铜锰协加载介孔材料ZrO2 TiO2复合材料催化合成及其CO催化氧化性(2)

2.4.Catalyticactivitymeasurement

Catalyticactivitytestswereperformedinacontinuous-?ow?xed-bedmicro-reactor.About100mgcatalystpowderwasplacedintothereactortube.Thereactiongasmixtureconsistingof1vol%CObalancedwith21vol%O2and78%N2waspassedthroughthecatalystbedatatotal?owrateof50mlminà1.Atyp-icalweighthourlyspacevelocity(WHSV)was30,000mlhà1gà1.TheconversionofCOwasmeasuredfromroomtemperaturetothetemperatureof100%COconversionbyaGC-2060gaschro-matographequippedwitha?ameionizationdetector.TheactivitywasexpressedbytheconversionofCO.3.Resultsanddiscussion

3.1.Structuralandmorphologicalanalyses

3.1.1.XRDanalysis

TheXRDpatternsoftypicalZTsamplesatdifferentcalcinationtemperaturespresentedinFig.1(a)indicatedthatZTmatrixwithaZr/Timolarratioof7:3hadanamorphousphasestructureaftercalcinationatnothigherthan600°C.Inaddition,themaindiffrac-tionpeakscorrespondingtocrystallineZrTiO4phase(JCPDS

No.

34-0415)couldbeclearlyobservedaftercalcinatedat700°Casre-ported[26],suggestingthatnophaseseparationstookplaceduringthethermaltreatment.

X-raydiffractionpatternsofthepreparedZTmatrixafterfur-therloadingCuand/orMnoxidespecieswereshowninFig.1(b).AlltheCuand/orMnloadedcatalystswereamorphousandnosep-arateCuand/orMnrelatedcrystallinecompoundscouldbede-tected,whichimpliedthatCuandMnspeciescouldhavebeensuccessfullyincorporatedintomesoporousZTframeworkand/orhomogeneouslydispersedintothemesoporouschannelswithverysmallparticlesize.

3.1.2.N2sorptionanalysis

TheresultsofN2sorptionanalysisofthepreparedmesoporous

fromthestrongoxidabilityofKMnO4solutiontowardsZTmeso-porouswall,aswellasitsinsituredoxreactionwiththeCTABsur-factantwithintheporechannelsofmatrix.Beforefurtherthermaltreatment,themesoporousZTframeworkwouldcontainalargeamountofchemicalbondsofZr–OHandTi–OHasaresultofincompletehydrolysisandpolycondensation,whichweresubjecttothecorrosionofKMnO4andconsequentporeenlargement.Interestingly,however,itcouldbefoundthatfortheco-loadedCuMnZTsamples,boththesurfaceareaandthemostprobableporesizeshowedonlyaveryslightchangecomparedtoZTmatrix,whichcouldbeattributedtothegreatlyincreasedstabilityofZTinorganicwallowingtoincorporationofCuspecies.TheresultsofN2adsorptionanalysiscon?rmedthatfortheco-loadedsamples,Mnspeciescouldbehomogeneouslydispersedintothemesopor-

Y.Gongetal./MicroporousandMesoporousMaterials173(2013)112–120

Table1

BETsurfaceareasofmesoporousZTandZT-basedcatalystswithdifferentCuand/orMncontents.SampleZT

CuZT-aCuZT-bCuZT-cMnZT-aMnZT-bMnZT-cCuMnZT-aCuMnZT-b115

Cucontent(wt%)*–0.50.962.20–––0.50.96Mncontent(wt%)*––––

15.6218.1920.3218.1918.79SBET(m2/g)196131152115329359348177148ThetypicalSEMimageofmesoporousZTmatrixinFig.4(a)exhibitedaninterestingmorphologycomposedofuniformspheri-calnanoparticlesabout50nminsize,whichwasbene?cialfortheloadingofactivecomponentswithhighdispersion.Surprisingly,theMnZT-csampleshowedaneedle-likemorphology(Fig.4(b)),totallydifferentfromthatofZTmatrixwhichcouldbeattributedtotheindependentgrowingofMnspecieswithoutbeingcon?nedbythemesoporouschannelsofZTmatrixduetothestrongcorro-sioneffectofKMnO4towardsZTmesoporouswall.Inaddition,bothCuZT-candMnCuZT-csamplesshowedverysimilarmorphol-ogiestothatofZTmatrix,asshowninFig.4(c)and(d)respectively,indicatingthateitherCuorMnspeciescouldbewelldispersedinto

visspectrainFig.6.ThiscouldbeduetotheinteractionbetweenCu2+andMn3+ionsintheco-loadedsamples,andresultedinthevalencereductionofCu2+toCu+species.SimilarresultswerealsoreportedbyCharyetal.[14,29]andLietal.[29].

3.2.2.XPSresultsanalysis

Fig.7(a)showedaXPSsurveyspectrumofalltheelementsforCuMnZT-csample,con?rmingtheco-existenceofCuandMnspe-ciesinthissample.Interestingly,itcouldbefoundthattheco-loadedsampleCuMnZT-cexhibitedaslightlylowerpeakofCu2pwithabindingenergyatabout934.1eVandagreatlyweakenedshake-uppeak(about945–943eV)inFig.7(b),ascomparedtothesingle-loadedCuZT-csample.Thisindicatedtheexistenceofreductionstateofcopperspecies(Cu+)intheCuMnZT-ccatalyst[32,33]duetothestronginteractionbetweenCuandMnspeciesintheco-loadedsampleasinagreementwiththeabovesection.

theabsorptionbandsdistinctivelydecreasedoncedopedwithCuspecies,suggestingthatCuprecursorscouldreactwithmostofOHgroupsinZTframeworkbeforecalcinationresultedfromtheincompletepolymerizationofZTprecursor.Therefore,itwasbe-lievedthattheincorporationofCuspeciescouldremarkablypro-motethehydrolysisandcondensationofZTprecursorbyreactingwithOHgroups,andinthemeantime,bewellincorpo-ratedintoZTframework,resultinginalmostcompletedcondensa-tionandgreatlyenhancedstabilityofmesoporousZTwall.Furthermore,itwasconsideredthatZTinorganicwallwithmuchenhancedstabilitybecamehighlyresistanttothecorrosionbyKMnO4.Therefore,suchco-loadedCuMnZTsamplespresentedcomparablestructuralandmorphologicalcharacterstoZTandCuZTsamples,ratherthantothesingleMnloadedsamples.

ataround184°Ccompaniedbyaweakshoulderpeakataround240°C,whichwasattributedtothehighlydispersedCuspeciesinthemesoporousZTmatrix.Moreinterestingly,fortheCuM-nZT-csample,theTPRpro?leshoweddistinctivelydifferentH2consumptionpeaksfromeitherCuZT-corMnZT-csample,withtwoapparentpeaksataround194and250°C,accompaniedbyasmallshoulderpeakat215°C,denotedasa,c,b,respectively[24].IthasbeenreportedthattheapeakcouldbeassignedtothereductionofthehighlydispersedCuOcomponent,andthebpeakcouldbeattributedtotheredoxinteractionbetweenthedis-persedCuspeciesandMnspeciesintheCuMnZT-csample.Thecpeakataround250°CcouldbeowingtothereductionofMnspe-ciesontheCuMnZT-csample,whichexhibitedmuchlowerreduc-tiontemperaturesthanthatofsingleMnloadedsampleMnZT-c[29].Importantly,itcouldbefoundthattheTPRpeakintensityoftheCuMnZT-cdemonstratedasigni?cantincreasecomparedwiththereferenceCuZTandMnZTsamples,whichcouldbeattrib-utedtothedistinctivesynergisticeffectbetweentheCuandMnspecies.Fig.8(b)presentedtheTPRpro?les

sampleswithdifferentCu/Mnweightratios.CuMnZT-bshowedsimilarTPRpro?lestoMnZT-catrelativelylowloadingamountsthan1wt%).Attheincreasedcontentof2.2wt%,thepeakintensityofH2consumptioncreaseanditspositionshiftedtolower?rmingtheexistenceofsynergisticeffectspeciesoncetheCucontentwashighenough.

3.3.2.COcatalyticoxidation

AmodelCOcatalyticoxidationreactiongatethecatalyticactivitiesofthepreparedcatalysts.ItcouldbefoundfromFig.9that,thealystshowedanincreasedcatalyticactivityamountofCuspecies,andtheCOcompleteturewasashighas350°C.WhileforMnZTactivityshowednoapparentchangeswithloading,andtheCOcompleteconversion250°C,whichwasmuchlowerthanthatofMnloadedmesoporoussilicasamplewithamountandsyntheticprocedure[25],ZTsupporthasasigni?cantcontributiontoalyticactivitybyhighlydispersedMnspecies,bythespecialMn-loadingstrategyoftheinployedinthisstudy.

Moreinterestingly,thepreparedgreatlydifferentCOoxidationactivityForexample,theCuMnZT-awiththelowestCucontentorCu/MnratioshowedthesimilarCOcatalyticactivitytothatofMnZT,andCuMnZT-bpresentedalimitedenhancementintheactivity(i.e.,decreaseinthereactiontemperature).Remarkably,CuMnZT-candCuMnZT-dcatalystswithCucontentsofhigherthan2.2wt%showedthegreatlyenhancedcatalyticactivity,andtheCOcompleteoxidationtemperaturewasaslowas180°C,whichwasmuchlowerthanthoseofCuZT(350°C)andMnZT(250°C)samples.Furthermore,itcouldalsobenotedthatthecatalyticactivitycouldnotfurtherenhancewithfurtherincreasedCucon-tentfrom2.2to5.35wt%.Thisindicated,aswebelieve,thatthereisasynergisticeffectbetweenCuandMnspeciesinthemesopor-ousZTsupportwhentheCuamountwashigherthan2.2wt%,andsuchasynergeticeffectwasabsentorinsigni?cantattheCucon-tentlessthan2.2wt%.Inbrief,itwasthesynergeticcatalyticeffectthatstronglyenhancedthecatalyticactivityinCOoxidation,whichwillbefurtherdiscussedindetailinthefollowingsection.

treatment.Inthelaterhydrothermaltreatmentandaftercalcina-tion,thesurplusCuspeciescouldsegregateoutoftheframeworkintothemesoporechannels,dependentonthesolidsolubilityofCuOxintheZTmatrix.ItwasthereforeinferredthatattheCucon-tentofnohigherthan2.2wt%,CuspeciesmainlyremainedwithintheframeworkintheCuMnZTsampletosupportthestabilityofZTinorganicwall,andincontrast,attheCucontentof2.2wt%,acon-siderableamountofthemhavesegregatedfromtheZTframeworkandthereforedispersedintotheinnerporesurface,i.e.,inthemes-oporechannelswithverysmallparticlesizes,whichcouldbeinclosecontactwiththelaterintroducedMnspecies,asinagreementwiththeforegoinganalysis.Thereforenodistinctivelysynergeticcatalyticeffectcouldbedetectedat<2.2wt%CucontentofsamplesCuMnZT-aandCuMnZT-b,andcomparativelythesynergeticeffectbecameprofoundwhentheCuamountwasat2.2wt%orabove,whereCuOxwaswellincontactwithMnspeciesinthemesopores.Inthecaseofsigni?cantsynergeticeffectbetweenCuandMnspecies,theunderlinesynergeticmechanismwillbediscussedasfollowsbasedonourexperimentalresultsonthecatalystcharac-terizationandcatalyticperformance,asindicatedinScheme2.IthasbeenproposedthatCOoxidationmayproceedviatheinteractionbetweenadsorbedCOandO(Langmuir–Hinshelwoodmechanism)[37,38]asfollows: