optical by using response mesoporous the silica foam Room temperature sensor with ethanol sub-ppm detection limit:Improving Sensors and Actuators B: Chemical

(1)

SensorsandActuatorsB243(2017)1205–1213

Contents lists available atScienceDirect

Sensors and Actuators B: Chemical

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

Room temperature ethanol sensor with sub-ppm detection limit:

Improving the optical response by using mesoporous silica foam

Dániel Seb ˝ok

^a,∗

, László Janovák

^a

, Dániel Kovács

^a

, András Sápi

^b

, Dorina G. Dobó

^b,c

, Ákos Kukovecz

^b,c

, Zoltán Kónya

^b,d

, Imre Dékány

^a,e

aDepartmentofPhysicalChemistryandMaterialsScience,UniversityofSzeged,1Rerrichsquare,H-6720Szeged,Hungary

bDepartmentofAppliedandEnvironmentalChemistry,UniversityofSzeged,1Rerrichsquare,H-6720Szeged,Hungary

cMTA-SZTE“Lendület”PorousNanocompositesResearchGroup,UniversityofSzeged,1Rerrichsquare,H-6720Szeged,Hungary

dMTA-SZTEReactionKineticsandSurfaceChemistryResearchGroup,UniversityofSzeged,1Rerrichsquare,H-6720Szeged,Hungary

eMTA-SZTESupramolecularandNanostructuredMaterialsResearchGroup,UniversityofSzeged,8Dómsquare,H-6720Szeged,Hungary

a r t i c l e i n f o

Articlehistory:

Received25August2016 Receivedinrevisedform 17December2016 Accepted19December2016 Availableonline21December2016

Keywords:

Thinﬁlm RIfS

Mesoporoussilica Roomtemperature Sub-ppm Ethanolsensor

a b s t r a c t

Inthispaper,theimprovementinroomtemperatureethanolsensingcharacteristicsofzincperoxide (ZnO2)basedhybridthinfilmsispresentedbythecombinationofthebeneficialsensingproperties ofmesoporous materialsandreflectometricinterference spectroscopy(RIfS).Thehybridthin films werepreparedbyLayer-by-Layer(LbL)self-assemblymethodfromZnO2nanoparticles,polyelectrolyte [poly(acrylicacid),PAA]and/ormesoporoussilica(MPS).Theexpectedimprovedsensingpropertieswere attributedtothefractalpropertiesandhighspecificsurfacearea(a^s)ofthemesoporouscoating/interlayer material,whichwasevidencedbysmallangleX-rayscattering(SAXS)andN2sorptionmeasurements (a^s>650m²/g).Thesensortestsshowedthatthedetectionlimitofthethinfilmsisinthesub-ppmrange (<500ppb).Applyingsilicafoam(SF)assurfacecoatingorinterlayermaterialinthesandwich-structured thinfilm(ZnO2/SF)improvedtheopticalresponse(l:wavelengthshift)comparedtotheZnO2/PAAthin layer,butthesensitivityshowednon-linearcharacteristicandsignaldrift.Thethinfilmwithmixedstruc- ture(ZnO2/PAA/ZnO2/SF)showedlinearsensitivity(␭/c=0.6nm/ppm)inthe0.5–12ppmrangewith anacceptableselectivityandstablebaseline.Testingthesensorinextended(upto40ppm)concentration rangeshowedonlyaslightquadraticdeviationfromlinearbehaviorwithR²=0.9987.

1. Introduction

Sensorsforvolatileorganiccompounds(VOCs) (e.g.alcohols, benzeneetc.)playanimportantroleineverydaylifeandindus- trialsafety. Indisputable fact is that thesechemical agents are harmfulandunhealthy,sothedetectionofthesemoleculeshas agreatimportanceinenvironmentalandhealthprotection,such asinairandwaterqualitycontrol,foodindustryor–especiallyin thecaseofethanol–the“drivingunderinﬂuence”(DUI)control.

Considering a comprehensive, although not complete overview ofthearticlespublishedinrecentyearsinethanolsensorstopic (Fig.1)wecanconcludethattheprinciples,technicalsolutions, thematerialsused, theoperating temperatureranges and con- centrationlevelsarefairlydiversiﬁed.Themostcommonlyused

∗Correspondingauthor.

E-mailaddress:sebokd@chem.u-szeged.hu(D.Seb ˝ok).

sensor materialsare SnO₂ [1–3], ZnO [4–7], SiO₂ [8],In₂O₃ [9]

TiO2[10],Fe₂O₃ [11,12]andothernanostructures[13–16],com- posites[17–25]orcoatings[26,27].Themeasurementprincipleis mainlybasedontheresistivemethod,butalsocapacitive[8,15], optical[2,26–28],quartzcrystalmicrobalance(QCM)[26,28]and piezo(self-poweringdevice)[7,21,22]applicationscanbefound.

Fig.1showstheprinciples,studiedconcentrationrangesandoper- atingtemperaturespresentedintheworkscitedabove.Itcanbe seenthatthestudiescanbedividedintotwomajorgroups:room temperature(RT)andhightemperature(around200and300^◦C) applications.Mainlyelectricalmethodsandmesoporoussensing materialsarepreferredinthelattercase,therebybroadconcen- tration ranges withexcellent detection limits can be achieved.

However,ithastobenotedthatVOCpollutantseasilyevaporate atroomtemperatureand canbeveryharmfulandcarcinogenic alreadyatlow concentration.ItcanbeseenonFig.1thatmost oftheRTtechnicalsolutions[2,7,8,15,16,26–28]areabletodetect ethanol vapour onlyabove 10ppm concentration.In this work, http://dx.doi.org/10.1016/j.snb.2016.12.097

(2)

Fig.1.Diagramsummarizingthecomprehensive,althoughnotcompleteoverviewofthearticlespublishedinrecentyearsinethanolsensorstopic(groupedbythe measurementprinciples,thestudiedconcentrationrangesandoperatingtemperatures;note:roomtemperaturehasnotdetailedscale!).

wemadeanattempttocombinethebeneﬁcialsensingproperties ofmesoporousmaterials[3,29,30]andreﬂectometricinterference technique[31]toconstructahighlysensitiveethanolsensoroper- atingatroomtemperature.

Reflectometricinterferencespectroscopy[28,31–36]isanopti- cal method which is based on the spectral (red) shift of the interferencepatternreflectedfroma(fewhundrednanometersof layerthickness)thinfilm.Thewavelengthshiftiscausedbythe adsorptionoradhesionofmoleculesorcolloidalunits,soitcanbe utilizedinantigen-antibodyreactionsortodetecttheadsorptionof volatilecompounds[].ThesensorsurfaceofRIfStechniquecanbe preparedbyusingthewetcolloidchemicalprocedure,theso-called Layer-by-Layermethod[37,38].TheLbLmethodis widelyused forthinfilmpreparationdirectlyfromcolloidalsystems(nanoparticles, polymer solutions, etc.): it is an easy, non-instrumental techniqueanditresultsahomogeneoussurfaceandwell-ordered, transparentstructurewithcontrollablefilmthicknessandafine andporousmicrostructure[39],likethesimilarLangmuir-Blodgett method[40,41].TheseareessentialconditionsforapplyingRIfS technique,andthesensitivity,aswellas,thelimitofdetectioncan beimprovedbytheadditionofvarioussurface-modifyingagents [28,31].

In the present work we demonstrate the beneficial effect of using mesoporous silica materials on the sensitivity and detection limit of RIfS sensor in the gas phase. We show that applyingmixed(nanoparticle/polyelectrolyte/mesoporoussilica) nanostructureresultslinearsensitivityandsub-ppmethanoldetec- tionlimitwithoutresponsedrift,whileboththeresponsetimeand selectivityremainstableandadequate.Furthermore,firstlyinthis workwecarriedoutreflectionintensitymeasurementinaddition tothewavelengthshiftmonitoring:thetwotypesofresponsesdif- fersignificantly,whichmayhighlight–byfurtherstudies–the

differencesbetweentheadsorptionmechanismsontothevarious surfaces.

2. Experimental 2.1. Materials

Zincperoxidenanoparticleswithanaveragediameterof80nm were synthesized by the photolysis of zinc acetate dehydrate (C4H6O4Zn·2H2O,Fluka,a.r.)describedin[38].Poly(acrylicacid) (PAA,M_W=100000,Sigma,a.r.)wasusedasanegativelycharged polyelectrolyte.Furthermore,SBA-15andsilicafoamwereusedas coatingsornegatively chargedinterlayermaterials.Synthesisof SBA-15silicaiswell-known[42].MesoporousSFwereprepared byamodifiedsol-gelroutebasedonthetechniquesuggestedby Bagshaw[43].Inatypicalsynthesis,13.9gTEOSwasslowlyadded to30mL 10 w% TritonX114 aqueoussolutionand thesynthe- sismixturewerevigorouslystirredfor24h.Theobtainedsilica suspensionwascollectedbyvacuumfiltrationandlefttodryat roomtemperaturefor24h.Thedriedsamplewasintroducedinto aTeflon-linedstainlesssteelautoclavewithavolumeof100mL where 10mLwater was alsoadded separatelyto ensurewater vapourenvironment.Aftertheassemblyoftheautoclave,itwas heldat140^◦Cfor24h.Finally,thesilicafoamwascalcinedinairat 450^◦Cfor4h.

2.2. Thinﬁlmpreparation

Five types of hybrid thin ﬁlms were prepared byusing the ZnO2 nanoparticles,negatively chargedPAApolyelectrolyte and themesoporous silicasamples (see Fig. 2): (1.) 20 zinc peroxide/poly(acrylicacid)bilayers([ZnO₂/PAA]₂₀);(2–3.)[ZnO₂/PAA]₂₀ ﬁlms with silica foam and SBA-15 coatings ([ZnO2/PAA]20 +SF

(3)

D.Seb ˝oketal./SensorsandActuatorsB243(2017)1205–1213 1207

Fig.2. Theschematicviewofthepreparedandappliedhybridthinﬁlms.

and [ZnO2/PAA]20 +SBA); (4.) a thin film containing 20 bilayers of zinc peroxide/silica foam ([ZnO₂/SF]₂₀) and (5.) a thin film containing 10 mixed, zinc peroxide/poly(acrylic acid)/zinc peroxide/silicafoammultilayers([ZnO2/PAA/ZnO2/SF]10).During pre-experiments,SBA-15wasnotsuitableasinterlayermaterial inthinfilmsconsistingof40layers,therefore,itwasonlyapplied asacoating.ThethinfilmswerepreparedbytheLbLdeposition methodbythealternatedadhesion/adsorptionofZnO2nanoparti- cles,poly(acrylicacid)[44]andmesoporoussilicananostructures (SBA-15andSF)onthesurfaceofglasssubstrate(microscopeslides, MarienfeldSuperior,Germany).Thethinfilmpreparationwascar- riedoutbyusingc=8g/LZnO₂,c=0.1g/LPAAandc=10g/Lsilica solutions.Theimmersiontimewas10minforeachstep,whichwas followedbyrinsingwithdeionizedwatertoremovethesurplus (non-electrostaticallyattached)colloidunits.Duringthecoating processthepreviouslypreparedzinc-peroxide/poly(acrylicacid) hybridfilmswereimmersedintothesilicasuspensionandwere driedwithoutrinsingthesurplus.

2.3. Instrumentalmethods

TransmissionElectronMicroscopy(TEM)measurementswere carriedoutbyaFEITECNAIG²20X-Twinhigh-resolutiontrans- missionelectronmicroscope(equippedwithelectrondiffraction) operatingatanacceleratingvoltageof200kV.Thesampleswere drop-castontocarbonﬁlmcoatedcoppergridsfromethanolsus- pension.

Thespeciﬁcsurfacearea(BETmethod)andthetotalporevolume weredeterminedbytheBJHmethodusingaQuantachromeNOVA 2200gassorptionanalyzerbyN₂gasadsorption/desorptionat77K.

Beforethemeasurements,thesampleswerepre-treatedinvacuum at200^◦Cfor2h.Thedensityofthesilicapowderswasmeasured usingahelium

gaspycnometer(Micromeriticstype1305).

SAXStechniquewasusedtoinvestigatethefractalproperties andstructuralparametersofthemesoporoussilicacomponents.

SAXScurves were recorded witha slit-collimated Kratky com- pact small-angle system(KCEC/3 Anton-PaarKG,Graz, Austria) equippedwitha position-sensitive detector(PSD50Mfrom M.

Braun AG Munich, Germany) containing 1024 channels 55␮m inwidth.CuK_␣radiation(␭CuK␣=0.1542nm)wasgeneratedbya PhilipsPW1830X-raygeneratoroperatingat40kVand30mA.The fractaldimensionof atwo-phasesystemcanbedeterminedby usingthefollowingequation:I(h)=I₀h⁻^p,whereh=4sin␭⁻¹ isthescatteringvector,isone-halfofthescatteringangle,isthe wavelengthofCuK_␣radiation,I(h)isthescatteringcurve,I₀isthe scatteredintensityath=0,andpistheslopeoftheﬁttedlinein thehigherh-range(Porodregime)inlog-logplotofthescattering curve.If3<p<4thenthesampleissurfacefractal,andthesam- plehasmassfractalpropertiesinthecaseof1<p<3.Thespeciﬁc surfacearea(a^s)valueswerecalculatedbyusingequationsin[45].

The optical properties of the thin films were studied by a Nanocalc2000spectrophotometer withADC1000-USBA/D con- verter (Ocean Optics). The reflectionspectra of the films were measuredina special,home-builttestcellatdetectionangleof 45^◦.Thethickness(d)andeffectiverefractiveindex(n₁)ofthethin filmswerecalculatedbasedonthemodelpresentedinFig.3.aand byusing(andfitting)Eq.(1)(moredetailsin[38]):

R(,ne,d)=c1+c2·cos

_4n

1dcos ε₁

(1) whereε₁ istheangleofrefractionatair/thinﬁlminterface,is thewavelength,c₁andc₂areconstantswhichcontainthet_ijandr_ij transmissionandreﬂectionamplitudescalculatedbyFresnelequa- tions(i,j=0,1,2,seeFig.3.a).

Thesametestcellwasusedduringthesensorialtestsindynamic conditions(Fig.3.b):itwasconnectedtoagasflowsystemwhich consistsofthecarriergas(N2)holder,thetemperaturecontrolled liquidsampleholderandanumberofflowcontrollers(MFC)(Cole- Parmer,USA). The vapourconcentration inthe test cellcan be controlledbytheMFCunitsandV1-V5valvesviathemixingrateof pureandvapourcontainingN₂flows(theflowcontrollerscanreg- ulatemaximum3922,844and49mL/mingasflow).Theaccuracy oftheflowadjustingontheMFCscaleis±0.5division,sothepreci- sionoftheconcentrationis±0.04ppmor±0.68ppminthecaseof themax.49mL/minorthemax.844mL/mindevices,respectively.

Theethanol dosageandrinsing(N2)timeswere3–3min,alter- nately.Thesensorresponses,␭(nm)andR(a.u.)weredefined asthewavelengthshiftofthegivenextremeofthereflectionspec- traandthechangeofthereflectionvaluecorrespondingtothis extreme,respectively.The measurementsineach concentration steps(475-11880ppb)wererepeatedthreetimes,theresponses weredeterminedastheaverageofthethreevalue.Duringselec- tivitymeasurements2␮Lofliquids(methanol,ethanolasalcohols;

n-hexaneasaliphatic;toluene,xyleneasaromaticmolecules)was droppedintotheliquidsampleholder(inthiscaseT=70^◦C)with a mixingrateof4.9mL/minsub-branchand1000mL/minmain branchﬂowrates.

3. Resultsanddiscussion

3.1. Characterizationofthemesoporoussilicamaterials

The porosity,pore systemcharacteristic and speciﬁcsurface areaare of great importancein thecase of mesoporous adsor- bentsusedinsensorialapplications,thereforeseveralstructural parametersweredeterminedandcalculatedbyusingSAXStech- nique.Asitisknownintheliterature,SBA-15has2Dhexagonally orderedporesystem[38],whichcanbeidentiﬁedbyTEMandSAXS technique(Fig.4.a:AandB).Bothmeasurementsclearlyshowthe porestructure;inthelattercasethepeaksath=0.705,1.235and 1.41nm⁻¹correspondtothe1:√3:2ratio,therebytheP6mmsym-

(4)

Fig.3.aTheusedthinfilmmodelforthicknessandrefractiveindexcalculations(indices:0–air,1–thinfilm,2–substrate).bSchemeoftheusedexperimentalsetup(gasflow systemandreflectometrictestcell)andthemeasurementprinciple.

metryisclearlyidentiﬁed[46].Thedoublelogarithmicplotofthe scatteringcurveissuitablefordeterminingthefractalproperties ofthematerial.InthecaseofSBA-15theslopeisapproximately

−2inthehigherh-range,whichindicatesaframe-likesurfacefrac- talstructure,aswellas,p≈−3.5forSFsampleischaracteristicfor thesmoothsurfacefeatureofthemesocellularfoamstructure(see Fig.4.a:C).ThespeciﬁcsurfaceareavaluesdeterminedbySAXS measurementsare820and730m²/gforSBA-15andSF,respectively.

TheN2adsorption/desorptionstudiesofthesilicasamplesshow isothermswithhysteresisloopandporesizedistributionsbetween 3and10nmcharacteristicformesoporousmaterials(Fig.4.b).The specificsurfaceareasandaverageporediametersare798m²g⁻¹, 4.2nmand666m²g⁻¹,4.6nmforSBA-15andSF,respectively.In summary,theSBA-15hashigherspecificsurfacearea,howeverthe SFshowedhighertotalpore volume(Vp=0.76and 1.30cm³g⁻¹ for SBA-15and SF,respectively). The overall conclusionis that althoughthestructural and fractal natureofthese mesoporous materialsaresignificantlydifferent,buttheaverageporediameters andspecificsurfaceareasaresimilar,andthesevaluesappeartobe sufficientlyhighforconsiderableadsorptioncapacityandsensorial applications.

3.2. Thinﬁlmcharacterization

Therecordedandfitted(inthe␭=550–850nmrange)reflection spectraandthecalculatedrefractiveindexcurvesofthreetypesof thinfilms(withoutcoating)canbeseenonFig.5.Thelayerthick- nessesandeffectiverefractiveindices(at␭=589nm)are782nm and1.286,894nmand1.258,989nmand1.251for[ZnO2/PAA]20, [ZnO₂/PAA/ZnO₂/SF]₁₀ and [ZnO₂/SF]₂₀, respectively. It can be establishedthatusingSFsilicaasinterlayermaterialincreasesthe thickness(d)anddecreasestheeffectiverefractiveindex(n1)ofthe thinfilms.Inouropinion,themainreasonsforthisisthefollow- ing:thesetypeofhighlyporoussilicananostructuressignificantly increasethemicro-andmacro-levelporosityofthehybridfilms, therebyconsiderablydecreasetheeffectiverefractiveindex,while thepolyelectrolyteformsultrathinlayersinthemultilayerstruc- ture,therebyensuringacloserpackingfortheZnO₂nanoparticles.

Increasingtheporosityandformingthickerandporousinterlayers inthesandwich-likestructureexplaintheslightlylowerrefractive indices,higherfilmthicknessesandtheadvantageouseffectinsen- sorialtests(presentedlaterin3.3).Inthecaseof[ZnO2/PAA]20+SF and[ZnO₂/PAA]₂₀+SBAfilmsthesilicamonolayerhasnosignifi- cantcontributiontotherefractiveindexandlayerthickness,and giventhefactthatforthesesamplesthecoatinghadnosignificant

(5)

Fig.4.a(A)ArepresentativeTEMimageofthemesoporousSBA-15sample,(B)theSAXScurvesofSBA-15andSFpowdersamplesinlog–logrepresentation(thepower-law exponentsareindicatedbydottedlines)and(C)theTEMimageoftheSFsilicasample.b(A,C)BETisothermsfortheSBA-15andSFsilicasamplesshowthemesoporous characteristic,(B,D)poresizedistributionofSBA-15andSFsamples,respectively.

Fig.5.(A)Themeasured(solidlines)andcalculated(dottedlines)reﬂectionspectraof[ZnO2/PAA]20,[ZnO2/PAA/ZnO2/SF]10and[ZnO2/SF]20thinﬁlms,and(B)thecalculated refractiveindexcurves.

effectinthesensorialtests,thedetaileddiscussionoftheoptical propertiesisignoredinthisparagraph.

3.3. Sensorialtestofthehybridthinﬁlms

Thethin filmsweresubjected to reflectometricinterference measurements for testing sensorial applications. The measure- mentswerecarriedoutbymeasuringtheshiftofthelocalminimum ofreflectedintensitynear␭=500nmwavelength.Itis␭min=457nm in thecase of [ZnO2/PAA]20,and ␭min=507nm and 568nm for [ZnO₂/PAA/ZnO₂/SF]₁₀and[ZnO₂/SF]₂₀,respectively(thesevalues arevalidinthet=0measurement point).Therawresults(sensorgrams),i.e., the vs. t and R vs. t curves are presented in Fig. 6.a and b, respectively. Conspicuous differences can be noticedviewing either thetwo sensorgrams or thecurves one byone.It canbeimmediatelyconcludedthatinthecaseof

curves the responsesare positive becauseof the optical thick- nessincreasesduetothevapouradsorption.Signiﬁcantsignaldrift canbeobservedinthecaseofZnO2/PAA,ZnO2/PAA+coatingand ZnO/SFthinﬁlms,whichisratherdisturbingphenomenon:should bedrift compensationapplied? In thiscase, not in general.f it isignoredtheneach measurementstep(atthesameconcentra- tion)resultsinhigher responsethan thepreviousoneand this factmakesimpossibletoaccuratelydetermineunknownconcen- trations.As it can beseen onthe calibration (responsevs.

concentration)curves(Fig.7.a),wedidnotapplycompensation,the responsesincreasewithconcentration,evenifnotlinearly.How- ever,inthecaseof[ZnO2/PAA/ZnO2/SF]10hybridthinﬁlmresponse driftwasnotobservedwhichresultedalinearcalibrationcurve (0.586nm/ppm).InthecaseofRvs.tcurves(Fig.6.b)positive responsecanbeobservedonlyinthecaseof[ZnO2/PAA]20 sensor,butif thethinﬁlm containsMPS (asinterlayer materialor

(6)

Fig.6. aEthanolsensingtests:vs.tcurvesforthetestedthinﬁlms(labels:

structureofthethinﬁlmandtheethanolconcentrationsteps);inset:responseof [ZnO2/PAA/ZnO2/SF]10mixedstructureforc=475ppbEtOH.bEthanolsensingtests:

Rvs.tcurvesforthetestedthinﬁlms(labels:structureofthethinﬁlmandthe ethanolconcentrationsteps).

coating)thenthereﬂectivitydecreasesduetotheethanoldosage.

Furthermore,theresponsesofthecoatedthinﬁlmsarelowerthan theoriginal,aswellas,thesignalisnotevaluableinthecaseof [ZnO₂/SF]₂₀sensorsurface.ItcanbestatedthatevaluableRsig- nalandlinearcalibrationcanonlybeattributedtoPAAcontaining multilayers(withoutcoating),andtheresponsecanbeimprovedby interlayeredmesoporoussilicafoam(Fig.7.b:forclarity,theabso- lutevaluesoftheresponsesareplotted).Insummary,itwasfound thatapplying[ZnO₂/PAA],[ZnO₂/PAA]+MPScoatingor[ZnO₂/SF]

structuredthinlayersmostlyfailedduetothesignaldriftandnon- linearsensitivity.Themixedstructureof[ZnO2/PAA/ZnO2/SF]was devoidofdriftandshowedlinearcalibrationcurves,sothistype ofhybrid(nanoparticle/polyelectrolyte/mesoporoussilica)multi- layerisan appropriate structure toapplyassensing surface in reﬂectometricinterferencesensoringasphase.Furthermore,we canconcludethatduringRIfSmeasurementsonsolid/gasinterface themonitoringofR(t)besidetheconventional(t)signalmay revealthecomplexityoftheadsorptionprocessesandmechanisms inmeso-andmacroporoushybridthinﬁlms.

Basedonthepresentedresultsthe[ZnO₂/PAA/ZnO₂/SF]₁₀thin ﬁlmwasselectedforfurtherexperiments,suchasreproducibil- ity,responsetimeanalysisandselectivity(Fig.8).Itcanbestated that the sensors signal is well reproducible (Fig.8A),the sen- sorsresponsereachesthe90%ofmaximumvaluewithin40sand it is relaxed to10% within 80s(t_90% and t_10% on Fig.8B) (the totalresponseandrecoverytimeswere180–180s).Selectivitywas testedbydropping2␮Lofdifferentliquidsintoa70^◦Cliquidsam-

Fig. 7.a Ethanol sensing (nm) vs. c(ppm) calibration curves for the pre- paredthinlayers(labels:structureofthethinﬁlmsandcalibrationequationfor [ZnO2/PAA/ZnO2/SF]10mixedstructure).bEthanolsensingRvs.c(ppm)calibration curvesforthepreparedthinlayers(labels:structureofthethinﬁlmsandcalibration equationsfor[ZnO2/PAA]20and[ZnO2/PAA/ZnO2/SF]10mixedstructures).

pleholder, which wasconnected into the 1000mL/min carrier gasstreamtowardstheRIfStestcell.Theliquidsweremethanol, ethanol,n-hexane,tolueneandxylene.Thestudiedthinﬁlmswere [ZnO₂/PAA]₂₀and[ZnO₂/PAA/ZnO₂/SF]₁₀toinvestigatetheeffect ofsilicafoamontheselectivity.Itwasestablishedthatinthecase ofZnO2/PAA/SFmixedstructure2–3timeshigherresponse was observedforethanolthanfortheothervolatileorganiccompounds (VOC),althoughalsotheafﬁnitytoaromaticmoleculesincreased comparedtoZnO2/PAAstructure(Fig.8C).

3.4. Extendingtheconcentrationrange

ThesensorsignalandcalibrationcurveofZnO₂/PAA/ZnO₂/SF mixedstructureinthelowppmrange(∼0.48-11.9ppm)waspre- sentedonFig.6.aand7.a,respectively.Itwasdemonstratedthat thelowest set and detected ethanol vapour concentration was 475±40ppb,andthevs.ccalibrationcurvewaslinearinthe c=475–11880ppbrange.Nextthe[ZnO₂/PAA/ZnO₂/SF]₁₀thinﬁlm wassubjectedtosensorialtestinahigher,c=2.46-37(±0.68)ppm range(Fig.9.a).Theconcentrationsteps(0–50min)wererepeated aftera50minlongbaselinestabilitytest.Thestatementwasthat thesensorhasafairlystablebaseline(withoutdrift),butthe␭ calibrationcurveshowedaslightquadraticdeviationfromlinear behavior(Fig.9.b).Themostimportantparametersaresummarized inTable1.

(7)

Table1

Structure,layernumberandlayerthicknessofthinﬁlmsusedinsensorialapplication,andconcentrationrange,calibratingequation(asthefunctionofconcentration, c),R²parameteranddriftpropertiesobtainedfromsensorialtests(*:LODandLODassumingerror-freecalibrationvalueswerecalculatedby[47]for[ZnO2/PAA/ZnO2/SF]10

inc=0.48-11.9ppmrange).Thetablecontainsearlierreflectometricinterferenceresultsforcomparison(originalandsurfacemodified–bybutyltrichlorosilane,BTS– [ZnO2/poly(styrenesulfonate)]20thinfilms).

Thinﬁlm d(nm) crange(ppm) Calibration,␭=f(c) R² Drift LOD(ppm)

[ZnO2/PAA]20 782 0.48–11.9 −0.0086c²+0.339c 0.997 yes –

[ZnO2/SF]20 989 3rdorder – yes –

[ZnO2/PAA/ZnO2/SF]10 894 0.586c^a 0.996 no 0.61^b

[ZnO2/PAA/ZnO2/SF]10 894 0.573c^c 0.996 no 0.63^d

[ZnO2/PAA/ZnO2/SF]10 894 2.46–37 0.0107c²+0.427c 0.999 no –

[ZnO2/PSS]20e 514 0−128 0.038c+0.536 0.998 no 29.2

[ZnO2/PSS]20+BT^e 0−106 0.331c–0.652 0.997 no 10.3

[ZnO2/PSS]20+BTS^e 0−50 0.293c–0.140 0.999 no 5.5

Theresultsindicatedbyboldfontarethemostimportantresultsofthisarticle,andasigniﬁcantpartofthediscussionisdetailedaboutthis(ZnO2/PAA/ZnO2/SF)typeof nanostructure.

acalibrationbyusingcontinuouslyselectedamountsofconcentrations.

bLODefc(assumingerror-freecalibration)=0.27ppm;[47].

c calibrationbyusingrandomlyselectedamountsofconcentrations.

d LODefc(assumingerror-freecalibration)=0.29ppm;[47].

eD.Seb ˝ok,I.Dékány,Sensor.Actuat.B-Chem.206(2015)435–442.[28].

Fig.8.Responseanalysisfor[ZnO2/PAA/ZnO2/SF]10mixedstructure:(A)repro- ducibility, (B) response and recovery times and (C) selectivity compared to [ZnO2/PAA]20thinﬁlm.

Fig.9.aEthanolsensingtestinextendedconcentrationrange:vs.tcurvesfor [ZnO2/PAA/ZnO2/SF]10thinﬁlminthe2.46–37ppmrange(blackline)compared tothe0.48-11.9ppmconcentrationrange(grayline)(labels:ethanolconcentration steps).bEthanolsensingtestinextendedconcentrationrange:(nm)vs.c(ppm) calibrationcurvesfor[ZnO2/PAA/ZnO2/SF]10thinﬁlminthe2.46–37ppmrange (whitesquares)comparedtothe0.48-11.9ppmconcentrationrange(blackcircles) (labels:calibrationequations).

(8)

structures were subjected to sensorial tests in the gas phase.

We showed that thedetection limit of the sensor is sub-ppm (<500ppb),but onlythemixed(ZnO₂/PAA/ZnO₂/SF)nanostruc- tureshowedlinearsensitivityinthe0.4-11.9ppmrangewithout responsedrift,whileboththeresponsetimeandselectivityremain reasonablegood.Testingthesensorin extended(upto37ppm) concentrationrangeshowedaslightquadraticdeviationfromlin- ear behavior. In the future the functionalization of the sensor surfacebydifferentmodifyingagentsisexpectedtoenhancethe selectivityandsensitivityofthesensor.

Acknowledgements

Theauthorsareverythankful forthefinancialsupportfrom TheHungarianScientificResearchFund(NKFIHOTKA)PD116224 andGINOP-2.3.2-15-2016-00013.Theworkwasearlierpartially supported by the European Union and the State of Hungary, co-financed by the European Social Fund in the framework of TÁMOP4.2.4.A/2-11-1-2012-0001‘NationalExcellenceProgram’.

ASisgratefulforthesupportofJánosBolyaiResearchScholarship oftheHungarianAcademyofSciences.

AppendixA. Supplementarydata

Supplementarydataassociatedwiththisarticlecanbefound,in theonlineversion,athttp://dx.doi.org/10.1016/j.snb.2016.12.097.

References

[1]X.Feng,J.Jiang,H.Ding,R.Ding,D.Luo,J.Zhu,Y.Feng,X.Huang, Carbon-assistedsynthesisofmesoporousSnO2nanomaterialashighly sensitiveethanolgassensor,Sensor.Actuat.B-Chem.183(2013)526–534, http://dx.doi.org/10.1016/j.snb.2013.04.006.

[2]S.Sharifpour-Boushehri,S.M.Hosseini-Golgoo,M.-H.Sheikhi,Alowcostand reliableﬁberopticethanolsensorbasedonnano-sizedSnO2,Opt.Fiber Technol.24(2015)93–99,http://dx.doi.org/10.1016/j.yofte.2015.05.002.

[3]W.Tan,Q.Yu,X.Ruan,X.Huang,DesignofSnO2-basedhighlysensitive ethanolgassensorbasedonquasimolecular-clusterimprintingmechanism, Sensor.Actuat.B-Chem.212(2015)47–54,http://dx.doi.org/10.1016/j.snb.

2015.01.035.

[4]L.Zhang,J.Zhao,H.Lu,L.Li,J.Zheng,H.Li,Z.Zhu,Facilesynthesisand ultrahighethanolresponseofhierarchicallyporousZnOnanosheets,Sensor.

Actuat.B-Chem.161(2012)209–215,http://dx.doi.org/10.1016/j.snb.2011.

10.021.

[5]L.Wang,Y.Kang,X.Liu,S.Zhang,W.Huang,S.Wang,ZnOnanorodgassensor forethanoldetection,Sensor.Actuat.B-Chem.162(2012)237–243,http://dx.

doi.org/10.1016/j.snb.2011.12.073.

[6]F.Meng,S.Ge,Y.Jia,B.Sun,Y.Sun,C.Wang,H.Wu,Z.Jin,M.Li,Interlaced nanoﬂake-assembledﬂower-likehierarchicalZnOmicrospherespreparedby bisolventsandtheirsensingpropertiestoethanol,J.Alloy.Compd.632(2015) 645–650,http://dx.doi.org/10.1016/j.jallcom.2015.01.289.

[7]P.Wang,Y.Fu,B.Yu,Y.Zhao,L.Xing,X.Xue,Realizingroom-temperature self-poweredethanolsensingofZnOnanowirearraysbycombiningtheir piezoelectric,photoelectricandgassensingcharacteristics,J.Mater.Chem.A 3(2015)3529–3535,http://dx.doi.org/10.1039/C4TA06266C.

[8]X.J.Li,S.J.Chen,C.Y.Feng,Characterizationofsiliconnanoporouspillararray asroom-temperaturecapacitiveethanolgassensor,Sensor.Actuat.B-Chem.

123(2007)461–465,http://dx.doi.org/10.1016/j.snb.2006.09.021.

[9]S.Zhang,P.Song,H.Yan,Z.Yang,Q.Wang,Asimplelarge-scalesynthesisof mesoporousIn2O3forgassensingapplications,Appl.Surf.Sci.378(2016) 443–450,http://dx.doi.org/10.1016/j.apsusc.2016.04.019.

[14]T.T.LeDang,M.Tonezzer,PolycrystallineNiOnanowires:scalablegrowthand ethanolsensing,Proc.Eng.120(2015)427–434,http://dx.doi.org/10.1016/j.

proeng.2015.08.658(Eurosensors2015).

[15]M.S.Hosseini,S.Zeinali,M.H.Sheikhi,Fabricationofcapacitivesensorbased onCu-BTC(MOF-199)nanoporousﬁlmfordetectionofethanolandmethanol vapors,Sensor.Actuat.B-Chem.230(2016)9–16,http://dx.doi.org/10.1016/j.

snb.2016.02.008.

[16]A.Renitta,K.Vijayalakshmi,Anovelroomtemperatureethanolsensorbased oncatalyticFeactivatedporousWO3microspheres,Catal.Commun.73 (2016)58–62,http://dx.doi.org/10.1016/j.catcom.2015.10.014.

[17]Z.Zhu,C.-T.Kao,R.-J.Wu,AhighlysensitiveethanolsensorbasedonAg@TiO2

nanoparticlesatroomtemperature,Appl.Surf.Sci.320(2014)348–355, http://dx.doi.org/10.1016/j.apsusc.2014.09.108.

[18]X.L.Xu,Y.Chen,S.Y.Ma,W.Q.Li,Y.Z.Mao,S.H.Yan,T.Wang,Facilesynthesis ofSnO2mesoporoustubularnanostructurewithhighsensitivitytoethanol, Mater.Lett.143(2015)55–59,http://dx.doi.org/10.1016/j.matlet.2014.12.

064.

[19]M.Bagheri,A.A.Khodadadi,A.R.Mahjoub,Y.Mortazavi,Gallia–ZnO nanohybridsensorswithdramaticallyhighersensitivitytoethanolin presenceofCO,methaneandVOCs,Sensor.Actuat.B-Chem223(2016) 576–585,http://dx.doi.org/10.1016/j.snb.2015.09.137.

[20]S.Luo,Y.Shen,Z.Wu,M.Cao,F.Gu,L.Wang,Enhancedethanolsensing performanceofmesoporousSn-dopedZnO,Mat.Sci.Semicon.Proc.41(2016) 535–543,http://dx.doi.org/10.1016/j.mssp.2015.10.001.

[21]M.NaderiNasrabadi,Y.Mortazavi,A.A.Khodadadi,Highlysensitiveand selectiveGd2O3-dopedSnO2ethanolsensorssynthesizedbyahigh temperatureandpressuresolvothermalmethodinamicroreactor,Sensor.

Actuat.B-Chem.230(2016)130–139,http://dx.doi.org/10.1016/j.snb.2016.

02.045.

[22]D.Zhu,Y.Fu,W.Zang,Y.Zhao,L.Xing,X.Xue,Room-temperature self-poweredethanolsensorbasedonthepiezo-surfacecouplingeffectof heterostructured␣-Fe2O3/ZnOnanowires,Mater.Lett.166(2016)288–291, http://dx.doi.org/10.1016/j.matlet.2015.12.106.

[23]Y.Lin,P.Deng,Y.Nie,Y.Hu,L.Xing,Y.Zhang,X.Xue,Room-temperature self-poweredethanolsensingofaPd/ZnOnanoarraynanogeneratordrivenby humanﬁngermovement,Nanoscale6(2014)4604–4610,http://dx.doi.org/

10.1039/C3NR06809A.

[24]B.-Y.Kim,J.S.Cho,J.-W.Yoon,C.W.Na,C.-S.Lee,J.H.Ahn,Y.C.Kang,J.-H.Lee, ExtremelysensitiveethanolsensorusingPt-dopedSnO2hollownanospheres preparedbyKirkendalldiffusion,Sensor.Actuat.B-Chem.234(2016)(2016) 353–360,http://dx.doi.org/10.1016/j.snb.2016.05.002.

[25]Q.Wang,X.Li,F.Liu,Y.Sun,C.Wang,X.Li,P.Sun,J.Lin,G.Lu, Three-dimensionalﬂake-ﬂowerCo/Snoxidecompositeanditsexcellent ethanolsensingproperties,Sensor.Actuat.B-Chem.230(2016)17–24,http://

dx.doi.org/10.1016/j.snb.2016.01.147.

[26]M.Penza,G.Cassano,P.Aversa,F.Antolini,A.Cusano,M.Consales,M.

Giordano,L.Nicolais,Carbonnanotubes-coatedmulti-transducingsensorsfor VOCsdetection,Sens.ActuatorsB:Chem.111–112(2005)171–180,http://dx.

doi.org/10.1016/j.snb.2005.06.055.

[27]M.Consales,A.Crescitelli,M.Penza,P.Aversa,P.DelliVeneri,M.Giordano,A.

Cusano,SWCNTnano-compositeopticalsensorsforVOCandgastrace detection,Sensor.Actuat.B-Chem.138(2009)351–361,http://dx.doi.org/10.

1016/j.snb.2009.02.041.

[28]D.Seb"ok,I.Dékány,ZnO2nanohybridthinﬁlmsensorforthedetectionof ethanolvapouratroomtemperatureusingreﬂectometricinterference spectroscopy,Sensor.Actuat.B-Chem.206(2015)435–442,http://dx.doi.org/

10.1016/j.snb.2014.09.087.

[29]Q.Qi,T.Zhang,X.Zheng,L.Wan,Preparationandhumiditysensingproperties ofFe-dopedmesoporoussilicaSBA-15,Sensor.Actuat.B-Chem135(2008) 255–261,http://dx.doi.org/10.1016/j.snb.2008.08.036.

[30]Y.Liu,J.Chen,W.Li,D.Shen,Y.Zhao,M.Pal,H.Yu,B.Tu,D.Zhao,Carbon functionalizedmesoporoussilica-basedgassensorsforindoorvolatile organiccompounds,J.ColloidInterf.Sci.477(2016)54–63,http://dx.doi.org/

10.1016/j.jcis.2016.05.040.

[31]D.Seb"ok,EditCsapó,NóraÁbrahám,ImreDékány,Reflectometric measurementofn-hexaneadsorptiononZnO2nanohybridfilmmodifiedby hydrophobicgoldnanoparticles,Appl.Surf.Sci.333(2015)48–53,http://dx.

doi.org/10.1016/j.apsusc.2015.01.150.

[32]G.Gauglitz,J.Krause-Bonte,H.Schlemmer,A.Matthes,Spectralinterference refractometrybydiodearrayspectrometry,Anal.Chem.60(1988) 2609–2612,http://dx.doi.org/10.1021/ac00174a015.