Comparison of the photocatalytic efficiencies of bare and doped rutile and anatase TiO2 photocatalysts under visible light for phenol degradation and E. coli inactivation

(1)

ContentslistsavailableatSciVerseScienceDirect

Applied Catalysis B: Environmental

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

Comparison of the photocatalytic efﬁciencies of bare and doped rutile and

anatase TiO 2 photocatalysts under visible light for phenol degradation and E. coli inactivation

G. Veréb

^a

, L. Manczinger

^b

, G. Bozsó

^c

, A. Sienkiewicz

^d

, L. Forró

^d

, K. Mogyorósi

^a,∗

, K. Hernádi

^a

, A. Dombi

^a

aResearchGroupofEnvironmentalChemistry,InstituteofChemistry,FacultyofSciencesandInformatics,UniversityofSzeged,H-6720,Szeged,TiszaLajoskrt.103,Hungary

bDepartmentofMicrobiology,FacultyofSciencesandInformatics,UniversityofSzeged,H-6701,P.O.Box533,Szeged,Hungary

cDepartmentofMineralogy,Geochemistry,andPetrology,FacultyofSciencesandInformatics,UniversityofSzeged,H-6701,P.O.Box651,Szeged,Hungary

dFSB,IPMC,LPMC,Station3,EcolePolytechniqueFédéraledeLausanne,CH-1015Lausanne,Switzerland

a r t i c l e i n f o

Articlehistory:

Received15June2012 Receivedinrevisedform 22September2012 Accepted26September2012 Available online 8 October 2012

Keywords:

Photocatalysis Visiblelight Disinfection Hydroxylradical Dopedtitania ESR Singletoxygen

a b s t r a c t

ThisstudyaimedatcomparingthephotocatalyticefficienciesofvariousTiO2basedphotocatalystsfor phenoldegradationandbacteriainactivationunderilluminationwithvisiblelight.Commercialundoped anataseandrutile(bothfromAldrich),AeroxideP25(EvonikIndustries),nitrogen-dopedanatase(Sumit- omoTP-S201,SumitomoChemicalInc.),nitrogenandsulphurco-dopedanatase(KronosVLP7000,Kronos TitanGmbH),andourcustom-synthesizednitrogen-andiron-dopedTiO2,aswellasnitrogenandsulphur co-dopedAeroxideP25andsilver-andgold-depositedAeroxideP25werestudied.Thephotocatalytic efficiencyofdifferenttypesoftitaniumdioxidebasedphotocatalystswasdeterminedbyinactivationof EscherichiacoliK12bacteriaandbyphenoldecomposition.Electronspinresonance(ESR)incombination withspintrappingwasusedtogetinsightintothereactiveoxygenspecies(ROS)-mediatedphotocatalytic processesinthepresenceofTiO2-basedphotocatalysts.ESRresultsconfirmedthattitaniaswhichgener- atedOH^•radicalswereefficientinE.colidisinfection,whereastitaniasthatwereunabletoproduceOH^• radicalsdidnotrevealsignificantbactericidalaction.Threeofourhome-madetitanias(iron-,nitrogen-, nitrogen/sulphur)aswellasthecommercialnitrogen/sulphurcodopedKronosVLP7000TiO2showed higherefficiencyofphenoldegradationthanthewell-establishedreferencephotocatalyst,AeroxideP25, butshowedmuchlower(ifany)activityforbacteriainactivation,includingKronosVLP7000,which revealedextremelyhighefficiencyforphenoldecomposition.InterestinglyundopedAldrichrutile(with largeparticles-100–700nm)hadthehighestefficiencyforinactivationofE.coliandalsohadfairlyhigh activityofphenoldegradation.

1. Introduction

Photocatalysisisanintensivelyinvestigatedalternativewater treatmentmethodnowadayswhich isapromisingtechniqueto decomposeorganicpollutantsinwater.Averyimportantadvan- tageofphotocatalyticpurificationproceduresisthatitisapplicable forawiderangeoforganicpollutants.Themethodissuitablefor deactivationofvariousmicroorganismsaswell.Duetoitshigh efficiency,lowtoxicity,excellentphysico-chemicalstability,and lowrelativecosts,titaniumdioxide(titania,TiO₂)isconsideredas themostpromisingphotocatalystforenvironmentalpurification.

TheﬁrstreportonTiO₂basedwaterdisinfectionwaspublishedin 1985byMatsunagaetal.[1].Sincethiswork,inactivationstudies

∗Correspondingauthor.Tel.:+3662544334;fax:+3662420505.

E-mailaddress:k.mogyorosi@chem.u-szeged.hu(K.Mogyorósi).

ofmanymodelmicroorganismswerecarriedoutinthepresence ofTiO2-basedphotocatalysts.

Inthemajorityofthesestudies,UVirradiationwasappliedto purifywaterandinactivatemicroorganismsinthepresenceofvar- iousformsoftitania[1–14].Howeveritiswellknownthatinthe solarspectrumthereisamuchhigherintensityinthevisiblelight range(∼43%oftotalsolarenergy)thanintheUVrange(∼3%)[8,15], andinanenvironmentallyfriendlyandeconomicalprocess,solar irradiationcouldbeappliedtoactivatethephotocatalyst[16–21].

Thereforeitisimportanttodevelopvisiblelightactivephotocata- lystswhichareabletodecomposeorganicpollutantsandtokill bacteria.Visiblelightactivephotocatalystsarealsoimportantfor indoorapplications,e.g.inair andsurface purificationatwhich naturallyUVlightispracticallyabsent.Inparticular,visiblelight activateddetoxification and/ordisinfectionprocesseshavebeen reportedforTiO₂ dopedwithnitrogen[13,22–25],iron[26–28], iodine[8,29–33],sulphur[13,22,34],aswellasformetal-modified TiO2,e.g.withsilver[8,23,35]orgold[36].

http://dx.doi.org/10.1016/j.apcatb.2012.09.045

(2)

Themaingoalofthisstudywastogiveanoverallpictureof theperformanceof differenttypeof bareand modiﬁedtitanias usingvisiblelightforthedecompositionoforganicpollutantsand killingbacteria.Theefﬁciencyofseveralbare,doped(N,Fe,S)and noblemetaldeposited(Ag,Au)titaniumdioxideswerecompared byphenoldegradationandbydeactivationofEscherichiacoli.

We also aimed to investigate the correlation between the efficiencyoforganicpollutiondegradationand disinfectionperformance, as well as to get an insight into the underpinning photocatalyticalmechanisms.Itiswellknownthatexcitationof TiO2 byphotonswithenergiesgreaterthanthebandgapyields formation of electron–hole (e⁻_cb)/(h⁺_vb) pairs, which can either recombine or take part in redox reactions. As a result, differ- enttypes of reactiveoxygenspecies (ROS) can beformed that arecapableofdegradingorganicpollutantsor damagingbacte- rialcells[2,8,24].Irelandetal.[2]emphasizedtheimportanceof OH^• radicalsbecauseincontrastwithothertypeofwatertreat- menttechnologies(e.g.:ozonation,directphotolysisofhydrogen peroxide,andradiolysis)whichinherentlyproduceOH^•radicalin verysmallquantities(<10⁻¹²M)[37],inaphotocatalyticsystem 10⁻⁹MOH^•concentrationwasmeasured[38].Authorspublished thattherewasnoinactivationofE.coliinthepresenceoftheOH^• radicalscavenger[2].Manyotherauthorsrecognizedtheimpor- tanceofOH^• generationinphotocatalyticdisinfectionprocesses [2,3,5,6,8,24,34,39–43].Kikuchietal.[40]publishedthatthepho- tocatalyticbactericidal effectof E.colionilluminated TiO₂ was confirmedonboth oxidationandreduction sites,corresponding to OH^• and O2•− production, respectively. However theactual lethalagentisH₂O₂,subsequentlyproducedfromOH^•andO₂^•−, particularlyinthelong-rangebactericidaleffect[40].Inaddition Rengifo-Herreraetal.[13]reportedthatsuperoxideradical(O2•−) anditsoxidationproduct:¹O₂–singletoxygenwereresponsible forE.coliinactivationbyN,Sco-dopedTiO₂nanoparticlesunder visiblelightillumination.Inthisstudy,weusedelectronspinres- onance(ESR)spectroscopytoexplorethepossibledifferencesin efficienciesandthegeneratedROSbyselecteddopedandundoped titania-basedphotocatalysts.

2. Experimental

2.1. Commercialphotocatalysts

Aeroxide P25 (TiO₂-P25, Evonik Industries), Aldrich anatase (TiO₂-AA),andAldrichrutile(TiO₂-AR)wereselectedascommer- ciallyavailableundopedphotocatalysts.

TiO₂-TP-S201(SumitomoChemicalInc.)andTiO₂-VLP700(Kro- nos Titan GmbH) doped with nitrogen and nitrogen/sulphur, respectively,werechosenascommerciallyavailabledopedTiO2

basedphotocatalysts.

2.2. HomemadeTiO₂basedphotocatalysts

Thehome-madetitaniaswerenitrogen(TiO2-N),iron(TiO2-Fe), sulphur and nitrogen (TiO₂-P25-NS) doped and silver (TiO₂- P25-Ag) and gold (TiO₂-P25-Au) deposited titanium dioxide photocatalysts.

Nitrogendopedtitania(TiO₂-N)waspreparedbythehydroly- sisoftitanium(IV)chlorideinnitricacidsolutionfollowedbythe additionofaqueoussolutionofammonia.Theprecipitatewasthen driedandcalcinatedinournewlydevelopedcalcinationprocess usingrapidheating(∼60^◦C/min)andshortexposuretothehot environmentat400^◦Cfor10min[25].

Irondopedsample (TiO₂-Fe)wasprepared by theoxidative hydrolysisoftitanium(III)chlorideinthepresenceofiron(III)chlo- rideunderair[27,44].Thissamplewasalsocalcinatedapplying

rapidheatingandshortexposuretothehotenvironment(at600^◦C for10min).

SilverandgoldnanoparticleswerephotodepositedunderUV irradiation fromsilver acetate and HAuCl₄ solutions on Aerox- ideP25toproduceTiO2-P25-AgandTiO2-P25-Auphotocatalysts [44,45].Thesetwosamplescontained1wt%ofthenoblemetals.

AeroxideP25wasalsoco-dopedwithsulphurandnitrogenby calcinationwiththioureafor1h,at400^◦Cinstaticair(TiO2-P25- NS)similarlylikeinthemethodappliedbyRengifo-Herreraetal.

[22].

Allofthetitanias(includingthecommercialtitaniumdioxides) werewashedthreetimesbycentrifugationin0.1mMNaCl(Spek- trum3D,99.0%)aqueoussolution,resuspendedinMilliQwater, thendriedat80^◦Cfor24handregroundinagatemortarbefore thephotocatalyticexperiments.

2.3. Methodsandinstrumentation 2.3.1. XRD

ARigakudiffractometerwasappliedforX-raydiffraction(XRD) measurements (Cu K␣=0.15406nm, 30kV, and 15mA,in the 20–40^◦(2)regime).Theaveragediametersoftheparticleswere derivedusingtheScherrerequation.Theweightfractionofanatase andrutilewascalculatedfromthepeakareasoftheanataseand rutilepeaksat25.3^◦(2)and27.5^◦(2),respectively.

2.3.2. DRS

TheDRspectraofthesamples(=220–800nm)weremeasured byaJASCO-V650diodearraycomputercontrolled(SpectraManager Software)spectrophotometerwithanintegrationsphere(ILV-724).

2.3.3. TEM

TEMmicrographswererecordedonaPhilipsCM10instrument operatingat100kVusingFormvarcoatedcoppergrids.

2.3.4. BET

Thespeciﬁcsurface areaof thecatalystswasdeterminedby nitrogenadsorptionat77KusingaMicromeriticsgasadsorption analyser(GeminiType2375).Thespeciﬁcsurfaceareawascalcu- latedusingtheBETmethod.

2.3.5. X-rayphotoelectronspectroscopy(XPS)

X-rayphotoelectronspectraweretakenwithaSPECSinstru- mentequippedwithaPHOIBOS150MCD9hemisphericalelectron energyanalyseroperatedintheFATmode.Furtherdetailsofthe measurementsaredescribedelsewhere[46].

2.3.6. X-rayﬂuorescencespectroscopy(XRF)

AHoribaJobinYvonXGT-5000X-rayﬂuorescentspectrometer, equippedwithRhX-raysourcewasusedtomeasuretheelement contentofthesamples.Therecordsweremadeat30kVexcitation voltage,0.5mAanodecurrentand1000smeasuringtime.

2.3.7. ESRspintrappingmeasurements

The ESR measurements were performed at room tempera- ture by using a Bruker ESP300E spectrometer (Bruker BioSpin, Germany),operatingattheX-bandfrequencyandequippedwith astandardrectangularTE₁₀₂cavity.Aftereachilluminationstep, small aliquots of ∼20␮L were transferred into 0.7mm ID and 0.87mmODglasscapillarytubes(VitroCom,NJ,USA).Tomaximize thesamplevolumeintheactivezone oftheESR cavity,assem- bliesofseventightlypackedcapillarieswerebundledtogetherand insertedintoa wide-bore quartzcapillary (standardESRquartz tubewith2.9mmIDand 4mmOD,Model707-SQ-250M,from Wilmad-LabGlassInc.,Vineland,NJ,USA).Suchsetupresultedin ca.140␮LsamplevolumeintheactivezoneoftheTE102cavity,

(3)

Fig.1. Photographofusedphotoreactorsystemequippedwithconventional24W energysavingcompactﬂuorescencelamps.

which,togetherwiththedivisionoftheaqueoussampleintoseven physically-separatedvolumes,markedlyimprovedtheoverallsen- sitivityofmeasurements[47,48].Thetypicalinstrumentalsettings were:microwavefrequency9.77GHz,microwavepower10.1mW, sweepwidth100G,modulationfrequency 100kHz,modulation amplitude0.5G,receivergain2×10⁴,timeconstant81.92ms,con- versiontime40.96msandtotalscantime41.9s.

2.4. Measurementsofphotocatalyticactivity 2.4.1. Experimentalconditionsforphenoldegradation

Theexperimentsofphenol(Spektrum3D,99.0%)degradation werecarried out in a special photoreactorwhich was anopen glass vesselwithdouble walls, surroundedby a thermostating jacketat 25.0^◦C. Aroundthereactor fourcompactfluorescence lamps(DÜWI25920/R7S-24Wtype–conventional 24Wenergy savingcompactfluorescencelamps)weremounted(Fig.1).The spectrum of thelamp wasslightly modified bycirculating 1M NaNO₂ (MolarChemicals, min.99.13%)aqueoussolution inthe thermostatingjacket. This cut-off solution absorbs UV photons below400nm,providingvisiblelightirradiationforthesamples (Fig. 2).Theradiation intensitywas determinedby ferrioxalate actinometryfortheVISlampswithNaNO2cut-offfiltersolution asI_VIS,1=1.07±0.03×10⁻⁵einstein/dm³/sinthephotoreactor.(It shouldbenotedthattheaveragequantumyieldofferrioxalateacti- nometryisabout0.9inthewavelengthrangebetween400and 540nm,anditisnearlyzeroabove540nm[49].)

Theefﬁciency ofdifferent type ofphotocatalysts was deter- minedbydecomposingphenol(0.1mM)inNaClsolution(0.9wt%) whichcontainedthetitaniapowdersin1.0g/Lconcentration.The suspension(100mL)wassonicatedbeforethephotocatalytictests (for5min) then it was stirredby magnetic stirrerand air was bubbledduringtheexperiments.Changesinphenolconcentration werefollowedusinganAgilent1100seriesHPLCsystemequipped withLichrospherRP18columnapplyingmethanol/watermixture aseluent(thedetectionwascarriedoutat210nm).

2.4.2. ExperimentalconditionsforinactivationofE.coli

Thesamephotoreactorandsimilarconditionswereappliedlike forthephenoldecompositionmeasurements.Anothertypeoflight ﬁltrationwasalsocarriedoutinsomeexperiments.Applying5mM

0 5000 10000 15000 20000 25000 30000

300 400 500 600 700 800

Wavelength (nm)

Intensity (a.u.)

1M NaNO2 filtration

5mM K2Cr2O7 filtration

0 1000 2000 3000 4000 5000

400 402404 406408 410

Fig.2.Spectraoftheappliedconventionalenergysavingcompactﬂuorescence lamps(24W)with1MNaNO2and5mMK2Cr2O7lightcut-offﬁltrationsbythe recirculationofthesesolutionsinthethermostatingjacket.

K2Cr2O7(Reanal,analyticalgrade)aqueoussolutioninthethermo- statingjacket,thelightintensitywasreducedto4%below420nm Fig.2.Theradiationintensitywasdeterminedbyferrioxalateacti- nometryfortheVISlampswiththeK2Cr2O7cut-offﬁltersolution asI_VIS,2=1.75±0.01×10⁻⁶einstein/dm³/sinthephotoreactor.

TheE.colisuspensionswerepreparedusingthefollowingpro- cedure.Firstly,E.coliculturesweregrownfor24hin0.9%NaCl solutionsupplementedwithnutrients:1%Tripton (Reanal,analyticalgrade)and0.5%yeastextract(Scharlau,analyticalgrade).

Thenthecultureswerewashedtwotimeswitha0.9%salinesolu- tionbycentrifugationat4000rpmfor2minandthesedimentwas re-suspendedin0.9%NaClsolution.Priortothedisinfectionexper- iments,theTiO₂-basedphotocatalystswereaddedtoa0.9%NaCl solutionsandsonicatedtoformhomogeneoussuspensions,which werethensupplementedwiththepreviouslypreparedE.colisus- pensions.

Alowvalueofinitialcolonyformingunit(10⁴CFU/mL)wasset inthetitaniumdioxidesuspensionsinordertodetermineverylow disinfectionefﬁcienciesaswell.Sampleswereplatedonagargels duringtheexperimentsandthecolonieswerecountedafter24hof incubationat37^◦Cindark.Allofthesampleswereplatedon2–3 agargelsforgettingreliabledata.Allofthepresentedresultswere takenfromtheaverageoftwoparallelexperiments(resultswere fairlyreproducible).

2.4.3. ExperimentalconditionsforESRmeasurements

IntheESRmeasurementsweusedtwo(ROS) scavengers,i.e.

10mMconcentrationof2,2,6,6-tetramethyl-4-piperidinol(TMP- OH)and50mMconcentrationof5,5-dimethyl-1-pyrrolineN-oxide (DMPO).ThesolutionsofROSscavengerswerepreparedeitherin H2Oordeuteratedwater(D2O).TMP-OH,andD2O(isotopicpurity of99.9at%D)werepurchasedfromSigma–Aldrich(Switzerland) andusedasreceived.DMPOwasalsoobtainedfromSigma–Aldrich (Switzerland).BeforetheESRmeasurementsthesuspensionswere sonicatedfor5minandthen,duringthesubsequentphotocatalytic experimentsunderexposuretovisiblelight,theywerevigorously stirredbymagneticstirrerandairwasbubbledthrough.Theaque- oussuspensionscontainingthetitaniumdioxidenanoparticlesand ROS scavengers were exposed to the visible light in the same photoreactor in which phenol decomposition and disinfection experimentswerecarried out.Similarly,thesame conventional (24W)energysavingcompactﬂuorescencelampsand1MNaNO₂ solutioninthethermostatingjacketforlightcut-offﬁltrationwere

(4)

0.0 0.2 0.4 0.6 0.8 1.0 1.2

200 300 400 500 600 700 800

Absorbance

TiO2-N TiO2-TP-S201

TiO2-Fe

TiO2-VLP7000 0.0

0.2 0.4 0.6 0.8 1.0 1.2

200 300 400 600 700 800

(nm)

Absorbance

TiO2-P25-Au TiO2-P25-Ag TiO2-P25-NS TiO2-P25

0.0 0.2 0.4 0.6 0.8 1.0

300 350 400 450 500

TiO2-AA TiO2-AR TiO2-P25

TiO2-P25

TiO2-P25-NS TiO2-P25-Ag

TiO2-P25-Au

a

b

Fig.3.Diffusereﬂectancespectraofinvestigatedphotocatalysts:(a)non-doped commercialandP25basedmodiﬁedhome-madetitaniumdioxides;(b)dopedcom- mercialandhome-madetitanias.

used.Theconcentrationofthetitaniumdioxidesuspensionswas of1.0g/L,andthetotalvolumewasof20mL.

3. Resultsanddiscussion

3.1. Characterizationofthephotocatalysts

The photocatalysts were characterized by several methods, including: XRD,transmission electron microscopy (TEM),X-ray fluorescencespectroscopy(XRF),diffusereflectancespectroscopy (DRS),X-rayphotoelectronspectroscopy(XPS)and nitrogengas adsorptionanalysis(BET)method.Themostimportantphysico- chemical parameters of the studied photocatalysts, i.e. anatase andrutilecontent,theparticlesize,thedopantcontent(derived fromXRF,and/orXPS)aswellastheBETspecificsurfacearea,are showninTable1.Thereareinvestigatedtitaniaswhichcontain onlyanatasephase(suchasTiO2-AAand TiO2-VLP7000),inthe P25basedtitaniasmorethan90%oftheparticlesareinanatase phase,inthenitrogendopedtitania5%brookitecontentwasdeter- minedbyXRD,whileAldrichrutilecontains96%rutile.Theparticle sizesareverydifferent:TiO₂-VLP7000hasthesmallestparticles (D_XRD=7.8nm)withthelargestspecificsurfacearea(297m²/g), whileAldrichrutilecontainslargeparticles(100–700nm),which resultsinverylowspecificsurfacearea(2.7m²/g).Thelightabsorp- tionwasmeasuredbydiffusereflectancespectroscopyforallof thephotocatalysts(Fig.3aandb).Thesefiguresshowthatdoped

Fig.4. TEMimagesof(a)Aldrichrutile,(b)SumitomoTP-S201and(c)Kronos VLP7000titanias.TEMimagesofotherinvestigatedphotocatalystsarepublished elsewhere[44].

titaniasabsorbvisiblelight.Aldrichanatasedoesnotabsorbany photonsbeyond400nm,howeverAldrichrutilehasabandgap at about420nm. These titaniaswere allexamined withtrans- missionelectronmicroscopytoobservethesizedistributionand themorphologyoftheparticles.TheTEMimagesofTiO₂-TP-S201, TiO₂-VLP7000and TiO₂-ARareillustratedin Fig.4,while other imagesofthephotocatalystsarepublishedelsewhere[44].Aldrich rutilesamplecontainsrelativelylargeparticleswhicharelarger than100nm(D_TEM≈315nm)whichresultsinverylowspeciﬁc surfacearea.Theparticlediametervaluescalculatedfromtheline

(5)

Table1

Structuralparametersoftheinvestigatedphotocatalysts(phasecontent,particlesize,theconcentrationofdopantsandspeciﬁcsurfacearea),theinitialdegradationratesof phenol,irradiationtimefortotalsterilizationandtheresultsofESRmeasurementsareshown.Thesamplesarelistedintheorderofdecreasingactivityforphenoldegradation formthetoptothebottom.Greyrowsindicatethehome-madetitaniumdioxides,thewhiterowsareforcommercialtitanias.

Sample Anatase

(wt%) Rutile (wt%)

DA(nm) DR(nm) Dopantcontent (at%)^f

a^S_BET(m²/g) r0,phenol

(×10⁻⁸M/s)

tsterilization

(min)^g

ESRMeasurements

WithTMP-OH scavenger

WithDMPO scavenger

1O2 O2•- OH^•

TiO2-VLP7000 100 – 7.8 – S:0.33^c/N:1.21^d 297 29.9 – High No No

TiO2-AR 4 96 ∼315^b – 2.7 4.2 20 No No High

TiO2-P25-NS 94 6 25.4 ∼40 S:0.13^c/N:n.d. 55 3.7 – Notmeasured

TiO2-N 95^a – 6.5 – 1.32^d 139 2.4 – No No No

TiO2-Fe 29 71 ∼35 ∼31 0.37^c 28 1.7 – Notmeasured

TiO2-TP-S201 100 – 17.3 – N:0.82^d 80 1.5 60 Notmeasured

TiO2-P25 90 10 25.4 ∼40 – 49 1.4 60 No No Yes

TiO2-P25-Ag 90 10 24.5 ∼42 0.24^c,e 51 1.3 60 Notmeasured

TiO2-P25-Au 90 10 24.1 ∼37 0.13^c,e 51 0.4 – Notmeasured

TiO2-AA 100 – >85 – – 9.9 0.4 – Notmeasured

aCalculatedcontentofbrookiteis5wt%,DB=14.4nm.

b AverageparticlediameterwascalculatedfromTEMpictures.

c MeasuredbyXRF.

d EstimatedbyXPS.

eNominal(added)metalcontentis1.0wt%asmD/mcatalyst×100;measuredvaluesforAgandAucontentsare0.94wt%and0.96wt%,respectively.

f ExpressedasnD/ntotal×100.

gCFU=10⁴;1MNaNO2lightcut-offﬁltration(>400nm).

broadeningoftheanatasepeakforTiO2-TP-S201(DXRD=17.3nm) andTiO₂-VLP7000(D_XRD=7.8nm)samplesareingoodagreement withthenanoparticlesizesobservedontheirTEMimages(Table1, Fig.4bandc).

3.2. Phenoldegradation

ThedecompositionofphenolwasfollowedbyHPLC.Thedecay curvesarepresentedinFig.5aandb.Therewasnotanydecrease intheconcentrationunderirradiationwithoutphotocatalyst,and only very slow degradations were observed applying Aldrich anataseorTiO₂-P25-Au.Itmeansthatinthelatercasedepositing Aunanoparticlesontitaniumdioxidereducesthephenoldegrada- tionefﬁciencyinthevisiblelightrange.TheTiO₂-TP-S201andthe TiO₂-P25-Agsamplesshowedsimilarperformancethanthewell knownreferenceAeroxideP25(TiO2-P25),whichdecomposed17%

ofthephenolfromthe0.1mMsolutionafter4hofirradiation.Three home-madetitaniasshowedhigherefﬁciencythanTiO₂-P25.Only commerciallyavailableKronosVLP7000(co-dopedbynitrogenand sulphur)hadbetterperformancethanthenon-dopedAldrichrutile (TiO₂-AR).Itshouldbenotedthatrutileabsorbslightbetween400 and420nm(incontrastwiththeanatase),inwhichwavelength rangetherearetwopeaksinthespectrumofthelamp(Fig.2).

Nevertheless,themostactivephotocatalyst(TiO₂-VLP7000)con- tains100% anatasephase. The highperformance of this titania wasresultedbytheefﬁcientdopingprovidingtheabilitytoacti- vatetheparticlesbyvisiblelight.Thedecaycurves,andalsothe initialdegradationratesofphenoldemonstrateaswell,thatTiO₂- VLP7000hasextremelyhighefﬁciencyforphenoldegradation,94%

ofphenolwasdecomposedin4h.KRONOSVLP7000isanitrogen andsulphurco-dopedtitaniumdioxidelikeourhome-madeTiO₂- N,S,but thecommercialtitaniaownsvery highspecificsurface areawhichcanresultinthishighperformancebesidestheefficient doping(KimandChoi[50]publishedthathighspecificsurfacearea isbeneficialforthedegradationofphenol).

3.3. Disinfectionperformance

Investigating Aldrich anatase, TiO₂-P25-Au, TiO₂-Fe, TiO₂-N;

TiO₂-P25-NSandTiO₂-VLP7000titaniumdioxides,therewasnot anydisinfectioneffectobservedafter2hofirradiation.Theresults

0.0 0.2 0.4 0.6 0.8 1.0

0 60 120 180 240

Irradiation time (min) cphenol (10-4 M)

Light - without TiO2 TiO2-P25 TiO2-Fe TiO2-P25-NS TiO2-N TiO2-AR TiO2-VLP7000 0.80

0.85 0.90 0.95 1.00

0 60 120 180 240

Irradiation time (min) cphenol (10-4 M)

Light - without TiO2 TiO2-AA TiO2-P25-Au TiO2-TP-S201 TiO2-P25-Ag TiO2-P25

a

b

Fig.5.(aandb)DecaycurvesofphenolunderVISirradiation(conventional24W energysavingcompactﬂuorescencelampswith1MNaNO2lightcut-offﬁltration (>400nm).

(6)

1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05

0 20 40 60 80 100 120

Irradiation time (min)

CFU (1/mL)

Light without photocatalyst TiO2-P25 in dark TiO2-TP-S201 in dark TiO2-AR in dark TiO2-P25-Ag in dark TiO2-TP-S201 + light TiO2-P25 + light TiO2-P25-Ag + light TiO2-AR + light

> 400 nm λ

Fig.6. Disinfectionexperimentswith1MNaNO2ﬁlteredvisiblelightirradiation (initialcolonyformingunitwas10⁴).

oftheﬁveotherphotocatalystsactivitiesarepresentedinFig.6.All oftheexperimentswererepeatedtwotimes,andtheaveragesare showed.ApplyingTiO2-TP-S201,TiO2-P25andTiO2-P25-Agtita- nias,thecolonyformingunitwasreducedtozeroin1h.Fromthe shapeofthemortalitycurve,itseemsthatthesilvercontaining samplehasalittlebithigherdisinfectionabilitythanthesilverfree P25.Itismostlikelyduetothewellknowndisinfectioneffectof silverions[51–57]whichcanbereleasedfromthesilvernanopar- ticlestothesolutionin lowquantity(inthedarkexperiment a slowdisinfectioneffectwasalsodeterminedonthisphotocatalyst).

TiO₂-ARsamplehadmuchhigheractivityforinactivationofE.coli.

Thewaterwassterilizedtotallybythistitaniain20min(Fig.6).

A seriesof experimentswasalso carried outtoexcludethe possibilitythatthedetermineddisinfectioneffectswereresulted bytoxiccompoundsdissolvedfromtheirradiatedphotocatalysts.

Suspensionsofthetitanias(withoutbacteriaaddition)wereirra- diatedforthesametimeatwhichtheydisinfectedpreviouslythe solution,andthenthenanoparticleswereseparatedbycentrifuga- tionandﬁltration.Aftertheseparation,10⁴CFU/mLwasadjusted tothissolution,anditwaspouredbackintothephotoreactorand thechangesincolonyformingunitwasfollowedfor2hofirradia- tion.InthecaseofTiO2-P25,TiO2-TP-S201,TiO2-ARtitanias,there wasnotseenanynotableCFUreduction.Onlyinthecaseofsilver containingTiO₂wasnoticedasimilarCFUreductionlikeinTiO₂-Ag suspensionindark.

Investigationofdisinfectionperformancewascarriedoutalso at higher initial CFU (10⁵CFU/mL) for the most active titania (TiO2-AR).Thistitaniatotallysterilizedthewaterafter30minof irradiationinthiscase.

Disinfectionexperimentswerealsocarriedout withanother type oflight cut-off ﬁltration.Applying5mMK₂Cr₂O7 aqueous solutioninthethermostatingjacketthelightintensitywasreduced to4%below420nm(Fig.2).Resultsofdisinfectionexperimentsare presentedinFig.7.Fig.8representsaphotographwhichdemon- stratestheE.colicoloniesofthephotocatalytictestwithAldrich rutile(after24hofthermostatingin37^◦C,indark).ApplyingAerox- ideP25orTiO₂-TP-S201therewasnotseenanydisinfectioneffect after2h.ThesilvercontainingTiO2 sampleshowedalowactiv- ityforkillingbacterialikeindarkcondition(duetothepresence ofAg⁺ ions).Non-dopedAldrichrutileshowedanotableactivity consideringtheverylowlightintensitybelow420nm(onlylight <420nmcanexcitethisphaseoftitaniumdioxide).WithK₂Cr₂O₇ ﬁltrationthreetitaniumdioxides(TiO₂-P25,TiO₂-P25-Ag,TiO₂-TP- S201)losttheirphotocatalyticdisinfectionproperty,howeverwith

1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05

0 20 40 60 80 100 120

Irradiation time (min)

CFU (1/mL)

TiO2-TP-S201 + light TiO2-P25 + light TiO2-P25-Ag + light TiO2-AR + light

λ > 420 nm

Fig.7.Disinfectionexperimentswith5mMK2Cr2O7ﬁlteredvisiblelightirradiation (initialcolonyformingunitwas10⁴).

NaNO2cut-offﬁltration,thesetitaniasdisinfectedcompletelythe water in1h. Thismeansthat theintensityof thecommercially availablelightsources inthewavelengthrangefrom400nmto 420nmiscrucialforefﬁcientindoorair/surfacecleaning.

3.4. Comparisonofphenoldegradationandthedisinfection performance

Theresultsofphenoldecompositionanddisinfectionexperi- mentsaresummarizedinTable1.Thevaluesofinitialdegradation ratesofphenolforthedifferenttitaniasaredecreasingfromthetop tothebottom.TiO2-VLP7000showedextremelyhighefficiencyfor phenoldecomposition,butithadnotanynotabledisinfectionprop- erty.Aldrichrutilehadhighefficiencytokillbacteriaandalsohad highefficiencyforphenoldegradation.Therewerethreetitanias whichdegradedphenolwithhighefficiencybutdidnotshowany antibacterialproperty.Moreover,therewerethreeothertitanias whichkilledE.coliafter1hofirradiation,howeverthesetitanium dioxideshadrelativelylowperformanceforphenoldegradation.It seemsfromTable1,thereisnocorrelationbetweenthebacteria killingabilityofcatalystandtheircrystallinestructure.Interpre- tationfortheseinterestingobservationsbasedontheparticlesize oronthespecificsurfaceareacannotbeprovided.Togetsome

Fig.8. PhotographofE.colicoloniesofthephotocatalytictest(>420nminitial CFU=10⁵)withAldrichrutile(after24hthermostatingin37^◦C,indark).

(7)

-1.0E-03 -5.0E-04 0.0E+00 5.0E-04 1.0E-03

3440 3450 3460 3470 3480 3490 3500 3510 3520 3530 Magnetic Field (G)

ESR signal (a.u.)

TiO2-VLP7000 - TMP-OH - D2O - 30 min Irradiation TiO2-VLP7000 - TMP-OH - D2O - NaN3 - 0 min Irradiation TiO2-VLP7000 - TMP-OH - D2O - NaN3 - 30 min Irradiation

-1.0E-03 -5.0E-04 0.0E+00 5.0E-04 1.0E-03

3440 3450 3460 3470 3480 3490 3500 3510 3520 3530 Magnetic Field (G)

ESR signal (a.u.)

TiO2-VLP7000 - TMP-OH - H2O - 60 min Irradiation TiO2-VLP7000 - TMP-OH - D2O - 60 min Irradiation

-1.5E-04 -5.0E-05 5.0E-05 1.5E-04

3440 3450 3460 3470 3480 3490 3500 3510 3520 3530

Magnetic Field (G)

ESR signal (a.u.)

TiO2-AR - DMPO - H2O - 3 min Irradiation TiO2-VLP7000 - DMPO - H2O - 0 min Irradiation TiO2-VLP7000 - DMPO - H2O - 3 min Irradiation

-1.0E-03 -5.0E-04 0.0E+00 5.0E-04 1.0E-03

3440 3450 3460 3470 3480 3490 3500 3510 3520 3530

Magnetic Field (G)

ESR signal (a.u.)

TiO2-AR - TMP-OH - H2O - 60 min Irradiation TiO2-AR - TMP-OH - D2O - 60 min Irradiation

a b

d c

Fig.9. ResultsofESRmeasurements:(a)visiblelightilluminatedaqueous(H2OandD2O)suspensionsofTiO2-ARcontaining10mMconcentrationofTMP-OH;(b)visiblelight illuminatedaqueoussuspensionsofTiO2-ARandTiO2-VLP7000containing50mMconcentrationofDMPO;(c)visiblelightilluminatedaqueous(H2OandD2O)TiO2-VLP7000 suspensionscontaining10mMconcentrationofTMP-OH;(d)visiblelightilluminatedD2OsuspensionofTiO2-VLP7000containing10mMTMP-OH(inthepresenceand absenceofNaN3singletoxygenquencher).

insightintoROS-mediatedbacteriainactivationmechanisms,ESR spin-trappingmeasurementswerecarriedoutforselectedphoto- catalysts.

3.4.1. ESRmeasurements

Four selected titanium dioxides were investigated. TiO2- VLP7000and TiO₂-N showed rapidphenol degradation,but no E.colideactivation.TiO₂-ARhadhighefﬁciencyforphenoldegrada- tionandfordisinfectionaswell,whileTiO2-P25hadantibacterial propertybutithadrelativelylowperformanceforphenoldegra- dation.TheresultsoftheESRmeasurementsaresummarizedin Table1.

3.4.1.1. TiO₂-AR. InthepresenceofTMP-OHscavengerintheVIS irradiatedTiO₂-ARsuspensionastrongESRsignalofTEMPOLwas observedafter1hofillumination(Fig.9a).TEMPOLmightbepro- ducedbyanattackofsingletoxygen(¹O₂)orhydroxylradical(OH^•) onTMP-OH[58].

TheESRsignalofTEMPOLisslightlydistorted.Asitisseenin Fig.9athecentralESRfeatureismarkedlyhigherthanthelow-and high-ﬁeldfeatures.Thisdistortionoriginatesfromtheformationof anothernitroxideradical,TEMPONE.OH^•radicalscanreactwith TEMPOLbyattackingontheOH-groupswhichyieldthegeneration ofTEMPONE[59].

InD2O,asitcanbeseeninFig.9a,thesignalamplitudeofthe ESRsignalofTEMPOLwasreducedca.7.3timesascomparedtoH₂O (forthesameilluminationtime).ItisknownthatD₂Osuppresses theformationofOD^•radicalsattheTiO2/D2Ointerface[60,61].

InthepresenceofDMPOanintenseDMPO-OH^•signal(inH₂O) wasobservedundervisiblelightirradiationwhichprovedtheOH^• radicalgenerationinthecaseofAR(Fig.9b).Thisresultcorrobo- rateswiththemarkeddiminishmentoftheTEMPOLsignalinD₂O, andsuggeststhatthephotocatalyticactivityofthissamplemightbe relatedtothephotocatalyticgenerationofhydroxylradicals(OH^•).

ItshouldbenotedthatTiO₂-ARprobablyproducessome¹O₂via theclassicalmechanismsuggestedbyNosakaetal.[62]butsince applying DMPO scavenger no DMPO-OOHsignal was detected, TiO₂-ARdonotproduceO₂^•−radicalinmeasurablequantity.

3.4.1.2. TiO₂-VLP7000.Under VISlight illumination in H₂O,the TiO2-VLP7000samplegaveastrongTEMPOLsignal.Thissignalis evenstronger(byca.10%)thanthesignaloftheTiO2-AR(Fig.9c).

InD₂O,theESRsignalamplitudeofTEMPOLincreasedca.1.4 times as compared to H₂O (Fig. 9c) which suggested that the detectedsignalcanbeduetothegenerationofsingletoxygen(¹O2) onthesurfaceofTiO₂-VLP7000.Togetmoreevidenceofthegen- erationof¹O₂ anexperimentwithsingletoxygenquencherwas alsocarriedout.InD2OaverystrongsuppressionoftheTEMPOL signalwasobserved(itwasreducedca.19times)inthepresence of100mMNaN₃(whichisa¹O₂quencher)(seeFig.9d).

SoboththeESRsignalenhancementinD2Oandsignalquench- ingbyNaN₃suggestedthatthestrongsignalofTEMPOLisdueto theformationof¹O₂.Howevertheseresultsdonotexcludethepos- sibilitythatthescavengerwasdirectlyoxidizedonthesurfaceof thephotocatalyst,assuggestedbyNosaka(in2006)forothertita- niaphotocatalysts[63],consideringthatthistitaniahasextremely highspeciﬁcsurfacearea.

(8)

InthepresenceofDMPO scavengernoDMPO-OHor DMPO- OOHsignalwasdeterminedforVISlightirradiatedTiO2-VLP7000 (Fig.9b)whichprovedthatthereisnoOH^•radicalgenerationon thistitania.Thisobservationcorroborateswiththeresultsofthe experimentswithTMP-OHscavenger.

3.4.1.3. TiO₂-P25. AtthecaseofTMP-OHscavengerinthevisible lightilluminatedTiO2-P25suspensionawellmeasurablesignalof TEMPOLwasobserved,whichwasabout30%ofthesignalofTiO₂- AR.ThissignalwasstronglydecreasedinD₂Olikeinthecaseof TiO2-AR,andtheexperimentswithDMPOscavengeralsoproved againthegenerationofOH^•radical(butinsmalleramountsthan inthecaseofTiO₂-AR)andtheabsenceofO₂^•−.

3.4.1.4. TiO2-N.NomarkedformationofOH^•,O2•−orsingletoxygen(¹O₂)wasobservedforthenitrogendopedhome-madetitania (TiO₂-N).

4. Conclusions

Undervisiblelightillumination,threeofourhome-madetitanias(TiO₂-Fe,TiO₂-N,S,TiO₂-N,)andcommercialnitrogen/sulphur- dopedKronosVLP7000titaniumdioxideshowedhigherefﬁciency forphenoldegradationthanthewellknownreferenceAeroxide P25, however onlyVLP7000 had betterperformance than non- dopedAldrichrutile.

AldrichrutilehadhighefficiencyofE.coliinactivationandalso hadsignificantactivityforphenoldegradation;howeverfourdoped titaniashadhigherefficiencyforphenoldegradationthanTiO₂-P25 buthadnotanyactivityforinactivationofbacteria.

ESR measurements pointed out that titanias that generated OH^•radicalswereactiveforkillingE.coli,andthose,whichwere unabletoproduceOH^•radicalsdidnotshowanydisinfectionprop- erty.Thisisconsistentwiththestatementsofmanyauthorswho described that OH^• radicals play a majorrole in photocatalytic disinfectionexperiments[2,5,6,24,34,39,42,43],andrecentstudy provedthatthispersistinthecaseofvisiblelightirradiationas well.ThisisingoodagreementwiththeresultsofthestudyofYu etal.[34]whodemonstratedthedeactivationofMicrococcuslylae byvisiblelightirradiatedTiO2duetothegenerationofhydroxyl radical.

KronosVLP7000showedextremelyhighefﬁciencyforphenol decompositionduetoitshighspeciﬁcsurfacearea,butthistitania didnotshowanydisinfectionpropertyunderourcircumstances whichmightbeduetoabsenceofOH^•radicalgeneration.

TiO2-N titania showedhighperformance for phenol decomposition,howeverneitherOH^•radicalnorsuperoxideradicalion (O₂^•−)orsingletoxygen(¹O₂)wasdetected.Thismeansthatphe- noldegradationoccurredviathereactionbytheholeonthesurface, andthehighefficiencyismostlikelyduetothehighspecificsurface area(thistitaniumdioxidehasthesecondhighestspecificsurface areaamongtheinvestigatedsamples).

The experiments with K2Cr2O7 cut-off filtered light source pointed outthat therutile couldhave high-performanceappli- cability forutilizing visible light in self-cleaningor disinfecting processes.Rutilehadanotableefficiencyalsowiththisverylow intensitybelow420nm.Thisstudyfirstlyinvestigatedthedisin- fectionperformanceofrutileparticleswithrelativelylargeparticle size(d∼315nm)usingsolelyvisiblelightforactivatingthephoto- catalyst.

PureandsilverdopedP25andSumitomotitaniasterilizedwater in1hwhen>400nmconditionwasused,butthesephotocata- lystslosttheirdisinfectionpropertywhenK₂Cr₂O₇ light cut-off ﬁltrationwasapplied(>420nm).Theseresultshighlightedthat theintensityofthecommerciallyavailablelampsinthewavelength

range from400nmto420nm iscrucialtoapplyeffectively the photocatalystsforindoorair/surfacecleaning.

Acknowledgements

This work was partially financed by the European Union throughtheHungary-SerbiaIPACross-borderCo-operationPro- gram,HU-SRB/0901/121/116.Itwasalsoco-financedbythegrant from the Hungarian National Office of Research and Technol- ogy (OTKACK 80193), by theEuropean Regional Development Fund(TÁMOP-4.2.1/B-09/1/KONV-2010-0005andTÁMOP-4.2.2/B- 10/1-2010-0012)andtheSwissContribution(SH/7/2/20).

KM thanksthe ﬁnancial supportof the Hungarian Research Foundation(OTKAPD78378)andtheJánosBolyaiResearchSchol- arshipoftheHungarianAcademyofSciences.

A.Sand L.F. acknowledgethe ﬁnancialsupport of theSwiss NationalScienceFoundationthroughtheNano-TeraNTFproject

“Core-shellsuperparamagneticandupconvertingnano-engineered materialsforbiomedicalapplications–NanoUp”.

TheauthorsareindebtedtoEvonikIndustries,toKronosGmbh., toSumitomoChemicalsInc.forsupportingourworkbysupplying freeTiO₂samplesforthesestudies.

References

[1] T.Matsunaga,R.Tomoda,T.Nakajima,H.Wake,FEMSMicrobiologyLetters29 (1985)211–214.

[2]J.C.Ireland,P.Klostermann,E.W.Rive,R.M.Clark,AppliedandEnvironmental Microbiology59(1993)1668–1670.

[3] J.A.Herrera-Melian,J.M.D.Rodriguez,A.V.Suarez,E.T.Rendon,C.V.D.Campo,J.

Arana,J.P.Pena,Chemosphere41(2000)323–327.

[4]K.Kühn,Chemosphere53(2003)71–77.

[5]M.Cho,H.Chung,W.Choi,J.Yoon,WaterResearch38(2004)1069–1077.

[6] A.Vohra,D.Y.Goswami,D.A.Deshpande,S.S.Block,AppliedCatalysisB64 (2006)57–65.

[7]A.K.Benabbou,Z.Derriche,C.Felix,P.Lejeune,C.Guillard,AppliedCatalysisB 76(2007)257–263.

[8] C.Hu,J.Guo,J.Qu,X.Hu,Langmuir23(2007)4982–4987.

[9] C.Guillard,T.Bui,C.Felix,V.Moules,B.Lina,P.Lejeune,ComptesRendusChimie 11(2008)107–113.

[10]D.M.A.Alrousan,P.S.M.Dunlop,T.A.McMurray,J.A.Byrne,WaterResearch43 (2009)47–54.

[11] L.Caballero,K.A.Whitehead,N.S.Allen,J.Verran,JournalofPhotochemistry andPhotobiologyA:Chemistry202(2009)92–98.

[12]F.Chen,X.Yang,Q. Wu,EnvironmentalScience &Technology43(2009) 4606–4611.

[13]J.A.Rengifo-Herrera,K.Pierzchała,A.Sienkiewicz,L.Forró,J.Kiwi,C.Pulgarin, AppliedCatalysisB88(2009)398–406.

[14]E.A.Kozlova,A.S. Safatov,S.A.Kiselev,V.Y.Marchenko,A.A.Sergeev,M.O.

Skarnovich,E.K.Emelyanova,M.A.Smetannikova,G.A.Buryak,A.V.Vorontsov, EnvironmentalScience&Technology44(2010)5121–5126.

[15]S.Malato,P.Fernández-Ibá ˜nez,M.I.Maldonado,J.Blanco,W.Gernjak,Catalysis Today147(2009)1–59.

[16]A.Vidal,A.I.Dıaz,A.E.Hraiki,M.Romero,I.Muguruza,F.Senhaji,J.González, CatalysisToday54(1999)283–290.

[17]E.Duffy,SolarEnergy77(2004)649–655.

[18]C.Sichel,J.Blanco,S.Malato,P.Fernández-Ibá ˜nez,JournalofPhotochemistry andPhotobiologyA:Chemistry189(2007)239–246.

[19]C.Sichel,J.Tello,M.deCara,P.Fernández-Ibá ˜nez,CatalysisToday129(2007) 152–160.

[20]A.I.Gomes,J.C.Santos,V.J.P.Vilar,R.A.R.Boaventura,AppliedCatalysisB88 (2009)283–291.

[21]C.Karunakaran, G.Abiramasundari,P.Gomathisankar, G. Manikandan,V.

Anandi,JournalofColloidandInterfaceScience352(2010)68–74.

[22]J.A.Rengifo-Herrera,J.Kiwi,C.Pulgarin,JournalofPhotochemistryandPhoto- biologyA:Chemistry205(2009)109–115.

[23] P.Wu,J.A.Imlay,J.K.Shang,Biomaterials31(2010)7526–7533.

[24]P.Wu,R.Xie,J.A.Imlay,J.K.Shang,AppliedCatalysisB88(2009)576–581.

[25]Z.Pap,L.Baia,K.Mogyorosi,A.Dombi,A.Oszko,V.Danciu,CatalysisCommu- nications17(2011)1–7.

[26]K.Nagaveni,G.Sivalingam,M.S.Hegde,G.Madras,EnvironmentalScience&

Technology38(2004)1600–1604.

[27]Z.Ambrus,N.Balazs,T.Alapi,G.Wittmann,P.Sipos,A.Dombi,K.Mogyorosi, AppliedCatalysisB81(2008)27–37.

[28] E.G.Bajnoczi,N.Balazs,K.Mogyorosi,D.F.Sranko,Z.Pap,Z.Ambrus,S.E.Canton, K.Noren,E.Kuzmann,A.Vertes,Z.Homonnay,A.Oszko,I.Palinko,P.Sipos, AppliedCatalysisB103(2011)232–239.