Spectroscopic investigation of sulfur-resistant Pt/K 2 O/ZrO 2 /TiO 2 /Al 2 O 3 NSR/LNT catalysts

ContentslistsavailableatScienceDirect

Catalysis Today

jo u rn al h om ep a g e :w w w . e l s e v i e r . c o m / l o c a t e / c a t t o d

Spectroscopic investigation of sulfur-resistant Pt/K 2 O/ZrO 2 /TiO 2 /Al 2 O 3 NSR/LNT catalysts

Z. Say, M. Tohumeken, E. Ozensoy

DepartmentofChemistry,BilkentUniversity,06800Ankara,Turkey

a r t i c l e i n f o

Articlehistory:

Received6October2015 Receivedinrevisedform 25November2015 Accepted4December2015 Availableonline18January2016

Keywords:

NSR/LNT

DeNO_xAl2O3/ZrO2/TiO2

FTIR TPD

a b s t r a c t

AnalternativeternarysupportoxidematerialanditsK2OandPtfunctionalizedcounterpartsintheform ofPt/K2O/Al2O3/ZrO2/TiO2withdifferentK2Oloadingsweresynthesized.Structuralandmorphological propertiesofthecatalystswerecharacterizedviaXRDandBETtechniquesincomparisontoaconven- tionalPt/20Ba/AlbenchmarkNSR/LNTcatalyst.Comprehensivein-situFTIRandTPDanalysisrevealed thatincreasingtheK2OloadinginthePt/K2O/AZTsystemleadstoanincreaseinNO_xStorageCapacity (NSC)attheexpenseoftheformationofbulk-likesulfatesrequiringhighertemperatureforcomplete sulfureliminationwithH2(g).Observeddelicatetrade-offbetweenNSCandsulfurpoisoningtenden- ciesofthecurrentlyinvestigatedfamilyofAZT-basedNSR/LNTcatalystsimpliesthatPt/5.4K2O/AZTisa promisingcatalystrevealingcomparableNSCwithinthetemperaturerangeof473–673Ktothatofthe conventionalPt/20Ba/Albenchmarkcatalyst,whileexhibitingsuperiorsulfurtoleranceandregeneration characteristics.

©2016ElsevierB.V.Allrightsreserved.

1. Introduction

NO_x emitted from mobile sources have serious destructive effectsontheatmosphere,globalecosystemandespeciallyonthe humanhealth.AboutonehalfofthetotalNO_xemissionsresults frommobilesources[1].Whiletheveryfirstregulationsfordiesel engineemissionswereprimarilyfocusing onparticle emissions, otherhazardouspollutantssuchasCO,SO₂,NO_xand unburned hydrocarbonsfrommobilesourcesarecurrentlybeingregulated withincreasingly stringentlimitations. Thisleadstoa constant pressureontheglobalautomotiveindustrytodevelopnoveland innovativeaftertreatmenttechnologiesthatcansatisfythecon- tinuously evolvingenvironmentallegislations and to lowerthe exhaustemissionlevels[2–4].Recently,itwasreportedthatNO_x emissionsofsomeofthecurrentlyexistingdiesel-enginepassenger carsequippedwithmodernDeNO_xaftertreatmentsystemsonthe highwaywereupto20–35timeshigherthanthatoftheallowed emissionlimits[5].Furthermore,averyrecentstudypublishedby theEuropeanEnvironmentAgency(EEA)[6]reportedthatwithout

∗ Correspondingauthor.

E-mailaddress:ozensoy@fen.bilkent.edu.tr(E.Ozensoy).

anyexception,eachEuropeanUnion(EU)memberstateviolates atleastoneormoreoftheexistingannualemissionlimitations associatedwithNO_x,SO_x,NH3andnon-methanevolatileorganic compounds(NMVOC).AmongtheseEUmemberstates,particularly Germany,AustriaandIrelandwerefoundtofailmeetingannual EuropeanNO_xemissionstandardsin2014.Thesestrikingexamples clearlycallforthedesignanddevelopmentofmoreefficient,more stableandmoreaffordableheterogeneouscatalyticarchitectures thatcanbeusedinmodernDeNO_xaftertreatmenttechnologies.

Forlean-burnengines,apromisingaftertreatmentmethodfor thecatalyticNO_xreductionisthesocalledNOxstorage/reduction (NSR)/LeanNOxTrap(LNT)technology[7,8].AtypicalNSR/LNTcat- alystiscomprisedofbasicoxides(e.g.BaO,K₂O),redoxsites(e.g.

Pt,Pdand/orRh)and ahighsurfaceareasupportmaterial(e.g.

-Al2O3)[2,3,9].

The conventional NSR/LNT catalyst, Pt/BaO/␥-Al2O3, exhibits efficientNO_xconversionandstorageperformancewithintheoper- ationaltemperaturewindowofthedieselemissiontailpipe(i.e.

473–673K)[10–18].However,recentengineapplicationssuchas thefuel-efficientgasoline directinjection(GDI)engines require catalyticaftertreatmentsolutionswhichshouldbeabletooper- ateattemperaturesabove400^◦C,wheretheconventionalNSR/LNT catalystscannotfunctioneffectively[19].ToyotaMotorCompany http://dx.doi.org/10.1016/j.cattod.2015.12.013

0920-5861/©2016ElsevierB.V.Allrightsreserved.

reportedthatK₂OandBaOaretwoofthemostpromisingNO_xstor- agecomponentstobeusedinNSR/LNTcatalysts[20].Amongthese twodifferenttypesofbasicmetaloxides,theuseofK₂Oattracted particularinterestduetoitssuperiorNO_xstoragecapacity(NSC)at elevatedtemperatures[21].OthernoteworthyadvantagesofK2O domainsareassociatedtotheirstrongerbasicityandthelackof unfavorablesolid-stateinteractionsbetweenK₂Oandthe-Al2O₃ supportmaterial,unlikethatofBaOwhichmayleadtothefor- mationofundesiredBaAl₂O₄athightemperatures[22].Luoetal.

investigatedtheeffectofK₂Oloading(within2–20wt.%)onthe NSCofthePt/K2O/␥-Al2O3system.Itwasfoundoutthatthecat- alystformulationcontaining10wt.%K2Oresultedinthehighest NSCvalueswithinawidetemperaturewindowof523–823K[22].

InadditiontothepromisingNSCofK₂O-functionalizedmate- rialsin high temperature DeNO_x applications, sulfur-poisoning tolerances as well as the sulfur regeneration characteristics of such systems should be also taken into consideration. It is knownthatK2Odomainsdispersedona␥-Al2O3 supportmate- rialarehighlypronetosulfurpoisoning,experiencingrapidand ratherirreversiblecatalyticdeactivation.AclassofnovelAl/Ti/Zr mixedoxides hasemerged in recent years with enhanced sur- faceandstructuralpropertiesassupportforK₂O-basedNSR/LNT catalysts [23–25]. In recent studies, ZrO₂/TiO₂, TiO₂/Al₂O₃ and Al2O3/ZrO2/TiO2-supportedNSR/LNTcatalystscanrevealsuperior sulfurregenerationandNO_xrecoveryperformancesascomparedto thatof␥-Al2O₃-basedsystems(i.e.Pt/BaO/Al₂O₃vs.Pt/K₂O/Al₂O₃) [26,27].Takashietal.reportedthataZrO₂:TiO₂supportmaterial withamassratioof70:30(whichalsorevealedthehighestSSA amongtheinvestigatedmaterialstherein)exhibitedthebestper- formanceintermsofsulfurresistance,thermaldurabilityandNO_x abatement[28].TheirstudieswhichalsoincludedPt/Rh/Ba/K/AZT catalyst with nano-composite ternary oxide Al₂O₃/ZrO₂/TiO₂- supportshowedexcellentNO_xstoragecapacity(NSC)comparedto thatof␥-Al2O3/ZrO2/TiO2-support,where␥-Al2O3wasphysically mixedwithZrO₂/TiO₂[29,30].Inaddition,Zouetal.[27]performed adetailedanalysisontheeffectofAl₂O₃dopingintotheZrO₂/TiO₂ matrix,suggestingthattheAl:(Ti+Zr)atomicratioof3:1exhibited thehighestNSCforfreshandsulfur-regeneratedcatalyst.Inamore recentwork,Zouetal.studiedtheeffectofKloadingontheNSCand sulfurregenerationperformanceofPt/K/Al₂O₃/ZrO₂/TiO₂catalyst underrealisticflowconditions[31].

However, these aforementioned comprehensive studies includedalimitednumberofspectroscopicinvestigationsonthe interactionsbetweenSO_xspeciesandthecorrespondingcatalyst surfaces.Thus, inthecurrent work,we focusonthemolecular levelinvestigationofthefundamentalinteractionsthattakeplace betweenSO_xspeciesandK-basednovelNSR/LNTcatalystsurfaces in a qualitative and a semi-quantitative manner. Along these lines, we investigate the SOx adsorption/uptakeas well as the SO_xreduction/regeneration/releasepropertiesofAl2O3/ZrO2/TiO2

(AZT)supportedPt/K/AZTcatalystsincomparisontoabenchmark NSR/LNTcatalyst (i.e.Pt/BaO/Al₂O₃)byutilizing in-situ spectro- scopictechniques.GenerationofS-containingsurfacefunctional groups,theirthermalevolution,reductionandreleasesafunction oftemperatureandK₂Oloadingaresystematicallymonitoredby meansofin-situFourierTransformInfraredSpectroscopy(in-situ FTIR)andTemperatureProgrammedDesorption(TPD).Moreover, structuralandmorphologicalpropertiesofthesynthesizedmate- rialsarealsoanalyzedviaX-rayDiffraction(XRD)andBrunauer, Emmettand Teller(BET)surface areaanalysis techniques.Cur- rent resultsprovide valuable molecular level insight regarding the interaction of SO_x species withK-based NSR/LNT catalysts supportedonnovelAZT mixed oxidesurfaces and thedelicate trade-offbetweentheNSCandsulfurpoisoningphenomena.

2. Experimental 2.1. Materialsynthesis

2.1.1. SynthesisofPt/Al2O3/ZrO2/TiO2

Al₂O₃/ZrO₂/TiO₂ (AZT) supportmaterial was synthesized as describedinoneofourformerpublicationswheretherelativecom- positionoftheternaryoxidesystem(i.e.Al2O3/ZrO2/TiO2)bymass was50:35:15[25].1wt. %platinum-incorporatedternary oxide materials were synthesized by incipient wetness impregnation methodusingasolutionofPt(NH3)2(NO2)2 (Aldrich,diammine- dinitritoplatinum(II),3.4wt.%solutionindiluteNH3(aq)).Priorto thePtaddition,Al₂O₃/ZrO₂/TiO₂wasinitiallycalcinedinairat973K for150mininordertoremovetheorganicfunctionalitiesinthe precursor.AfterthePt-incorporation,Pt/AZTmaterialwassubse- quentlycalcinedin airat973Kfor 150minin ordertoremove nitrite/nitrateoriginatingfromthePtprecursorandtostructurally stabilizethecatalystsurface.

2.1.2. SynthesisofPt/K₂O/Al₂O₃/ZrO₂/TiO₂

K₂O-based catalysts were also prepared via wetness impregnation. Pt/K2O/Al2O3/ZrO2/TiO2 catalysts with 2.7, 5.4 and 10.0wt. % K₂O loading (i.e. Pt/2.7K₂O/Al₂O₃/ZrO₂/TiO_2, Pt/5.4K₂O/Al₂O₃/ZrO₂/TiO₂ and Pt/10K₂O/Al₂O₃/ZrO₂/TiO₂; respectively)werepreparedviaimpregnationofAl2O3/ZrO2/TiO2

support(initially calcined at 973K for 150min) withan aque- ous solution of potassium nitrate (KNO₃·6H2O, >99.0 %, Fluka, France)followedbycalcinationat873Kfor150mininorderto thermallyremovethenitratecontentpresent intheprecursors.

Finally, K₂O/Al₂O₃/ZrO₂/TiO₂ structure was impregnated with thePt(NH₃)₂(NO₂)₂ precursorandcalcined at973Kfor150min under ambient conditions in order to attain 1wt. % nominal preciousmetalloading.Throughoutthecurrenttext,synthesized Pt/K₂O/Al₂O₃/ZrO₂/TiO₂ catalysts with 2.7, 5.4 and 10.0wt. % K2Oand 1wt.%PtloadingswillbeabbreviatedasPt/2.7K/AZT, Pt/5.4K/AZTandPt/10K/AZT,respectively.

2.1.3. SynthesisofPt/BaO/-Al2O₃

For the synthesis of the Pt/20BaO/Al benchmark catalyst,

␥-Al2O3supportmaterial(SASOLPuralox,210m²/g)wasimpreg- natedwithanaqueoussolutionofbariumnitrate(Ba(NO₃)₂,ACS Reagent,≥99%,Riedel-deHäen,Germany) whichwasfollowed bycalcination at873Kin air for 150min. Finally,20BaO/Al2O3

was impregnated with the Pt(NH₃)₂(NO₂)₂ precursor (Aldrich, diamminedinitritoplatinum(II),3.4wt.%solutionindiluteNH₃(aq)) toobtain1wt.%nominalpreciousmetalloading,followedbycal- cinationat973Kfor150min.Thiscatalystwillbeabbreviatedas Pt/20Ba/Althroughoutthecurrenttext.

2.2. Experimentalsetup

Comprehensivedescription ofthecustom-made batch-mode in-situ FTIR and TPD spectroscopic setup used in the current measurementscanbefoundelsewhere[10,14,25,32].Briefly,an FTIR spectrometer (Bruker Tensor 27) and a quadruple mass spectrometer(QMS,StanfordResearch Systems,RGA200)were simultaneouslyconnectedtoabatch-typespectroscopicreactor.

FTIRexperimentswereperformedintransmissionmode.TPDpro- fileswereobtainedundervacuumbyusingacomputer-controlled lineartemperaturerampof12K/minwithamaximumsampletem- peratureof1173K.

2.3. Experimentalprocedures

2.3.1. MonitoringSO_xadsorptionviain-situFTIR

Sulfuradsorption/poisoningcharacteristicsofeachmaterialwas investigatedbyexposingthecatalystsurfacestoa2.0TorrSO2+O2

gasmixture (SO₂:O₂=1:10, v/v) at 323K (SO₂ purity>99%, Air Products;O₂purity>99.999%,LindeGmbH).Aftertheintroduc- tionofSO_xmixtureat323K,sampleswereannealedto373,473, 573and673Kfor5mininthepresenceoftheSO_xmixture.FTIR spectraofthesesulfatedsurfaceswereacquiredaftercoolingto 323Kinthepresenceofthegasmixtureandsubsequentevacuation to<10⁻³Torr. It shouldbenoted that theeffectiveconcentra- tionofSO₂usedinthecurrentpoisoningexperimentscorresponds toca.263ppm (inabalancecarriergasunderflow conditions), whichtranslatesintoextremelyseverepoisoningconditionscon- sideringthetypicalsulfurcontent(15ppm) ofUltraLowSulfur Diesel(ULSD)fuel.Thus,thecurrentpoisoningexperimentscanbe assessedasacceleratedandextremesulfurpoisoningexperiments, wherethenovelK-AZTbasedcatalystswereexposedtoparticularly challengingconditions,wheretheycandemonstratetheirultimate sulfur-regenerationcapabilities.

2.3.2. MonitoringSO_xdesorptionviain-situFTIR

PriortoSO_x desorptionexperiments,materialsweresulfated asdescribed aboveby collectinga series ofin-situFTIR spectra inthepresenceoftheSO_xmixtureasafunctionoftemperature until673K.Afterthesaturationofthesurfaceswithsulfurat673K, thereactorwasevacuatedtoapressureof<10⁻²Torr,followedby theintroductionof15.0TorrofH2(g)(H2purity>99.999%,Linde GmbH) at323K.Next, poisonedcatalystswere annealedunder hydrogenatmosphereat473,673,773,873and973Kfor5min.

In-situFTIRspectrawereobtainedaftereachH2exposureandby coolingthesampleto323KinthepresenceofH₂.

2.3.3. SO_xdesorptionviaTPD

BeforetheSO_x-TPDexperiments,materialsurfaceswereini- tiallyexposedtoa2.0TorrSO2+O2 gasmixture(SO2:O2=1:10) at673Kfor30min.ThentheIRspectroscopicreactorwasevacu- atedtoapressurelowerthan10⁻³Torrfollowedbyheatingunder vacuumto1173Kwithalinearheatingrateof12K/min.IntheTPD experiments,m/z=32(correspondingtoO₂(g)desorptionandS(g) formationduetotheimpactionization-inducedfragmentationof desorbedSO2(g)speciesinQMS)andm/z=64(correspondingto SO₂(g)desorption)channelsweremonitoredviaQMS.

3. Resultsanddiscussion 3.1. Materialcharacterization

3.1.1. X-raydiffractionanalysis(XRD)

Fig. 1 illustrates the XRD patterns of Pt/AZT, Pt/2.7K/AZT, Pt/5.4K/AZT,Pt/10K/AZTandPt/20Ba/Al.Apartfromthepresence ofstructurallywell-orderedmetallicplatinum(JCPDS001-1190), Pt/AZTandPt/K/AZTcatalystsgiveninFig.1exhibithighlyamor- phouscharacteristics.On theotherhand,materialwith10.0wt.

% K2O (i.e. Pt/10.0K/AZT) reveals additional poorly discernible diffractionsignalsat2=30.48^◦,50.50^◦,60.91^◦correspondingto tetragonal ZrO₂ (JCPDS 80-2155) together with some suppres- sion of Pt diffraction features. XRD analysis of the benchmark Pt/20Ba/AlNSR/LNT catalyst reveals␥-Al2O3 (JCPDS 001-1303), BaAl₂O₄(JCPDS017-0306)andmetallicPt(JCPDS001-1190)fea- tures.OnthePt/20Ba/Alcatalyst,BaOdomainsinteractwiththe

␥-Al2O3supportatelevatedtemperaturesyieldingtheformation ofundesiredBaAl₂O₄phaseasaresultofthermalaging[2,3,33].

10 20 30 40 50 60 70 80

Pt/AZT Pt/20Ba/Al

Pt/2.7K/AZT Pt/5.4K/AZT Pt/10K/AZT

XRD Intensity (arb. u.)

2θ(degree)

t-ZrO2 Pt γ-Al2O3 BaAl2O4

Fig.1.XRDpatternscorrespondingtoPt/AZT,Pt/2.7K/AZT,Pt/5.4K/AZT,Pt/10K/AZT andPt/20Ba/Almaterialsuponcalcinationat973K.

Fig.2.BETspecificsurfacearea(SSA)valuesoftheinvestigatedmaterials.

3.1.2. BETspecificsurfacearea(SSA)measurements

Fig.2illustratesSSAvaluesforPt/AZT,Pt/2.7K/AZT,Pt/5.4K/AZT, Pt/10K/AZTandPt/20Ba/Al.SSAvaluesforPt/AZTisslightlyhigher than2.7wt.%K2O-modifiedcounterpart(i.e.Pt/2.7K/AZT).How- ever,increaseintheK₂Oloadingfrom2.7wt.%to5.4and10.0wt.

%monotonicallydecreasesSSAvaluesform177m²/gto155and 125m²/g;respectively.ItshouldbenotedthattheSSAvalueofthe catalystwiththehighestK₂Oloadingwascomparabletothatofthe benchmarkPt/20Ba/Alcatalyst(134m²/g),whileSSAvaluesofall oftheothercatalystswererelativelyhigher.

3.2. SO_xUptake/adsorptionviain-situFTIRspectroscopy

Fig.3representstemperature-dependentadsorbedSO_xspecies on Pt/AZT, Pt/2.7K/AZT, Pt/5.4K/AZT and Pt/10K/AZT material surfaces upon exposure to 2.0Torr of SO₂+O₂ gas mixture (SO2:O2=1:10).Whiletheblack-coloredspectraineachpanelcor- respondtothesurfaceSO_xspeciesgeneratedwithinatemperature rangeof323–573K,thetopmostredspectracorrespondtosulfur poisoningat673K.

1600 1400 1200 1000 1600 1400 1200 1000

1600 1400 1200 1000

1600 1400 1200 1000

Wavenumber (cm

)

A b so rb ance ( ar b . u) 0. 5

0. 5

(b)

(c)

Pt/2.7K/AZT

Pt/5.4K/AZT

673K

673K

323K

323K

1306 1 178 1019 1048 1 105

1281 1 179 1005 1046 1 106 1027

1306 1391 975

(a) ^Pt/AZT

673K

0. 5

Ab so rb an ce (a rb . u )

323K

Pt/10 K/AZT

Wavenumber (cm

) (d)

0. 5 1238 1 175 1 102 1051

673K

323K

Fig.3. FTIRspectrademonstratingtheSOxuptake/adsorptionpropertiesof(a)Pt/AZT,(b)Pt/2.7K/AZT(c)Pt/5.4K/AZTand(d)Pt/10K/AZTsurfaces.Blacksetofspectra ineachpanelwereacquiredafterSOxexposure(2.0Torr,SO2:O2=1:10)at323K,followedbyannealingat373,473and573KintheSOxgasmixturefor15minand subsequentevacuation.RedspectraineachpanelwererecordedafterSOxexposureat673Kandsubsequentevacuation.Allspectrawererecordedat323Kinvacuum.(For theinterpretationofthereferencestocolourinthisfigure,thereaderisreferredtothewebversionofthisarticle.)

Kimetal.reportedthatanincreaseinK2Oloadingfrom2wt.

%to 30wt. % ledto a boost in NSC of Pt/K₂O/Al₂O₃ materials within600–800K[34].Aswillbedemonstratedlatter,although suchextremelyhighK2Oloadingscouldbebeneficialtoenhance NSCintheabsenceofSO_x,theymayalsoleadtoirreversiblesulfur poisoninginthepresenceofSO_x.Therefore,inthecurrentstudy, welimitedtheK₂OloadingoftheAZT-basedNSR/LNTmaterialsto 10wt.%.

Fig.3ashowsthatSO_xadsorptiononPt/AZTatrelativelylower temperatures(i.e.323and373K)leadstotwomain vibrational featureslocatedat1027and975cm⁻¹whichcanbeassignedto sulfate(SO₄²⁻)andsulfite(SO₃²⁻)functionalgroups,respectively [35–39].Absorbanceintensitiesofthesetwoparticularvibrational frequenciesarecomparableatlowtemperatures,whilethesulfate featurestartstodominatethesulfitefeatureat highertemper- atures.Thermally-triggeredcatalyticoxidation of sulfitespecies tosulfatesonthePt/AZTsurfacecanalsobefollowedinFig.3a

bymonitoringthegrowthoftheantisymmetricstretchingmode ofsurfacesulfategroupslocatedat1391and1306cm⁻¹ [14,40].

Asillustrated inFig. 3b–d,additionof basic K₂Odomains onto AZTternaryoxidesystemresultsinalterationofthespectralline shapes.Pt/2.7K/AZT(Fig.3b)presentsfivemajorvibrationalfea- tureslocatedat1306,1178,1105,1048and1019cm⁻¹.WhileIR stretchingsat1306,1105,1048and1019cm⁻¹canbeattributedto thesurfacesulfate(SO42−)groupsonK2Oand/orontheAZTsup- port,vibrationalfeaturelocatedat1178cm⁻¹canbeattributedto bulk-likesulfategroupsonK₂O[41,42].Thislatterfeaturebecomes morediscerniblewithanincreaseintheK2Oloading(i.e.5.4wt.% and10wt.%)evidentbytheincreasingrelativeabsorbanceinten- sityof the1178cm⁻¹ signalinFig.3candd.SO_xadsorptionon Pt/5.4K/AZTandPt/10K/AZTmaterialsat673Kleadstovibrational featuresat1281and1238cm⁻¹alongwiththeabsenceofanysig- nificantvibrationalbandslocatedatca.>1300cm⁻¹ (Fig.3cand d).Thevibrationalsignalsat1281and1238cm⁻¹canbeassigned

1600 1400 1200 1000

1600 1400 1200 1000

1600 1400 1200 1000

Wavenu mber (cm

)

A b so rb ance ( ar b . u)

673K 473K 323K

773K

1306 1 178 1019 1048

673K 473K 323K

773K

1275 1 178 10 11 1049

(a)

(c)

Pt/2.7K/AZT

Pt/5.4K/AZT

0. 5

0. 5 1240

1240 1027

1306

1391

673K 473K 323K

773K

(a) Pt/AZT

0. 5 Ab so rb an ce (a rb . u )

1600 1400

120 0

Pt/10 K/AZT

673K 473K 323K

773K 873K

Wavenumber (cm

)

1238 1 175 1 102 1051

Fig.4. FTIRspectraassociatedwithSOxreductionandregenerationof(a)Pt/AZT,(b)Pt/2.7K/AZT(c)Pt/5.4K/AZTand(d)Pt/10K/AZTmaterialsviaH2(g).Catalystswere initiallysulfated(2.0Torr,SO2:O2=1:10for15minat673K)followedbyevacuationandsubsequentexposuretoH2(g)(15.0Torr)at323,473,673,773,873and973Kfor 5min.Allspectrawererecordedat323K.

topredominantlysulfatesonK₂Odomains[14,41–43].Itcanbe arguedthatwiththeincreasingK₂Oloading,anincreasinglylarger portionof the AZT surface is covered by K2O islands/domains, decreasingtheextentofexposed/uncoveredAZTsurface.Itisalso likelythattheincreaseintheK₂Oloadingalsoresultsinthegrowth oftheK2Oparticlesizeandformationof3Dagglomerates,enabling thestorageofSO_xintheformofbulk-likesulfatesinthesub-surface ofthese3Dnanoparticles.Thisargumentisalsoingoodagreement withthemeasuredSSAvaluespresentedinFig.2suggestingthat theincreaseintheK2Oloadinginthecatalystformulationleads toamonotonicdecreaseintheSSAasexpectedbysinteringofthe K₂Odomainsandparticlesizegrowth.

3.3. SulfurregenerationwithH2(g)viain-situFTIRspectroscopy

Asmentionedabove,SO_xreduction/regenerationperformance hasasignificantinfluenceonthecatalystlifetimeandNO_xstorage

capacity.Therefore, SO_x reductioncharacteristicsofsynthesized materialswerealsostudiedasafunctionoftemperaturebymeans of in-situ FTIR spectroscopy. Fig. 4 illustrates the evolution of theS-containingsurfacefunctionalgroupsonPt/AZT,Pt/2.7K/AZT, Pt/5.4K/AZTandPt/10K/AZTcatalystsurfacesasafunctionoftem- peraturewithin323–773Kinthepresenceofanexternalreducing agent,H2(g).

Inthesesetofexperiments,catalystswereinitiallysaturated witha2.0TorrofSO₂+O₂gasmixture(SO₂:O₂=1:10)at673Kfor 5minandthencooledto323Kfollowedbytheevacuationofthe spectroscopicreactor,introductionof15.0TorrH₂(g)at323Kand annealinginH₂(g)atthegiventemperatureswithin323–773K.

This particularly chosen initialsulfation/poisoning temperature (i.e.673K)isnot onlyrelevant torealisticNSR/LNT operational temperatures,but isalsohighenoughtoactivateSO₂ oxidation tosulfitesandsulfatesinacomprehensivemanner.Intheseries ofexperimentsgivenin Fig.4,spectral lineshapesdo nottypi-

300 400 500 600 700 800 900 1000

O₂ SO₂

300 400 500 600 700 800 9001000 O2 SO2

300 400 500 600 700 800 9001000

O2 SO2

Pt/2.7K/AZT

Pt/5.4K/AZT

Q M S In te n si ty (a rb . u. )

(a) Pt/AZT

1. 5E -6

(b)

(c)

1. 5E -6

Pt/10K/AZT

1. 5E -6 1. 5E -6

(d)

Q M S In te n si ty (a rb. u .)

Temperature (K) Temperature (K)

820 920 910

975 820 810

880 920 800

300 400 500 600 700 800 9001000 O2 SO2

Fig.5. TPDprofilesfor(a)Pt/AZT,(b)Pt/2.7K/AZT,(c)Pt/5.4K/AZTand(d)Pt/10K/AZTcatalystsafter2.0TorrSOx(2.0TorrSO2+O2,SO2:O2=1:10)adsorptionat673Kfor 30minandsubsequentevacuation.

callychangeinanoteworthymanneratreductiontemperatures

≤473K.However,increasingthereductiontemperatureto673K (grayspectrainFig.4a–c)leadstonoticeablealterationsintheFTIR spectra,wherebulkandsurfacesulfate/sulfitespeciessignificantly attenuateforPt/AZT,Pt/2.7K/AZTandPt/5.4K/AZT.Increasingthe temperatureto773KinthepresenceofH2 leadstothealmost complete elimination of the SO_x-related vibrational signatures onPt/AZT, Pt/2.7K/AZTandPt/5.4K/AZTsurfaces(redspectrain Fig.4a–c). It canbe seenin Supporting information Fig.1 that whenPt/20Ba/AlbenchmarkNSR/LNT catalystis exposed toan identicalsetofsulfationandsubsequentreductiontreatments,a significantlygreaterportionofsulfate/sulfitespeciescontinueto existonthePt/20Ba/Alcatalystevenat773Kinthepresenceof 15.0Torr H₂(g).This comparativeanalysis clearly suggests that Pt/AZT,Pt/2.7K/AZTandPt/5.4K/AZTcatalystsexhibit asuperior sulfurregenerationperformancethanthatofthePt/20Ba/Alcata-

lyst,asthisformersetofmaterialscanbefullyde-sulfatedat773K inthepresenceofH2.

However, AZT-based catalystswith thehighest K₂O loading usedinthecurrentstudy(i.e.Pt/10K/AZT)notonlyshowedunique sulfur uptake characteristics as presented in Fig. 3d which is dominated by bulk-like sulfates, but also revealed a fairly dif- ferent SO_x-reductionprofilein thepresence of H₂ (Fig. 4d).As can be seen in Fig. 4, at relatively low temperatures (i.e.T ≤ 673K), a significantportion of the SO_x species can already be eliminated from Pt/AZT, Pt/2.7K/AZT and Pt/5.4K/AZT surfaces.

However,atT≤673K,almostalloftheS-relatedsurfacefunctional groupsremainintactonPt/10K/AZT.Evenatareductiontemper- atureof773K,althoughPt/AZT,Pt/2.7K/AZTsurfacescanbefully regenerated(Fig.4a–c),Pt/10K/AZTsurfacestillremainspartially blocked/poisonedbysulfur-containingfunctionalgroupsandcom- pletelyeliminatedatonlyat≥873K(Fig.4d).

3.4. SulfurregenerationundervacuumviaTPDanalysis

TPDexperimentswerealsocarriedoutinvacuuminorderto investigatethethermalregenerationabilityofthesynthesizedcat- alystsaftersulfurpoisoningintheabsenceofareducingagent,as wellastocomparetherelativeadsorptionstrengthsofSO_xspecies residingonthepoisoned catalystsurfaces.Prior toTPDexperi- ments,eachcatalystwasexposedto2.0TorrSO2+O2gasmixture (SO₂:O₂=1:10)at673Kfor30min.

Fig. 5 shows the TPD spectracorresponding to the thermal decomposition of sulfates and sulfites on Pt/AZT, Pt/2.7K/AZT, Pt/5.4K/AZTandPt/10K/AZTcatalystsurfaces.IntheseTPDexper- iments,onlyO₂ and SO₂ desorptionchannels(correspondingto masstochargeratiosofm/z=32and64;respectively)revealedsig- nificantsignalsandotherSO_xorH2Sspecieswerenotdetectable.As inthecaseoftheBaO-basedconventionalNSR/LNTcatalyst(Sup- portinginformationFig.2),SO_x-relatedspeciesadsorbedonPt/AZT anditsK2O-incorporatedcounterpartsrevealhighthermalstabil- itywhichisevidentbytheappearanceofSO_xdesorptionsignalsat T>700K.AnalysisofthegeneralTPDlineshapesgiveninFig.5a–c suggeststhatatleasttwodifferentSO_xdesorptionsignalsexistfor Pt/AZT,Pt/2.7K/AZTandPt/5.4K/AZTcatalystsatT<1050K,reveal- ingdesorptionmaximalocatedatca.800–820Kandat900–975K.

Furthermore,SO2desorptiononthesesurfaceswithin700–1050K isaccompaniedbyO2 desorption,suggestingthat sulfate/sulfite decompositionoccursintheformofsimultaneousSO₂+O₂release.

It should benoted that thecontribution of theSO₂ gastothe m/z=32 signalduetoelectron-impactinducedfragmentationof SO₂intheQMSionizerchamberislessthan10%,suggestingthat m/z=32signalcanbealmostexclusivelyattributedtotheevolution ofO2(g)fromthecatalystsurfaces.

ItisvisibleinFig.5a–cthattheTPDdesorptionmaximatendto shifttowardshighertemperatureswithincreasingK₂Oloadingin thecatalystformulation.Itcanalsobenoticedthatwithincreas- ingK₂Oloadingto5.4wt.%(seeFig.5c),relativeintensityofthe 820KdesorptionfeaturewhichcanbemostlyassociatedwithSO_x speciesonAZTsurfaceissuppressed,alongwiththegenerationof ahigh-temperaturedesorptionshoulderat975K.Thisisinperfect agreementwiththein-situFTIRresultspresentedinFig.3suggest- ingthatwithincreasingK₂Osurfacecoverage,extentofexposed (uncovered)AZTsurfacedecreasesalongwithanincreaseinthe K₂Oparticlesize,facilitatingtheformationofbulk-likesulfatesthat arealsothermallymorestablethanthatofthesulfatesonAZT.

ItisalsoimportanttonotethattheSO_xdesorptionisnotcom- pleteinFig.5b–devenat1050K(i.e.thehighestexperimentally attainabletemperature inthecurrent TPDsetup)as evidentby thepresenceof adesorptiontailatT=1050Kwhichispresum- ablyextendingwell-beyondthistemperature(assupportedbythe in-situFTIRresultsthatwillbeprovided later inthetext).This observationimpliesthatwhilesurfacesulfates/sulfitespresenton AZTsupportandK2Odomainsfullydecomposeattemperatures below920K,bulk-like potassium sulfate species requirehigher desorptiontemperaturesforcompletethermaldecompositionand desorption.Inotherwords,presenceofbasicK2Odomainsyields strongbindingsitesforSO₂,leadingtotheformationofthermally stablesurfaceandbulk-likeSO_xspecies.

On theotherhand,a furtherincrease in theK2Oloadingto 10wt.%illustratesratherdifferentSO_xdesorptioncharacteristics (Fig.5d).ItisevidentthattheSO_xdesorptionfeaturesatT<1050K are suppressedto a great extent, leading toa relatively minor desorptionfeaturelocatedat910Kwithashoulderatca.810K.

ConsideringthesignificantSO_xuptakeofthePt/10K/AZTcatalyst surfacedemonstratedbythein-situFTIRdatagiveninFig.3d,itis clearthatmostofthesulfate/sulfitespeciesonPt/10K/AZTremain intactevenaftervacuumannealingupto1050K.Thisisalsoquan- titativelypresentedinFig.6,whichpresentstheintegratedSO2TPD

Fig.6.AnalysisofrelativeSO_xreleasefrominvestigatedcatalystscalculatedvia integratedTPDsignalsgiveninFig.5.

desorptionsignalsfortheinvestigatedAZT-basedcatalystswithin 323–1050K.Fig.6illustratesthattheintegratedSO_xdesorption signalofPt/AZTisroughlytwicegreaterthanthatofPt/2.7K/AZT and Pt/5.4K/AZTand alsoaboutfourtimes greaterthan thatof Pt/10K/AZT.

Figs.3,5and6,suggestthatafterthesulfationoftheAZT-based catalystsat673Kandasubsequentvacuumannealingupto1050K duringtheTPDexperiments,asignificantfractionofthesulfateand sulfitespeciesremainintactontheK-containingsamplesurfaces.

Thus,itiscrucialtoinvestigatetheresidualSO_xspeciesremaining ontheK-containingAZTsystemsaftertheTPDruns.Fig.7shows suchin-situFTIRexperimentscorrespondingtoalloftheinvesti- gatedsulfurpoisonedAZT-basedcatalystsbefore(blackspectra) andafter(redspectra)TPDexperiments.Fig.7aclearlyindicates thatintheabsenceofK₂O,sulfur-poisonedPt/AZTcatalystcanbe almostfullyregeneratedviavacuumannealingupto1050Kdur- ingtheTPDexperiments.Ontheotherhand,Pt/2.7K/AZTcatalyst which releasesabout%50lesser amountof SO_x speciesduring TPD(Fig.6),stillrevealsaminor,yetreadilydetectablequantityof SO_x(Fig.7b).Ontheotherhand,Pt/5.4K/AZTcatalystwhichhasa comparableintegratedSO_xdesorptionsignaltothatofPt/2.7K/AZT (Fig.6),revealsastrongerresidualSO_xsignalintheFTIRspectrum obtainedaftertheTPDrungiveninFig.7c.Thisobservationisin linewiththefactthat Pt/5.4K/AZTsurfacestores asignificantly greateramountofSO_xspecies(whicharealsothermallymoresta- ble)as compared tothatof Pt/2.7K/AZT.Finally, residualsulfur analysisofthePt/10K/AZTsurface(Fig.7d)indicatesthatalmost alloftheSO_xspeciesgeneratedduringinitialpoisoningprocess remainintactaftertheTPDrunandvacuumannealingat1050K.

Thus,itisapparentthattheminoramountofSO_xreleaseduringthe TPDexperimentforPt/10K/AZT(Fig.6)correspondstoatinyfrac- tionoftheoverallsulfurthatisstoredonthissurface.Thislatter resulthassomeresemblancetotheTPDdatacorrespondingtothat ofthePt/20Ba/AlbenchmarkcatalystgiveninSupportinginforma- tionFigs.2and3andalsoinFig.6whichalsorevealanincomplete thermalregenerationuponvacuumannealingupto1050Kduring theTPDrun.

WealsoperformedacomprehensiveinvestigationoftheNO_x storage,releaseandreductioncharacteristicsofPt/K/AZTsystems viain-situFTIR,TPDaswellasquantitativeflow-reactorexperi- ments[44].Adetailedaccountoftheseadditionalexperimentswill

1450 1250 1050

1450 1250 1050

1450 1250 1050

Wavenu mber (cm ^-1 )

A b sorb an ce (a rb . u) 0. 5

0. 5

(b)

(c)

Pt/2.7K/AZT

Pt/5.4K/AZT

1308 1 178 1019 1052

1238

1277 1 175 1049 1012

1225 111 5

A b sorb an ce (a rb . u)

(a)

1306

1391

Pt/AZT

1027

1450 1250 1050

Pt/10 K/AZT

Wavenumber (cm ^-1 )

0. 5

0. 5 1238 1 181 1 147 11 15 1051

(d)

Fig.7.FTIRspectracorrespondingtoSOxcontentof(a)Pt/AZT,(b)Pt/2.7K/AZT,(c)Pt/5.4K/AZTand(d)Pt/10K/AZTcatalystsbefore(black)andafter(red)SOx-TPDruns.(For interpretationofthereferencestocolourinthisfigurelegend,thereaderisreferredtothewebversionofthisarticle.)

bediscussedthoroughlyinaforthcomingreport.Nevertheless,itis instructivetopresentrelativeintegratedTPDNO_xdesorptionsig- nalsobtainedaftersaturationofthefreshlypreparedAZT-based catalystswithNO₂(5.0TorrNO₂at323Kfor10min)intheabsence ofsulfurascomparedtothatofthePt/20Ba/Albenchmarkcata- lyst(Fig.8).AscanbeseeninFig.8,relativeNO_xstorageamounts oftheAZT-basedcatalystsincreasemonotonicallywithincreasing K₂Oloadinguntil5.4wt.%K₂O,afterwhichitconvergestoavalue thatiscomparabletothatofthePt/20Ba/Albenchmarkcatalyst.

ItisworthmentioningthatNSCofthesynthesizedmaterialswere investigatedinaflow-modetubularreactorwheretheinletgasfeed wascomposedof500ppmNO,v.%5O2,v.%5CO2andv.%5H2O balancedwithAr(g),revealingsimilarNSCvaluesforPt/5.4K/AZT (0.165mmol/gcat)andconventionalPt/20Ba/Al(0.171mmol/gcat) at573K[44].

Acombinedanalysisofthestructuralcharacterizationresults aswellas thespectroscopicprobe moleculeadsorptionexperi- mentsgiveninthecurrentstudyallowsustoshedlightonsulfur poisoning,regeneration and NO_xstorage characteristicsofAZT- basedNSR/LNTcatalystsfunctionalizedwithK2O.Intheabsence oftheK2O,Pt/AZTsystemrevealshighSSA(191m²/g)andrela- tivelyweaklyboundsulfates/sulfiteswhichcanreadilyberemoved fromthesurfaceina completefashioneitherbyreductionwith H2(g)at 773Kor simplybythermalregeneration invacuumat ca.950K.However,duetolackofbasicK₂Odomains,Pt/AZTsuf- fersfromrelativelylow NSC.IncreasingtheK₂Oloadingto2.7, 5.4and10.0wt.%leadstoaprogressivelyincreasingNO_xadsorp- tionwhereNSCseemstobeconvergingtoavaluesimilartothat ofPt/20Ba/AlbenchmarkcatalystforPt/5.4K/AZTandPt/10K/AZT.

Furthermore,Pt/5.4K/AZTsampleallowscompleteremovalofSO_x viaH₂(g)at773KunliketheconventionalPt/20Ba/Albenchmark

Fig.8.RelativeintegratedNO_xdesorptionsignalsobtainedfromNO_x-TPDexperi- ments.

catalystwhosecompleteregenerationrequiresmuchhighertem- peratures(i.e.973K)underidenticalreducingconditions.Although increasingtheK₂Oloadingfrom5.4to10.0wt.%doesnotseem tohaveatremendousenhancementinNO_xadsorptionproperties ofthePt/K/AZTsystem,itdoesresultinunfavorableSO_xuptake, releaseandregenerationcharacteristics.TPDandFTIRdatasug- gestthatK₂OdomainstendtoagglomeratewithincreasingK₂O loadingand form 3D clusterswith growingK2O particle sizes.

Thesephenomena alsoexpeditetheformation of bulk-likesul- fatefunctionalitiesinthesubsurfaceofK₂Odomainswithmuch higherthermalstabilityandmuchstrongerresistanceagainstther- maldecompositionandreductionwithhydrogen.Consequently, Pt/5.4K/AZTsystemappearsasapromisingalternativewhichcan alsobeusedinconjunctionwithconventionalPt/20Ba/AlNSR/LNT catalysts.

4. Conclusion

Inthecurrentstudy,advancedternaryandquaternarymixed oxidematerialsintheformofPt/K₂O/Al₂O₃/ZrO₂/TiO₂weresyn- thesizedwithdifferentK₂Oloadings.Synthesizedmaterialswere structurallycharacterizedviaXRDandBETincomparisontoacon- ventionalPt/20Ba/AlbenchmarkNSR/LNTcatalyst.Interactionof thesecatalystsurfaceswithSO_x(i.e.SO₂+O₂)mixtureweremoni- toredspectroscopicallyusingin-situFTIRandTPD.Ourfindingscan besummarizedasfollows:

• BesidesthepresenceoforderedmetallicPt,Pt/AZT,Pt/2.7K/AZT and Pt/5.4K/AZT materialsrevealed disordered structures. On theotherhand,Pt/10K/AZTexhibitedadditionaldiffractionsig- nals corresponding to tetragonal ZrO2 domains. Unlike the AZT-supported materials,conventional Pt/20Ba/Albenchmark catalystwascomposedoforderedphasesincluding␥-Al2O₃and BaAl₂O₄.

• Increasein K2Oloadingfrom2.7 to5.4and10.0wt.%mono- tonicallydecreasestheSSA values from177m²/gto 155and 125m²/g,respectively.ApartfromthePt/10K/AZTcatalyst,SSA valuesofthecorrespondingPt/K/AZTcatalystsarehigherthan thatofthebenchmarkPt/20Ba/Alcatalyst(134m²/g).

• IncreasingtheK₂OloadinginthePt/K/AZTsystemleadstothe growthoftheK2Odomainsize(i.e.sintering),covering ofthe AZTsurface withK₂Oand anincreasein thebulk-likesulfate functionalgroups requiringhigher temperaturesfor complete sulfurelimination viathermal decompositionorviareduction withH₂(g).

• Increasein K₂Oloadingin thePt/K/AZTformulationincreases theNO_xadsorptionupto5.4wt.%ofK2O.HoweverK2Oloadings higherthanthisvaluedonothaveasignificantpositiveinfluence onNO_xadsorption.

• Thereis a delicate trade-offbetweenNSC and sulfur adsorp- tion/release/regenerationcharacteristics.NSCandSO_xtolerance ofAZTbasedNSR/LNTcatalystscanbeoptimizedsimultaneously bycarefullyfine-tuningtheK₂Oloading.

• Among the investigated catalysts, Pt/5.4K/AZT was found to revealsuperiorsulfurregenerationperformancethanthatofthe conventionalPt/20Ba/Albenchmarkcatalystalongwithacom- parableNSC(0.165vs.0.171mmol/gcat,repsectively).

Acknowledgements

AuthorsacknowledgethefinancialsupportfromtheScientific andTechnologicalResearchCouncilofTurkey(TUBITAK)(Project Code:111M780).AuthorsalsoacknowledgeProf.LouiseOlssonand OanaMihai(ChalmersUniversityofTechnology)forquantitative flow-modeNSCmeasurements.

AppendixA. Supplementarydata

Supplementarydataassociatedwiththisarticlecanbefound, intheonlineversion,athttp://dx.doi.org/10.1016/j.cattod.2015.12.

013.

References

[1]F.Klingstedt,K.Arve,K.Eränen,D.Y.Murzin,Acc.Chem.Res.39(2006) 273–282.

[2]W.S.Epling,L.E.Campbell,A.Yezerets,N.W.Currier,J.E.Parks,Catal.Rev.Eng.

46(2004)163–245.

[3]S.Matsumoto,CATTECH4(2000)102–109.

[4]A.Fritz,V.Pitchon,Appl.Catal.BEnviron.13(1997)1–25.

[5]Q.Schiermeier,ThesciencebehindtheVolkswagenemissionsscandal,in:

Nat.News,2015.

[6]<http://www.eea.europa.eu/publications/nec-directive-status-report-2014>.

[7]K.Kato,H.Nohira,K.Nakanishi,S.Iguchi,T.Kihara,H.Muraki,EuroPatent application0,573,672A1,1993.

[8]N.Myioshi,S.Matsumoto,K.Katoh,T.Tanaka,K.Harada,N.Takahashi,K.

Yokota,M.Sugiura,K.Kasahara,SAETechnicalPapersSeriesNo.950809, 1995.

[9]S.Roy,A.Baiker,Chem.Rev.109(2009)4054–4091.

[10]Z.Say,E.I.Vovk,V.I.Bukhtiyarov,E.Ozensoy,Appl.Catal.BEnviron.142–143 (2013)89–100.

[11]W.S.Epling,D.Kisinger,C.Everest,Catal.Today136(2008)156–163.

[12]D.H.Kim,Y.Chin,G.G.Muntean,A.Yezeretz,N.W.Currier,W.S.Epling,H.

Chen,H.Hess,C.H.F.Peden,Ind.Eng.Chem.Res.45(2006)8815–8821.

[13]S.M.Andonova,G.S.S¸ent ¨urk,E.Kayhan,E.Ozensoy,J.Phys.Chem.C113 (2009)11026.

[14]Z.Say,E.I.Vovk,V.I.Bukhtiyarov,E.Ozensoy,Top.Catal.56(2013)950–957.

[15]T.Szailer,J.H.Kwak,D.H.Kim,J.C.Hanson,C.H.F.Peden,J.Szanyi,J.Catal.239 (2006)51–64.

[16]I.S.Pieta,M.García-Diéguez,C.Herrera,M.a.Larrubia,L.J.Alemany,J.Catal.

270(2010)256–267.

[17]K.S.Kabin,P.Khanna,R.L.Muncrief,V.Medhekar,M.P.Harold,Catal.Today 114(2006)72–85.

[18]S.Morandi,F.Prinetto,G.Ghiotti,L.Castoldi,L.Lietti,P.Forzatti,M.Daturi,V.

Blasin-Aubé,Catal.Today231(2014)116–124.

[19]D.H.Kim,K.Mudiyanselage,J.Szányi,H.Zhu,J.H.Kwak,C.H.F.Peden,Catal.

Today184(2012)2–7.

[20]M.Takeuchi,S.Matsumoto,Top.Catal.28(2004)151–156.

[21]T.J.Toops,D.B.Smith,W.P.Partridge,Catal.Today114(2006)112–124.

[22]J.Luo,F.Gao,D.H.Kim,C.H.F.Peden,Catal.Today231(2014)164–172.

[23]E.Fridell,H.Persson,L.Olsson,B.Westerberg,A.Amberntsson,M.Skoglundh, Top.Catal.16/17(2001)133.

[24]J.Escobar,J.A.DelosReyes,T.Viveros,Ind.Eng.Chem.Res.39(2000)666.

[25]Z.Say,M.Tohumeken,E.Ozensoy,Catal.Today231(2014)135.

[26]M.Meng,L.Guo,J.He,Y.Lai,Z.Li,X.Li,Catal.Today175(2011)72.

[27]Z.Q.Zou,M.Meng,X.Y.Zhou,X.G.Li,Y.Q.Zha,Catal.Lett.128(2009)475.

[28]N.Takahashi,A.Suda,I.Hachisuka,M.Sugiura,H.Sobukawa,H.Shinjoh,Appl.

Catal.BEnviron.72(2007)187.

[29]H.Imagawa,T.Tanaka,N.Takahashi,S.Matsunaga,A.Suda,H.Shinjoh,J.

Catal.251(2007)315.

[30]H.Imagawa,N.Takahashi,T.Tanaka,S.Matsunaga,H.Shinjoh,Appl.Catal.B Environ.92(2009)23.

[31]Z.Q.Zou,M.Meng,J.He,Mater.Chem.Phys.124(2010)987.

[32]Z.Say,M.Dogac,E.I.Vovk,Y.E.Kalay,C.H.Kim,W.Li,E.Ozensoy,Appl.Catal.B Environ154–155(2014)51.

[33]L.E.Venegas,N.A.Mazzeo,A.L.P.Rojas,AirQual.Appl.,InTech.,2011,ISBN:

978-953-307-307-1.

[34]D.H.Kim,K.Mudiyanselage,J.Szanyi,J.H.Kwak,H.Zhu,C.H.F.Peden,Appl.

Catal.BEnviron.142–143(2013)472–478.

[35]M.Kantcheva,E.Z.Ciftlikli,J.Phys.Chem.B106(2002)3941–3949.

[36]C.Chang,J.Catal.53(1978)374–385.

[37]M.Waqif,SolidStateIonics95(1997)163–167.

[38]O.Saur,M.Bensitel,A.B.MohammedSaad,J.C.Lavalley,C.P.Tripp,B.A.

Morrow,J.Catal.99(1986)104–110.

[39]H.Yao,H.K.Stepien,H.S.Gandhi,J.Catal.67(1981)231–236.

[40]G.S.S¸entürk,E.I.Vovk,V.I.Zaikovskii,Z.Say,A.M.Soylu,V.I.Bukhtiyarov,E.

Ozensoy,Catal.Today184(2012)54–71.

[41]C.Sedlmair,K.Seshan,A.Jentys,J.Lercher,Catal.Today75(2002)413–419.

[42]Y.Liu,M.Meng,X.G.Li,L.H.Guo,Y.Q.Zha,Chem.Eng.Res.Des.86(2008) 932–940.

[43]H.Abdulhamid,E.Fridell,J.Dawody,M.Skoglundh,J.Catal.241(2006) 200–210.

[44]Z.Say,O.Mihai,M.Tohumeken,L.Olsson,E.Ozensoy,inpreperation.