangesinCOoxidationactivityasthosediscussedforH2O.
Thus,itisnotinconceivablethathydrogenispartlypresentasH2OorsurfaceOH(1,9,10).
Recently,thereisreportedspectroscopicevidenceofthepresenceofcationicgoldintheactivecatalysts(Ref.
8andRefs.
therein).
Inaddition,althoughXRD,HRTEMandXPSmeasurementsofthestudiedcatalystclearlypointtothepresenceofmetallicgold,thepresenceoftracesofoxidicgold(Au2OorAu2O3)hasnotbeenexcluded(1,9,10).
COsorptionprocessesarefundamentalelementarystepsforSCOandmoreover,theeffectofhydrogenontheirsorptionoversupportedAucatalystsremainsasubjectofsignificantinterest.
Inarecentwork,COadsorption,desorptionandsurfacebonding,werekineticallystudiedoversilicasupportedmonometallicRhandRh0.
50+Pt0.
50alloycatalysts,underawiderangeofhydrogenatmospheres:25%-75%H2,bymeansofRF-GC(14).
ThemainfindingofthisworkwasaH2-induceddesorption,whichexplainsthatdescribedintheliteratureenhancedrateofSCOoxidation.
TheinteractionofcarbonmonoxidewiththestudiedAu/-Al2O3catalystcanbeoutlinedinakineticscheme,inwhichtheelementaryreversiblestepsofadsorption,k1,anddesorption,k-1,arefollowedbycarbonmonoxide'ssurfacebonding,k2.
Although,irreversibleCOadsorptionoverAu/Al2O3hasbeenreported(22),itisgenerallyassumedthatthenoblenatureofgoldmakesthestrongchemisorptionofCO(aswellasOatomsandmolecules)unlikely,evenwhenlowcoordinatedanddefectiveAuatomsexist(19,20).
GoldBulletin200639/4196Table1Rateconstantsfortheadsorption,k1,desorption,k-1andsurfacebonding,k2,ofCOoverAu/-Al2O3atvarioustemperatures,intheabsenceaswellasinexcessofhydrogen,determinedbyRF-GCT/°CT/K10k1/s-1104k-1/s-1104k2/s-10%H275%H20%H275%H20%H275%H230303-0.
83-6.
88-2.
1440313-0.
97-6.
35-2.
23503230.
460.
993.
498.
013.
242.
4260333-1.
00-8.
13-2.
50703430.
541.
025.
278.
453.
432.
5280353-1.
08-9.
28-2.
5990363-1.
15-8.
79-2.
661003730.
821.
256.
788.
843.
652.
701504231.
511.
489.
0313.
03.
892.
762004731.
971.
7010.
117.
94.
352.
872505233.
102.
0811.
624.
35.
162.
963005735.
952.
6413.
928.
06.
033.
23Consequently,inthepresentkineticmodel,k2,providesmoreameasureofthestrengthofCObondingthanofthereversibilityorirreversibilityofCOadsorptionoverthestudiedcatalyst.
ThemeasuredratesrelatedtotheeffectofhydrogenonCOsorptionoverAu/-Al2O3aresummarizedinTable1.
ThevaluesofthefoundratesarecomparablewiththosedeterminedbyFrequencyResponseMethodfortheadsorptionofCOonsilicasupportedPtcatalysts(23)ascertainingthepotentialofRF-GCforaccurateandreliablekineticmeasurements.
Atlowtemperatures(T150°C)adsorptionratesintheabsenceofhydrogenarehigherthanthoseinexcessofhydrogen,whichmayberelatedtotheobservedCOdecomposition,whichactivitybecomesmoreintenseattemperatureshigherthan150°C.
Moreover,thesatisfactoryselectivityofAu/-Al2O3forSCOatlowtemperaturescanbefurtherexplainedfromthefactthattherateofadsorptionisobviouslyhigherinexcessofhydrogen,duetothewellknownblockingorpoisoningactionofCO,whichpreventsthedissociativeadsorptionofothergases,suchasO2andH2(6,7).
Incontrast,desorptionratevaluesinexcessofhydrogenarehigherthanthoseintheabsenceofhydrogen,indicatingabeneficialinfluenceofhydrogenonCOdesorptionrate.
Carbonmonoxide'sdesorptionisconsideredasthedeterminingstepforthecatalyticoxidationofCOongroupVIII(Pt,Rh,Pdetc.
)noblemetalcatalysts(6,7,25).
However,thebeneficialinfluenceofhydrogenonCOdesorptionisnotdeterminingforthelow-temperatureCOoxidationactivityofgold,sinceitisgenerallyassumedthatthemildbondingcharacterofgoldisthebasisofAu-basedcatalystsinoxidationprocessesatlowtemperatures(19,20).
Jiaetal.
(22),recentlyobservedtwokindsofadsorbedCOonulltrafinegoldparticlesofaAu/Al2O3catalyst.
TheultrafinegoldparticlescatalyzedtheoxidationofreversiblybondedCOat150K,butdidnotconverttheirreversibleCOtoCO2.
TherateconstantscorrespondingtoCOsurfacebondingoverthestudiedAu/-Al2O3catalystintheabsenceofhydrogen,k2,areobviouslyhigherthanthosecorrespondingtothesurfacebondingofCOonthecatalystsactivesites,inexcessofhydrogen,indicatingtheweakerandmorereversibleadsorptionofCO,inexcessofhydrogen.
Rossignoletal.
(26),suggestfortheeffectofhydrogenthatbeneficialeffectsofhydrogenontheCOoxidationreactionaremostlyrelatedtothepreventionofdeactivationandregeneration.
InthepresenceofH2,twophenomenashouldoccursimultaneously.
Oneofthese,havinganegativeeffectonCOoxidation,isacompetitionofadsorptionbetweenH2andCO.
Theotherone,whichhasabeneficialeffectonCOoxidation,wouldbetheappearanceofadditionalreactiveintermediatesproducedinthepresenceofH2.
Furthermore,thepresenceofH2intheCO+O2mixturenotonlypreventsdeactivationbutalsocontributesviahighlyoxidizingintermediatespecies,suchasHO-Ospecies,totheoxidationofCO.
ItisgenerallyacceptedthatthehighactivityofnanosizedAuforthelow-temperatureCOoxidationisduetoitsweakbondingabilitywithCOandatomicO(19,20).
Remediakisetal.
(20),concludethatthepresenceoflow-coordinatedAusitesiscrucialforCOoxidation.
Theyidentifiedtwotypesofmechanismsforsupportedcatalysts:i)Gold-onlymechanismsforwhichthereactiontakesplaceonlyonthenanoparticle,andii)Asecondbranchofsupport-assistedmetal/oxideboundarymechanismsforwhichthereactionsiteisatthemetal/oxideinterfaceandtheensembleneededforO2adsorptionrequiresbothlow-coordinatedAuatomsinthegoldparticleandothercentersontheoxide.
Liuetal.
(19),suggestthatsince,O2dissociation(O22Oad)onAusurfacesincludingparticlesisnotexpectedatlowtemperatures,COoxidationonAu/inactive-materials(e.
g.
-Al2O3)isexpectedtooccuronAustepsviaatwo-stepmechanism:aslowCO+O2CO2+Oreaction,andthenafastCO+OCO2reaction.
Ingeneral,thesamemechanismcanalsodescribeselectiveCOoxidationinexcessofhydrogen.
ThebeneficialeffectofhydrogenconcerningthelowtemperatureCOoxidationcanbesimilarlyexplainedasresultoftheweakeningofthestrengthoftheCOandO2bondingwhichintensifiesthepossibilityofCO+O2reactiononAu.
Inaddition,H2dissociationhasbeenreportedonAu/Al2O3(26).
TheadsorbedHatomscanalsoaffectneighbourgoldatomsinsuchawaythattheycanfacilitatetheactivationofoxygenmolecules(e.
g.
throughHO-Ospecies)ordirectlyreactattheGoldBulletin200639/4197%COtoCO2conversionT/°CCO2ReverseWaterGasShiftreactioninthepresenceofH2overAu/-Al2O3studiesbyRF-GC6070805040302010230240250260270290300280Figure4TemperaturevariationofCOformationduetoCO2ReverseWaterGasShiftreaction,inthepresenceofH2,overthesameAu/-Al2O3catalyst,measuredundernon-continuousfeedingconditions,byRF-GCperimeteroftheAunanoparticlewithoxygenboundtothesupport(26).
Hydrogenadsorptionongold,partlyblockedatlowtemperaturesbyCO,explainsthehighactivityandsatisfactoryselectivitytowardsCO2formation.
AthighertemperaturestheCOdesorptionrateincreasesallowingmoreactivessitesforH2activationandreaction,resultinginlossofselectivity(1,9,10).
Similarlytohydrogenactivation,thedissociativeadsorptionofoxygenonensemblesofmetallicgoldatomsbecomesmoreimportantathighertemperatures,reducingthepossibilityofCO+O2reaction,whileinthesametimethepossibilityofhydrogenoxidationbecomesmoreimportant.
Thisisoughttothehigherenergybarrierofthehydrogenoxidationreaction(O+HOH,H+OHH2O,OH+OHH2O+O)comparedtothatofCOoxidation(CO+OCO2)(27).
Thus,whileCOoxidationismorefavoredatlowertemperatures,athighertemperatureshydrogenoxidationbecomesmoreimportant,explainingthedrasticdecreaseofSCOactivityandselectivity.
TheactivityofCOselectiveoxidationinexcessofhydrogenoverthestudiedAu/-Al2O3catalystishigh(rangingfrom95to80%)atambienttemperatures(200°C)inexcessofH2.
RWGSreactionresultsinH2andCO2consumptiontowardsCOandH2Oformation:CO2+H2CO+H2O(3)TheexperimentallymeasuredtemperaturevariationofCO2conversiontowardsCO,inexcessofH2,duetoRWGSreactionoverthestudiedAu/-Al2O3,isshowninFigure4.
ConclusionsNewfindingsgivefurtherinformationonthemechanismofcarbonmonoxideselectiveoxidationover-Al2O3supportednanoparticlesizedAucatalysts.
ForfirsttimeisobservedthattheadsorptionofCO,overthestudiedAu/-Al2O3catalystintheabsenceofoxygen,resultsinCO2formation,pointingtoamodelofactivesitesconsistingofanensembleofmetallicAuatomsandacationicAuwithahydroxylgroup.
ThepresenceofsurfacehydroxylgroupsisalsostrengthenedfromthefactthatwaterhasabeneficialeffectenhancingtherateofCOoxidationoverthestudiedAu/-Al2O3catalyst,furtherindicatingareactionpathwayinwhichsurfaceOHgroupscanprovideOneededforCOoxidation.
ThelowtemperatureactivityofAu/-Al2O3fortheso-calledCOdecompositionindicatesthatAu-hydroxylsareinvolvedintheformationofCO2,whichhoweverarenotsufficientforcompletingCOdecomposition.
AcorrespondingreactionmechanismsuggestedinvolvingtheinsertionofCOintoAu+-OH-toformAu-hydroxycarbonyl,whichisthenoxidizedtoabicarbonyl,mostprobablyinteractingwithOfromneighbouringAl-hydroxyl.
DecompositionofthebicarbonateremovesahydroxylfromAlandproducesCO2andAu-hydroxylcompletingthereactioncycle.
Atlowertemperatures(T200°C).
AcknowledgementsDGavrilthanksWorldGoldCouncilforthefinancialsupportintheframeofaGROWsponsoredproject.
References1R.
J.
H.
Grisel,B.
E.
Nieuwenhuys,J.
Catal.
,2001,199,482M.
Haruta,Catal.
Today,1997,36,1533L.
M.
Molina,B.
Hammer,Phys.
Rev.
Lett.
,2003,90,2061024M.
Mavrikakis,P.
Stoltze,J.
K.
Norskov,Catal.
Lett.
,2000,64,1015N.
Lopez,T.
V.
W.
Janssens,B.
S.
Clausen,Y.
Xu,M.
Mavrikakis,T.
Bligaard,J.
K.
Norskov,J.
Catal.
,2004,223,2326S.
H.
Oh,S.
Sinkevitch,J.
Catal.
,1993,142,2547M.
J.
Kahlich,H.
A.
Gasteiger,R.
J.
Behm,J.
Catal.
,1997,171,938C.
K.
Costello,J.
H.
Young,H.
-Y.
Law,Y.
Wang,J.
-N.
Lin,L.
D.
Marks,M.
C.
Kung,H.
H.
Kung,Appl.
Catal.
A,2003,6399,19R.
J.
H.
Grisel,PhDThesisLeidenUniversity,TheNetherlands,200210R.
J.
H.
Grisel,C.
J.
Westrate,A.
Goosens,M.
W.
J.
Craje,A.
M.
vanderCraan,B.
E.
Nieuwenhuys,Catal.
Today,2002,72,22311M.
Haruta,N.
Yamada,T.
Kobayashi,S.
Iijima,J.
Catal.
,1989,115,30112P.
Landon,J.
Ferguson,B.
E.
Solsona,T.
Garcia,A.
F.
Carley,A.
A.
Herzing,C.
J.
Kiely,S.
E.
Golunski,G.
J.
Hutchings,Chem.
Commun,2005,27,338513S.
Tsubota,T.
Nakamura,K.
Tanaka,M.
Haruta,Catal.
Lett.
,1998,56,13114D.
Gavril,V.
Loukopoulos,A.
Georgaka,A.
Gabriel,G.
Karaiskakis,J.
Chromatogr.
A,2005,1087,158GoldBulletin200639/419815D.
Gavril,B.
E.
Nieuwenhuys,J.
Chromatogr.
A,2004,1045,16116V.
Loukopoulos,D.
Gavril,G.
Karaiskakis,Instrum.
Sci.
Technol.
,2003,31,16517D.
Gavril,V.
Loukopoulos,G.
Karaiskakis,Chromatographia,2004,59,72118C.
P.
Vinod,J.
W.
Niemantsverdriet,B.
E.
Nieuwenhuys,Apl.
Catal.
A,2005,291,9319Z-P.
Liu,P.
Hu,A.
Alavi,JACS,2002,124,1477020I.
N.
Remediakis,N.
Lopez,J.
K.
Norskov,Appl.
Catal.
A,2005,291,1321M.
Date,M.
Haruta,J.
Catal.
2001,201,22122J.
F.
Jia,J.
N.
Kondo,K.
Domen,K.
Tamaru,J.
Phys.
Chem.
B,2001,105,301723Y.
E.
Li,D.
Wilcox,R.
D.
Gonzalez,AIChEJ,1989,35,42324R.
J.
H.
Grisel,B.
E.
Nieuwenhuys,Catal.
Today,2001,64,6925B.
E.
Nieuwenhuys,AdvCatal,1999,44,25926C.
Rossignol,S.
Arrii,F.
Morfin,L.
Piccolo,V.
Caps,J.
L.
Rousset,J.
Catal,2005,230,47627S.
Kandoi,A.
A.
Gokhale,L.
C.
Grabow,J.
A.
Dumesic,M.
Mavrikakis,Catal.
Lett.
,2004,93,93GoldBulletin200639/4199