Thermal evolution of structure and photocatalytic activity in polymer microsphere templated TiO2 microbowls

(1)

ContentslistsavailableatScienceDirect

Applied

Surface

Science

j o u r n a l ho me p ag e : w w w . e l s e v i e r . c o m / l o c a t e / a p s u s c

Thermal

evolution

of

structure

and

photocatalytic

activity

in

polymer

microsphere

templated

TiO

2

microbowls

夽

Deniz

Altunoz

Erdogan

a

_,

_Meryem

_Polat

a

_,

_Ruslan

_Garifullin

b

_,

Mustafa

O. Guler

b

_,

_Emrah

_Ozensoy

a,∗,1

a_Department_of_Chemistry,_Bilkent_University,₀₆₈₀₀_Ankara,_Turkey

b_Institute_of_Materials_Science_and_{Nanotechnology,}_National_{Nanotechnology}_Research_Center_(UNAM),_Bilkent_University,₀₆₈₀₀_Ankara,_Turkey

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received22January2014

Receivedinrevisedform4April2014 Accepted12April2014

Availableonline21April2014

Keywords: TiO2

Photocatalyst

Cross-linkeddivinylbenzene NO(g)oxidation

RhodamineB

a

b

s

t

r

a

c

t

Polystyrenecross-linkeddivinylbenzene(PS-co-DVB)microsphereswereusedasanorganictemplate inordertosynthesizephotocatalyticTiO2microspheresandmicrobowls.Photocatalyticactivityofthe microbowlsurfacesweredemonstratedbothinthegasphaseviaphotocatalyticNO(g)oxidationbyO2(g) aswellasintheliquidphaseviaRhodamineBdegradation.Thermaldegradationmechanismofthe poly-mertemplateanditsdirectinfluenceontheTiO2crystalstructure,surfacemorphology,composition, specificsurfaceareaandthegas/liquidphasephotocatalyticactivitydatawerediscussedindetail.With increasingcalcinationtemperatures,sphericalpolymertemplatefirstundergoesaglasstransition, cover-ingtheTiO2film,followedbythecompletedecompositionoftheorganictemplatetoyieldTiO2exposed microbowlstructures.TiO2microbowlsystemscalcinedat600◦Cyieldedthehighestper-sitebasis pho-tocatalyticactivity.CrystallographicandelectronicpropertiesoftheTiO2microspheresurfacesaswell astheirsurfaceareaplayacrucialroleintheirultimatephotocatalyticactivity.Itwasdemonstratedthat thepolymermicrospheretemplatedTiO2photocatalystspresentedinthecurrentworkofferapromising andaversatilesyntheticplatformforphotocatalyticDeNOxapplicationsforairpurificationtechnologies. ©2014ElsevierB.V.Allrightsreserved.

1. Introduction

Shape-deﬁnednanoand microscale titaniumdioxide(TiO2)

structuresare widelyutilized asphotocatalytic systems;where

theyhaveattractedaparticularinterestinenvironmental

applica-tions.Ithasbeenreportedthatcontrollingparticleshape,geometry,

size,surfacemorphology,electronicstructure,relativeabundance

ofanatase/rutile surfacedomains and thenatureofthesurface

functionalgroups(suchas OH)aresomeofthekeyfactorsfor

designingefﬁcientTiO2photocatalyticarchitectures[1–5].

TiO2 materials can be produced with unique

morpholo-gies, shapes and structures at the micro/nanoscale revealing

extraordinaryphysical,chemical,electronicandopticalproperties,

renderingthesesystemsveryversatilephotocatalysts[3].Template

directedsynthesis isoneoftheapproachesfor ﬁne-tuningsize,

夽 ElectronicSupplementaryInformation(ESI)available:Gas-phaseand solution-phasephotocatalyticperformenceofP25.

∗ Correspondingauthor.Tel.:+903122902121;fax:+903122664068. E-mailaddress:ozensoy@fen.bilkent.edu.tr(E.Ozensoy).

1 _Web:_{http://www.fen.bilkent.edu.tr/∼ozensoy}_.

shape andporosityofTiO2 particles[6–8].In particular,

utiliza-tionoforganictemplatessuchaspolymersoffersvastopportunities

forcontrollingtheshapesofinorganicmaterialsatthe

microme-ter/nanometerscale.Suchstrategiescanbeexploitedtosynthesize

shape-deﬁnedTiO2 materialsexhibitingnano/microspheres[9],

hollowstructures[10],tubes[11],wires[3],core–shellstructures

[3],andegg-yolkstructures[12].

In thecurrent report,TiO2 microbowls weresynthesizedby

usingpolystyrenecrosslinkeddivinylbenzene(PS-co-DVB)

micro-spheres.Thepolymertemplatewasremovedbycalcinationand

TiO2 microbowls were produced. The effect of the calcination

temperatureonthestructuralpropertiesandactivityofthe

pho-tocatalystswerestudiedinthegasphaseaswellasinthesolution

phaseoxidationreactions.

2. Experimental

2.1. Samplepreparation

Acustomsol–gelmethodcombinedwithapolymertemplating

techniquewasusedforthesynthesisofTiO2microbowlstructures

http://dx.doi.org/10.1016/j.apsusc.2014.04.082

(2)

Scheme1. SyntheticprotocolforPS-co-DVB-templatedTiO2microspheresandmicrobowls.

[13–15].Commerciallyavailablepolystyrenecross-linkeddivinyl

benzene(PS-co-DVB)microspheres(Aldrich)withanaverage

parti-clesizeofca.8␮mwereusedasthetemplatematerial.Preparation

ofTiO2microspheresandmicrobowlsisshowninScheme1.First,

equalmasses(i.e.1.0g)ofpolymermicrospheresandtitanium(IV)

isopropoxide(TIP,97%,Aldrich)weremixedandstirredfor24h

underambientconditions.Then,100mLofdeionizedwater

(Milli-Q,18.2Mcm)wasaddedtothemixtureundercontinuousstirring

(24h),wherehydrolysisandcondensationreactionswerecarried

out.Then,microsphereswerevacuum-ﬁltered,washedwith

deion-ized water anddried for 24h at60◦C in air. Later,thesample

wascalcinedinairinordertoremove thepolymertemplateas

wellastocrystallizetheinorganiccomponent(i.e.TiO2).Samples

werecalcinedatvarioustemperatures(200,300,400, 500,600,

700◦C) inair for 2h(using a heatingrate of8◦C/min) to

con-trolcrystallinity and surface morphologyof TiO2 microspheres.

SynthesizedsampleswerenamedasPsTi-200,PsTi-300,PsTi-400,

PsTi-500, PsTi-600, and PsTi-700 depending on the calcination

temperature.

2.2. Structuralcharacterization

Themorphologyandtheparticlesizeofthepolymertemplated

TiO2 microspheresand microbowls wereinvestigated byusing

a Carl-ZeissEvo40environmentalscanningelectron microscope

(SEM) equipped with a Bruker energy dispersive X-Ray (EDX)

detector.Determinationofthecrystalstructureofthesynthesized

materialswerecarriedoutwithaRigakuMiniﬂexX-ray

diffrac-tometer(XRD)equippedwithCuK␣radiationoperatedat30kV,

1.54 ˚Aand15mA.TheXRDpatternswererecordedinthe2range

of10–60◦withastepwidthof0.02s−1.Ramanspectraofthe

sam-pleswerecollectedintherangeof200–1500cm−1witharesolution

of4cm−1usingaHoribaJobinYvonLabRAMHR800spectrometer

equippedwitha confocalRaman BX41microscope.The Raman

spectrometerwasequippedwitha Nd:YAGlaser (=532.1nm)

where the laser power was 20mW. The thermal properties

of the TiO2 systems were also investigated by using thermo

gravimetricanalysis(TGA).TGAmeasurementswerecarriedout

between30and800◦C(ataheatingrateof10◦C/minandunder

nitrogenﬂow)by usinga TAInstrumentsTGA-Q500 setup.The

speciﬁcsurfacearea(SSA)oftheTiO2sampleswasdeterminedby

conventional Brunauer–Emmett–Teller (BET) N2 adsorption

methodwithaMicromeriticsTristar3000surfaceareaandpore

sizeanalyzer.PriortotheBETmeasurements,allofthesamples

wereoutgassedinvacuumfor2hat150◦C.

2.3. Photocatalyticperformanceanalysismeasurements

2.3.1. Gas-phasephotocatalyticoxidationperformance

measurements

ReactivityoftheTiO2microstructureswasstudiedvia

photo-catalyticNOoxidation(NO(g)+½O2(g)→NO2(g)).Thegasphase

photocatalyticactivityoftheTiO2microstructureswasanalyzedin

acustom-madecontinuousﬂowreactionsystem,whichisshown

in Scheme2.Theexperimentalsetupwascomprisedofa

high-puritygasmixturecontainingNO(g)(100ppmNO(g)inN2(g),Linde

GmbH),O2(g)(99.998%,LindeGmbH)andN2(g)(99.998%,Linde

GmbH) which was humidiﬁed with70% RH(relative humidity,

measured viaa Hanna HI 9565humidity analyzer atthe

sam-plepositioninthephotocatalyticreactor).Inatypicalgasphase

photocatalytic performance analysis test, a total gas ﬂow rate

of1SLM(SLM,standard litersperminute)wasused,wherethe

volumetric ﬂow ratesof N2(g),O2(g) andNO(g) weresettobe

0.750SLM, 0.250SLM and 0.010SLM via mass ﬂow controllers

(MFCs, MKS,1479A),respectively.Beforetheperformancetests,

synthesizedTiO2 microsphere/microbowl powdersampleswere

dispersedonapoly-methylmethacrylate(PMMA)sampleholder

(2×40×40mm3₎_and_irradiated_with_UVA_illumination_(Sylvania

UV-lamp,black-light,F8W,T5,368nm)underambientconditions

for18hinordertoremovethesurfacecontaminationsandto

acti-vatethephotocatalysts.Afterthisactivationanddecontamination

procedure,sampleswereinsertedintothephotocatalyticﬂow

reac-torforperformanceanalysis.UVAilluminationsourceusedinthe

performanceanalysistests(SylvaniaUV-lamp,black-light,F8W,T5,

368nm)generatedaUVAphotonﬂuxof7.5W/m2_at_the_sample

positionundertypicalreactionconditions.Duringtheperformance

tests, reaction gases were swept over a 950mg photocatalyst

sample and the concentration of NO(g), NO2(g) and total NOx

(g)speciesinthephotocatalyticreactorwerequantitatively

mea-sured online with a Horiba APNA-370 chemiluminiscence NOx

analyzer.

(3)

Gasphasephotocatalyticactivitymeasurementsarereportedin

termsofpercentphotonicefﬁciencies(%)asdescribedinEqs.(1)

and(2).

%= nNOx

nphoton×

100 (1)

wherenNOxcorrespondstoeitherthedecreaseinthetotalnumber

ofmolesofallgaseousNOxspeciesorthenumberofmolesofNO2(g)

generatedina60min(i.e.3600s)photocatalyticperformancetest.

Ontheotherhand,nphotoncorrespondstothetotalnumberofmoles

ofincidentUVAphotonsimpingingonthecatalystsurfacein3600s,

whichcanbecalculatedthroughEq(2)as:

nphoton=

(ISt)

(Nhc) (2)

whereIrepresentsthephoton powerdensityoftheUVAlamp,

experimentallymeasuredatthesamplepositioninthe

photocat-alyticreactor(typically,7.5W/m2_),_is_the_{representative}_emission

wavelengthoftheUVAlamp(i.e.368nm),Sisthesurfaceareaof

thephotocatalystsampleholderinthereactorthatisexposedto

theUVAirradiation(i.e.4cm×4cm=16cm2_);_t_is_the_duration_of

theperformancetest(i.e.3600s),NistheAvogadro’snumber,his

Planck’sconstantandcisthespeedoflight.

2.3.2. Liquid-phasephotocatalyticoxidationperformance

measurements

Liquid-phase photocatalytic oxidation activity of the TiO2

microstructureswasdemonstratedbyphotodegradation[16–18].

Oxidative degradation of Sulforhodamine B (RhB, 95%, Sigma)

underUVA irradiation (SylvaniaUV-lamp, F8W, T5,Black-light,

8W,368nm)wasconductedinabatch-modephotocatalytic

reac-torofdimensions45×23×28cm3_._An_aqueous_RhB_solution_at

concentrationof1mg/Land 30mg ofTiO2 microstructures was

addedintothereactorandstirredcontinuouslyatastirringrateof

100rpm.Then,thephotocatalyticdegradationprocesswas

stud-iedbymeasuring thechangein thedyeconcentrationwithan

UV–visspectrophotometer(Carry300,Agilent).Attenuationofthe

majorabsorptionbandofRhB(564nm)associatedwiththeS0→S1

absorption[19]wasrecordedevery30minuntilthetestsolution

becamevisuallytransparent.BeforetheUV–visabsorption

mea-surements,testsolutionswerecentrifugedandtheabsorbanceof

theﬁltratewasrecorded.Byusingacalibrationcurve(R2₌₉₉₉₄₎_of

thedyesolution,thepercentdecolorizationefﬁciency(Def)ofthe

systematanirradiationtimet(min)wascalculatedasdescribedin

Eq.(3)[20].

Def(%)= (C0_C−Ct)

0 ×

100 (3)

InEq.(3),C0andCtrepresenttheconcentrationofthetest

solu-tionbeforeandafterirradiationattimet,respectively.AplotofC0/C

versusirradiationtime(t)determinesthedecolorizationdegreeof

thetestsolution.

3. Resultsanddiscussion

3.1. Structuralcharacterizationofpolymer-templatedTiO2

microstructures

The SEM imagesin Fig.1a–d illustrate themorphology and

theparticlesizeoftheTiO2 coatedPS-co-DVBmicrospheres.The

particlesizevariationinthemicrostructuresstemsfromthe

cor-respondingsizedistributioninthenascentcommercialPS-co-DVB

material.SEMimagesinFig.1a–dandthecorrespondingEDX

mea-surements(Fig.1e)oftheTiO2-coatedmicrospheresrevealedthat

thesurfaceofthepolymermicrosphereswascoatedwithathin

layerofTiO2andadditionalTiO2wasalsofurtherdeposited.

Fig.1.(a–d)SEMimagesand(e)arepresentativeEDXspectrumofTiO2-coated

PS-co-DVBmicrospheresbeforecalcination.

UponcalcinationoftheTiO2coatedPS-co-DVBmicrospheres

between200and700◦C,signiﬁcantmorphologicalchangeswere

observed. The microspheres were converted into microbowls

(Fig.2).Thisobservationwasalsoaccompaniedbyaconsiderable

weightloss,whichwillbediscussed furtherinthetext(Fig.3).

Fig.2showstheSEMimagesandthecorrespondingEDXspectrum

ofthepolymer-templatedTiO2microbowls,whichwerecalcined

at600◦Catambientconditionsfor2h.Duetodecompositionof

thepolymertemplateandtheassociatedformationofHxCy(g)and

HxCyOz(g),pressureaccumulationinsidethemicrosphereleadsto

theruptureofthesphericalmorphologyduringtheevolutionof

theentrappedgas.Theresultingopenmicrobowlstructuresare

shownintheinsetofFig.2b.Theinteriorcavitiesofthemicrobowls

haveanaveragediameterof8␮mwithanaveragewallthickness

of 600nm.The EDXspectrum ofthemicrobowls(Fig.2b)

indi-cates TiO2/TiOx content witha relatively minorcontributionof

carbon-basedspecies.Ontheotherhand,EDXspectrumofthesame

samplesbeforethecalcinationrevealedexcessiveCsignal(Fig.1e).

EvolutionofHxCy(g)andHxCyOz(g)andtheanticipatedweight

loss of the sample upon the decomposition/degradation of the

polymertemplatebelow600◦Cisinperfectagreementwiththe

(4)

Fig.2.(a)SEMimage,(b)EDXspectrumofPS-co-DVBtemplatedTiO2microbowls

aftercalcinationat600◦_C_for₂_h_(inset_shows_the_detailed_morphology_of_the

microbowlsinSEM).

within400–500◦C.TheTGAcurveofTiO2-coatedPS-co-DVB

micro-spheres(Fig.3)exhibitsa2.7wt%lossinthetemperaturerangeof

30–250◦Cduetotheevaporationofwaterandothervolatile

organ-ics. TiO2 revealsa negligiblegravimetric losswithin 30–800◦C,

whilepure/uncoatedpolystyreneundergoesalmost100wt%loss

within 300–500◦C due to decomposition/degradation [21–23].

Fig.3. TGAmeasurementforPS-co-DVBtemplatedTiO2microspheres.

Thus, TGA data in Fig. 3, suggest that after the 71wt% loss at

T>400◦C,alargeportionoftheremainingsample,which

corre-spondsto29%oftheoriginalsampleweight,isduetotheinorganic

content(i.e.TiO2).

In order toinvestigate theinﬂuenceof the calcination

tem-perature onthe photocatalyst structure and the photocatalytic

activity,sampleswerecalcinedatdifferenttemperatureswithin

200–700◦C. CorrespondingXRD patternsand Ramanspectraof

thesesamplesarepresentedinFig.4.Calcinationat200and300◦C

leadstotheformationofanamorphousTiO2/TiOxstructure,which

startstocrystallizeintoaratherdisorderedanatasephaseat400◦C

withasmallaverageparticlesize,evidentfromthecorresponding

broadanataseXRDdiffractionsignals(ICDDCardNo:21-1272)in

Fig.4aandthecharacteristicallyintenseanataseRamanscattering

observedat144cm−1 [24–26].At500◦C,awell-orderedanatase

phasewithalargeraverageparticlesizeisformedascanbeseen

fromthesharpandintenseanatasesignalsinbothXRD(Fig.4a)

andRaman(Fig.4b)results.Atthistemperature,rutilephasealso

appearsasasecondaryphaseinbothXRDresultsshowninFig.4a

(ICDDcardno:04-0551)aswellasintheRamandatainFig.4b.

For-mationoftherutilephaseleadstotheevolutionoftypicalRaman

scatteringfeaturesat236,447,612,826cm−1[24–26].Rutilephase

becomesmorecrystallineandabundantathighercalcination

tem-peratures.Uponcalcinationat700◦C,rutilebecomesthedominant

phase,althoughanatasephasecanstillbedetectedasasecondary

phase(Fig.4aandb).

3.2. Photocatalyticactivityofthepolymer-templatedTiO2

microspheresandmicrobowls

3.2.1. Gas-phasephotocatalyticoxidationperformance

ThephotocatalyticNO(g)oxidation withO2(g)wasusedasa

modelreaction[27–32].Fig.5illustratesatypicalgasphase

pho-tocatalyticperformanceanalysistestinwhichthephotocatalyst

sampleisexposedtoafeedgasmixturecontaining1ppmNO(g)

aswellasacertaincompositionofN2(g)andO2(g)witha70%RH

(seeSection2fordetails).Fig.5showsthetime-dependent

pro-ﬁlesforthetotalNOxconcentration(i.e.sumoftheconcentrations

ofalloftheNOxspeciesexistinginthereactor,i.e.bluecurvein

Fig.5)as wellasseparateNO(g)(blackcurve) and NO2(g)(red

curve)concentrationsinthephotocatalyticreactormeasuredby

thechemiluminiscenceNOxanalyzer.AsshowninFig.5,during

theinitial15minoftheperformancetest,gasmixturecontaining

1ppmNO(g)isfedtothephotocatalystwhileUVAlampisinoff

positionandthereactoris keptindarkinordertopreventany

exposuretosunlight.Thisleadstoaminortransientfallinthetotal

NOx(g)andNO(g)concentrations,whichisassociatedwiththe

dilu-tionofthegasinthereactorpipelineandthethermaladsorption

of NOxspecies onthegaslines,reactorwallsaswellasonthe

photocatalystsurface.Asthesystemiskeptindarkunderthese

conditions,nophotocatalyticactivityisobservedduringthis

ini-tialstage,whichisevidentbythelackofanyNO2(g)production.

Aftertheinitialtransientperiod,reactorwallsandthe

photocata-lystsurfacearesaturatedwithNOx,afterwhichNOx(g)andNO(g)

tracesquicklyreturntotheoriginalinletconcentrationvalueof

1ppm.

Next,UVAlampisturnedonandthephotocatalyticreaction

is started.UponUVAradiation,asharpand apermanentfallin

the NO(g) and total NOx(g) concentrations along with a quick

rise inNO2(g)signal,wereobserved. Thisiscaused by

conver-sionofNO(g)intoNO2(g)viaphotocatalyticoxidation.Inaddition,

generated NO2(g)can also adsorbonthe photocatalystsurface

intheformofchemisorbedNO2,nitric/nitrousacid,nitritesand

nitrates [24–26,33] and stored in the solid state, leading to a

furtherdecreaseintheNO(g)signal.Furthermore,direct

(5)

100 200 300 400 500 600 700 800 A A A PsTi-200 PsTi-300 PsTi-400 PsTi-500 PsTi-700

Raman Intensity (a.u.)

Raman Shift (cm-1) x20 R A R R A A PsTi-600 A R A R A 10 000 10 20 30 40 50 60 Intensity (a.u) 2θ(deg) PsTi-700 PsTi-600 PsTi-500 PsTi-400 PsTi-300 PsTi-200 1000 _R R R R A R R A A AA R A R R A R R A R A A A R R A R R A A _A _A A (a) (b) Ra m an In tensi ty (a.u.) Raman Shi (cm-1)

Fig.4. (a)XRDpatterns,(b)RamanspectraofPS-co-DVBtemplatedTiO2microspheres/microbowlsuponcalcinationat200◦C,300◦C,400◦C,500◦C,600◦C,and700◦Cfor

2hunderambientconditions(insethighlightsthedetailedRamanfeaturesofPsTi-600andPsTi-700samples).A:anatase,R:rutile.

and/orN2O(g)cannotberuledout[34].ThetotalNOx

concentra-tion(blue)curve(whichismostlycomprisedofthesumofNO(g)

andNO2(g)signals)inFig.5staysalwaysbelow1ppmduringthe

UVA-activatedregime,illustratingthecontinuousphotocatalytic

activity.

Gas-phasephotocatalyticperformancetestssimilartotheone

giveninFig.5werealsoperformedonotherPS-co-DVBtemplated

TiO2microsphere/microbowlphotocatalysts,whichwerecalcined

atvarioustemperaturesbetween200and700◦C.Percentphotonic

efﬁciencyvaluesderivedfromsuchexperimentsareshowninFig.6,

wherebluebarsrepresentthe%photonicefﬁciencyoftotalNOx(g)

decrease,whileredbarscorrespondtothe%photonicefﬁciencyof

NO2(g)production.

Fig.6showsthatPsTi-200samplerevealsbothconsiderableNOx

storage(bluebar) andNO2(g)production (redbar)capabilities.

0 20 40 60 80 0.0 0.2 0.4 0.6 0.8 1.0 Concentration (ppm) Time(min) Thermal NOx adsorpon

Light-on Light-oﬀ

NOx (g)

NO(g)

NO2(g)

Fig.5.Typicaltime-dependentconcentrationproﬁlesfortotalNOx(g),NO(g)and

NO2(g)overPS-co-DVBtemplatedTiO2microbowlphotocatalyst(PsTi-600)during

gas-phasephotocatalyticNOoxidationactivitytests.(Forinterpretationofthe ref-erencestocolorinthisﬁgurelegend,thereaderisreferredtothewebversionof thisarticle.)

Ontheotherhand,uponincreasingthecalcinationtemperature

to300◦C,bothNOxstorageandNO2(g)productionperformances

wereobservedtodeclinedrastically.Ontheotherhand,after

calci-nationat400◦C,NOxstoragecapabilityisrecoveredwhileNO2(g)

productionisstillnoticeablysuppressed.Above500◦C,although

NOx storagecapacitydecreasestoa certainextent,NO2(g)

pro-ductioncapabilityisfullyregained,reachingitshighestvalueat

600◦C.Increasingthecalcinationtemperatureto700◦Cleadstoa

decreaseintheNOxstorageandNO2(g)productionperformances

simultaneously.

Interestinggas-phasephotocatalyticperformancetrendsgiven

inFig.6canbeelucidatedbyusingthestructuralpropertiesofthe

polymer-templatedTiO2microstructuresshowninScheme3.The

crosslinkedpolystyrenesystemshavetypicalglasstransition

tem-peratures(Tg)within100–150◦C,abovewhichthesolidpolymer

tendstoswitchtoamobilemolten/glassystate[23].Ascanbeseen

fromthespeciﬁcsurfacearea(SSA)resultsshown inScheme3,

PsTi-200samplehasamoderatelyhighSSA(86m2_/g)_suggesting

that the mobilized PS-co-DVB microsphere template starts to

segregateontheverytopsurface,onlypartiallycovering/blocking

Fig.6. ComparisonofthephotonicefﬁcienciesofTiO2microspheres/microbowls.

(Forinterpretationofthereferencestocolorinthisﬁgurelegend,thereaderis referredtothewebversionofthisarticle.)

(6)

Scheme3.Temperature-inducedstructuralevolutionofTiO2microspheres/microbowls.

the amorphous TiO2/TiOx coating on the microsphere system.

Thus, at this calcination temperature, TiO2/TiOx coating is still

partiallyaccessibleforgasphasephotocatalyticNOxstorageand

NO2(g)production(Fig.6).

However,uponcalcinationat300◦C,theSSAwasobservedto

decreasebyabout50%,whichisaccompaniedbyatotallossof

pho-tocatalyticNOxstorageandNO2(g)productionactivities(Fig.6).

Apparently,calcinationat300◦Cleadstothesegregationofthe

mobilized PS-co-DVB microsphere template onto theTiO2/TiOx

coating(Scheme3).Hence,accesstothephotocatalyticactivesites

toNO(g)iscompletelyblockedandthephotocatalyticactivityis

entirelylost.

Increasingthecalcinationtemperatureto400◦Cshowsaunique

switchinthephotocatalyticactivity.Thisistheborderline

tem-perature, where the PS-co-DVB template starts to decompose

leadingtotheruptureofthemicrospheresandformationofthe

microbowls.Formationofmicrobowlsandeliminationofthe

car-bonaceous/polymericﬁlmat400◦Cisalsofullyconsistentwiththe

drasticincreaseintheSSAofthesystemto159m2_/g₍_Scheme₃_).

The increase in theSSA is also accompanied by theformation

of a cavity inside the microspheresdue to the degradation of

thePS-co-DVBtemplate,generatingadditionaladsorptionsites.At

thistemperature,Ti-coatingrevealsmostlyanamorphous/porous

nature, which also exhibits poorly crystallineanatase domains

(Fig.4).Thus,duetothedecomposition/removalofthepolymer

template,mostofthephotocatalyticactivesitesontheamorphous

Ti-coatingbecomereadilyaccessibleandphotocatalyticNO

oxida-tioncanbeperformedefﬁcientlywhichisevidentbytherecovery

ofthephotocatalyticNOxstorage(bluebarforPsTi-400inFig.6).

AlthoughPsTi-400samplecanefﬁcientlyperformphotocatalytic

NOxstorage,yetitgeneratesarelativelysmallamountofNO2(g).

ThiscouldbeduetothelargeSSAofthePsTi-400samplewith

alargenumberofadsorptionsitesthatcanimmediatelycapture

NO2(g)intheformofnitritesandnitratesontheTiO2surfaceand

preventNO2(g)slipintothegasphase.

Fig.6showsthatasthecalcinationtemperatureisincreased

from400◦C to500◦C, thephotocatalyticNOxstoragedecreases

signiﬁcantlyincontrasttothenoticeableincreaseintheNO2(g)

production.Within400–500◦C,PsTisamplesundergoasubstantial

crystallographictransformation(Fig.4),whereporousand

amor-phous TiO2 domains crystalize into ordered anatase and rutile

domainsresultinginasigniﬁcantlossintheSSA.Alongtheselines,

PsTi-500samplehasaSSAof13.9m2_/g₍_Scheme₃_)._Thus,_the

pho-tocatalyticNOxstoragecapacityfallsinlinewiththecorresponding

theSSAloss,suggestingthatNO2(g)generatedviaphoto-oxidation

readilyslipsintothegasphase.However,thisdoesnotmeanthat

thephotocatalyticactivitydecreasesuponincreasingthe

tempera-turefrom400◦Cto500◦C.BycomparingthecombinedNOxstorage

andNO2formationresults(i.e.sumoftheredandbluebarsinFig.6)

for400◦Cand500◦CalongwiththecorrespondingSSAvalues

sug-geststhatPsTi-500samplehasaconsiderablyhigherper-sitebasis

photocatalyticactivitywithrespecttoPsTi-400.

Fig. 6 indicates that the optimum gas-phase photocatalytic

activityisreachedforthePsTi-600sample,whichrevealsalower

anatase/rutileratio(Fig.4aandScheme3)estimatedbyXRDresults

byusingtheapproachdevelopedbySpurrandMyers[35].Onthe

otherhand,asthecalcinationtemperatureisincreasedto700◦C,

concomitanttothefurtherdecreaseintheanatase/rutileratio,

pho-tocatalyticactivitystartstodecrease.Thus,itisapparentthatrather

thanthesoleSSAvalues,crystallographicandelectronicproperties

oftheTiO2 microspheres/microbowlsplayamajorrolein

deter-miningtheirultimategas-phasephotocatalyticactivities.

3.2.2. Solution-phasephotocatalyticoxidationperformance

PhotocatalyticactivityofTiO2microstructurecalcinedat

(7)

Fig.7. Time-dependentUV–VisabsorptionspectrashowingUVA-induced pho-tocatalytic degradation ofRhB in thepresence of PS-co-DVBtemplated TiO2

microbowlscalcinedat600◦Cfor2h.

photocatalyticoxidationofRhB.Atypicalseriesoftime-dependent

UV–visabsorptionspectraobtainedduringtheUVAirradiationis

presentedinFig.7.Thisseriesofspectracorrespondstothe

PsTi-600samplewhichiscomprisedofTiO2microbowls(Fig.2).During

thephotocatalyticreaction,thecharacteristicRhBabsorptionband

locatedat 564nm graduallydecreasesindicating photocatalytic

degradation/oxidationof RhB.After330minof UVAirradiation,

thedyesolutionbecomesvisiblycolorlessandthe564nmsignal

vanishesalmostcompletely.

Time-dependent decolorization efﬁciency results for the

remainingsamplesaresummarizedinFig.8a.Thesolutionphase

photocatalyticoxidationexperimentscouldnotberealizedforthe

PsTi-200andPsTi-300samplesduetolowdensityofthe

corre-spondingsolidphotocatalysts(originatingfromtheirhighpolymer

content),whichresultsintheﬂoatingofthemicrospheresonthe

Fig.8.(a)Liquid-phasephotocatalyticreactivity ofPS-co-DVBtemplatedTiO2

microspheres/microbowlsinRhodamineBphotodegradationviaUVAirradiation, (b)photocatalyst-containing1mg/LRhBsolutionsafter18hUVAirradiation.

aqueousmediumpreventingtheirefﬁcientmixingand

homoge-nousUVAexposure.Fig.8ashowsthatRhBconcentrationinthe

solutiondecreasesmonotonicallywithincreasingirradiationtime

which isalsoillustratedinFig.8b (forphotocatalyst-containing

1mg/LRhBsolutionsafter18hUVAirradiation).Control

experi-mentsperformedbyexposing1mg/LRhBsolutiontoUVAinthe

absenceofaphotocatalyst(datanotshown)didnotleadtoany

decolorizationundertypicalreactionconditions.Theliquid-phase

photocatalyticactivityofthesynthesizedTiO2structuresexhibits

astrongdependenceonthecalcinationtemperature.Fig.8aclearly

indicatesthatPsTi-600samplewhichhasamicrobowlstructure

(Fig.2)andexhibitspredominantlyanatasephase(inadditionto

rutileasasecondaryphase)revealsthehighestliquid-phase

pho-tocatalyticactivity.ThePsTi-400sampleissigniﬁcantlylessactive

thanalloftheanalyzedsamples(Fig.8),andiscomprisedofapoorly

crystallineanatasephase(Fig.4).Thissuggeststhatsolution-phase

photocatalyticactivityrequiresformationoforderedanatase/rutile

crystallographicphases.Ontheotherhand,Fig.8aalsoshowsthat

thesolution-phasephotocatalyticactivitytendstodecreaseat

ele-vatedcalcinationtemperaturessuchas700◦C,suggestingthata

rutile-dominantTiO2microbowlstructureisnotfavorable.

Itisworthmentioningthatthesolution-phasephotocatalytic

reactivitytrendspresentedinFig.8acannotbeexplainedsolely

basedontheSSAvalues ofthesynthesizedmaterials.Although

PsTi-400 sample reveals a signiﬁcantly higher SSA than all of

the other synthesized materials, it has a considerably lower

liquid-phasephotocatalyticactivity(Fig.8).Inotherwords,

crys-tallographic and the electronic properties of the TiO2-coated

Ps-co-DVBmicrospheres/microbowlsseemtoplayamajorrolein

theirliquid-phasephotocatalyticreactivity.

It is worthmentioningthat wehave alsoperformed similar

liquid-phaseand gas-phase photocatalyticactivitytests usinga

benchmarkphotocatalyst(P25)(Figs.S1andS2,ESI†).Weobserved

thattotalphotocatalyticactivityforP25inbothliquidandgasphase

experimentswereabouttwo timeshigherthanthatofthebest

Ps-co-DVBtemplatedTiO2microsphere/microbowlphotocatalyst

(PsTi-600).TheSSAofP25isabout50m2_/g,_which_is_about_more

than5timesgreaterthanthatofPsTi-600.Thus,per-sitebasis

pho-tocatalyticactivityofPsTi-600isstill2.5timeshigherthanthat

ofP25.Thissuggeststhatbyoptimizingthepolymermicrosphere

templatingstrategy(forinstancebyusingpolymernanospheres

withsmalleraverageparticlesizesandthushigherSSA),advanced

photocatalyticsystemscanbedesigned,whichrevealhigher

photo-catalyticperformancebothintermsoftotalphotocatalyticactivity

aswellasper-site-basisphotocatalyticactivity.Inaddition,further

improvements in the photocatalytic performance of PS-co-DVB

templatedTiO2microsphere/microbowlphotocatalystscanalsobe

achievedbyincorporatingplasmonicmetalnanoparticlestothese

systems[36].Suchexperimentaleffortsarecurrentlyunderwayin

ourresearchgroup[37].

4. Conclusions

In this work, Ps-co-DVB microsphere templated TiO2

pho-tocatalysts were synthesized via sol–gel method. Inﬂuence of

thecalcinationtemperatureonthestructuralpropertiesandthe

photocatalytic activity of these systems under UVA excitation

wereinvestigatedboth inthegasphase(bystudying

photocat-alyticNO(g)oxidationbyO2(g))aswellasinthesolutionphase

(by monitoring Rhodamine B photocatalytic degradation). The

polymermicrosphereswerefoundtobecoveredwithathinﬁlmof

TiO2/TiOxaswellasTiO2/TiOxnanoparticles.Photocatalyticactivity

carriedoutinthesolutionphaseandinthegasphaseshowedthat

the photocatalyst calcined at 600◦C exhibiting a microbowl

(8)

activity which is even greater than that of the commercial

benchmarkP25.Our ﬁndingsindicatethatnot onlythespeciﬁc

surfaceareabutalsothecrystallographicandelectronicproperties

oftheTiO2microstructuresplayamajorroleindeterminingtheir

ultimate photocatalytic activities. This suggests that

polymer-templated TiO2 microstructures offer a promising versatile

syntheticplatformforphotocatalyticDeNOx applications,which

canbefurtherimprovedbyusingpolymernanospheretemplates

withhigherSSAorbyadditionalfunctionalizationwithtransition

metalnanoparticlesand/orplasmoniccomponents.

Acknowledgments

AuthorsgratefullyacknowledgeAssociateProf.Dönüs¸Tuncel

for fruitful discussions,and Zafer Say forperforming BET

mea-surements.E.O.alsoacknowledgesﬁnancialsupportfromTurkish

AcademyofSciencesthroughthe“TUBA-GEBIPOutstandingYoung

Scientist Prize” and from Fevzi Akkaya Science Fund (FABED)

throughEserTümenScientiﬁcAchievementAwardaswellasthe

Scientiﬁc and Technical Research Council of Turkey (TUBITAK)

(ProjectCode:109M713).

AppendixA. Supplementarydata

Supplementary material related to this article can be

found, in the online version, at http://dx.doi.org/10.1016/j.

apsusc.2014.04.082.

References

[1]U.Diebold,Thesurfacescienceoftitaniumdioxide,Surf.Sci.Rep.48(2003) 53–229.

[2]T.L.Thompson,J.T.Yates,SurfacesciencestudiesofthephotoactivationofTiO2

-newphotochemicalprocesses,Chem.Rev.106(2006)4428–4453.

[3]X.Chen,S.S.Mao,Titaniumdioxidenanomaterials:synthesis,properties, mod-iﬁcations,andapplications,Chem.Rev.107(2007)2891–2959.

[4]A.Fujishima,X.Zhang,D.A.Tryk,TiO2photocatalysisandrelatedsurface

phe-nomena,Surf.Sci.Rep.63(2008)515–582.

[5]M.A.Henderson,AsurfacescienceperspectiveonTiO2photocatalysis,Surf.Sci.

Rep.66(2011)185–297.

[6]Y.Liu,J.Goebl,Y.Yin,Templatedsynthesisofnanostructuredmaterials,Chem. Soc.Rev.42(2013)2610–2653.

[7]C.N.R.Rao,A.Govindaraj,Synthesisofinorganicnanotubes,Adv.Mater.21 (2009)4208–4233.

[8]A.S.Attar,M.S.Ghamsari,F.Hajiesmaeilbaigi,S.Mirdamadi,K.Katagiri,K. Koumoto,Sol–geltemplatesynthesisandcharacterizationofaligned anatase-TiO2nanorodarrayswithdifferentdiameter,Mater.Chem.Phys.113(2009)

856–860.

[9]F.Cesano,D.Pellerej,D.Scarano,G.Ricchiardi,A.Zecchina,Radiallyorganized pillarsinTiO2 andinTiO2/Cmicrospheres:synthesis,characterizationand

photocatalytictests,J.Photochem.Photobiol.,A:Chem.242(2012)51–58.

[10]J.Hu,M.Chen,X.Fang,L.Wu,Fabricationandapplicationofinorganichollow spheres,Chem.Soc.Rev.40(2011)5472–5491.

[11]T.Kang,A.P.Smith,B.E.Taylor,M.F.Durstock,Fabricationofhighly-ordered TiO2nanotubearraysandtheiruseindye-sensitizedsolarcells,NanoLett.9

(2009)601–606.

[12]I.Lee,J.B.Joo,Y.Yin,F.Zaera,Ayolk@shellnanoarchitectureforAu/TiO2

cata-lysts,Angew.Chem.Int.50(2011)10208–10211.

[13]U.Meyer,A.Larsson,H.Hentze,R.A. Caruso,Templating ofporous poly-mericbeadstoformporoussilicaandtitaniaspheres,Adv.Mater.14(2002) 1768–1772.

[14]A.S. Deshpande, D.G.Shchukin, E.Ustinovich, M. Antonietti,R.A. Caruso, Titania and mixed titania/aluminum, gallium, or indium oxide spheres: Sol–gel/templatesynthesisandphotocatalyticproperties,Adv.Funct.Mater. 15(2005)239–245.

[15]J.H. Schattka, D.G.Shchukin, J. Jia,M. Antonietti,R.A. Caruso, Photocat-alyticactivitiesofporoustitaniaandtitania/zirconiastructuresformedby usingapolymergeltemplatingtechnique,Chem.Mater.14(2002)5103– 5108.

[16]L.Yuan,S.Meng,Y.Zhou,Z.Yue,ControlledsynthesisofanataseTiO2nanotube

andnanowirearraysviaAAOtemplate-basedhydrolysis,J.Mater.Chem.A1 (2013)2552–2557.

[17]D.A.H.Hanaor,C.C.Sorrell,Reviewoftheanatasetorutilephasetransformation, J.Mater.Sci.46(2011)855–874.

[18]P.Zhang,C.Shao,Z.Zhang,M.Zhang,J.Mu,Z.Guo,Y.Sun,Y.Liu,Core/shell nanoﬁbersofTiO2@carbonembeddedbyAgnanoparticleswithenhanced

vis-iblephotocatalyticactivity,J.Mater.Chem.21(2011)17746–17753.

[19]N.K.M.N.Srinivas,S.V.Rao,D.N.Rao,Saturableandreversesaturableabsorption ofRhodamineBinmethanolandwater,J.Opt.Soc.Am.B:Opt.Phys.20(2003) 2470–2479.

[20]K.Sridharan,T.J.Park,Thorn-ballshapedTiO2nanostructures:InﬂuenceofSn2+

dopingonthemorphologyandenhancedvisiblelightphotocatalyticactivity, Appl.Catal.,B:Environ.134(2013)174–184.

[21]D.Yu,J.H.An,S.D.Ahn,S.Kang,K.S.Suh,Titaniumdioxide/P(St-co-DVB)-MAA hybridcompositeparticlespreparedbydispersionpolymerization,Colloid Surf.,A:Physicochem.Eng.Aspects266(2005)62–67.

[22]E.D.Mekeridis,I.A.Kartsonakis,G.S.Pappas,G.C.Kordas,Releasestudiesof corrosioninhibitorsfromceriumtitaniumoxidenanocontainers,J.Nanopart. Res.13(2011)541–554.

[23]S.E.Shim,S.Yang, H.H.Choi, S.Choe,Fully crosslinked poly(styrene-co-divinylbenzene) microspheres by precipitation polymerization and their superiorthermalproperties,J.Polym.Sci.,PartA:Polym.Chem.42(2004) 835–845.

[24]S.M.Andonova,G.S.S¸entürk,E.Kayhan,E.Ozensoy,NatureoftheTi–Ba inter-actionsontheBaO/TiO2/Al2O3NOxstoragesystem,J.Phys.Chem.C113(2009)

11014–11026.

[25]S.M.Andonova,G.S.S¸entürk,E.Kayhan,E.Ozensoy,Fine-tuningthedispersion andthemobilityofBaOdomainsonNOxstoragematerialsviaTiO2anchoring

sites,J.Phys.Chem.C114(2010)17003–17016.

[26]G.S.S¸entürk,E.I.Vovk,V.I.Zaikovskii,Z.Say,A.M.Soylu,V.I.Bukhtiyarov,E. Ozensoy,SOxuptakeandreleasepropertiesofTiO2/Al2O3andBaO/TiO2/Al2O3

mixedoxidesystemsasNOxstoragematerials,Catal.Today184(2012)54–71. [27]T.Giannakopoulou,N.Todorova,G.Romanos,T.Vaimakis,R.Dillert,D. Bahne-mann,C.Trapalis,Compositehydroxyapatite/TiO2materialsforphotocatalytic

oxidationofNOx,Mat.Sci.Eng.B177(2012)1046–1052.

[28]A.Mitsionis,T.Vaimakis,C.Trapalis,N.Todorova,D.Bahnemann,R.Dillert, Hydroxyapatite/titanium dioxidenanocomposites for controlled photocat-alyticNOoxidation,Appl.Catal.,B:Environ.106(2011)398–404.

[29]Y.Lin,Y.Tseng,J.Huang,C.C.Chao,C.Chen,I.Wang,Photocatalyticactivity fordegradationofnitrogenoxidesovervisiblelightresponsivetitania-based photocatalysts,Environ.Sci.Technol.40(2006)1616–1621.

[30]Y.Tseng,C.Kuo,C.Huang,Y.Li,P.Chou,C.Cheng,M.Wong, Visible-light-responsivenano-TiO2withmixedcrystallatticeanditsphotocatalyticactivity,

Nanotechnology17(2006)2490–2497.

[31]N.Negishi,K.Takeuchi,T.Ibusuki,SurfacestructureoftheTiO2thinﬁlm

pho-tocatalyst,J.Mater.Sci.33(1998)5789–5794.

[32]T.Sano,N.Negishi,K.Koike,K.Takeuchi,S.Matsuzawa,Preparationof visible-light-responsivephotocatalystfromcomplexofTi4+withnitrogen-containing

ligand,J.Mater.Chem.14(2004)380–384.

[33]R.Hummatov,O.Gülseren,E.Ozensoy,D.Toffoli,H.Üstünel,First-principles investigationofNOx andSOx adsorptiononanatase-supportedBaOandPt

overlayers,J.Phys.Chem.C116(2012)6191–6199.

[34]O.Carp,C.L.Huisman,A.Reller,Photoinducedreactivityoftitaniumdioxide, Prog.SolidStateChem.32(2004)33–177.

[35]R.A.Spurr,H.Myers,Quantitativeanalysisofanatase-rutilemixtureswithan X-raydiffractometer,Anal.Chem.29(1957)760–762.

[36]J.Lu,P.Zhang,A.Li,F.Su,T.Whang,Y.Liu,J.Gong,MesoporousanataseTiO2

nanocupswithplasmonicmetaldecorationforhighlyactivevisible-light pho-tocatalysis,Chem.Commun.49(2013)5817–5819.