Microstructural characterization of hydrophobic Ti1−xAlxN coatings with moth-eye-like surface morphology
Abstract
Ti1−xAlxN thin films with different Al content were deposited by magnetron sputtering. The combination of electron energy loss spectroscopy (EELS) and energy dispersive spectroscopy (EDS) was used to evaluate the composition of the coatings. The effect of Al content on the morphology and properties of the coatings was investigated. High resolution electron microscopy and related techniques revealed the formation of a pillared moth-eye-like nanostructure with variable size and distribution of meso- and nano-columns and different degree of open porosity that depends on the Al content on the coating. For low Al content (x ≤ 0.21) c-(Ti,Al)N highly porous columns ending in a sharp pyramidal shape present low reflectivity and high hydrophobicity. While the precipitation of h-AlN phase at the column boundaries for x = 0.71 suppresses the c-(Ti,Al)N columnar growth and produces a smother surface, with higher reflectivity and less hydrophobic character.
Full text
1 Post‐Printof:JournalofAlloysandCompoundsVolume536,Supplement1,25September 2012,PagesS398–S406 InternationalSymposiumonMetastable,AmorphousandNanostructuredMaterials, ISMANAM‐2011(26thJunetoJuly1st2011) MicrostructuralcharacterizationofhydrophobicTi1−xAlxNcoangswithmoth‐eye‐like surfacemorphology V.Godinho(a),C.Lopez‐Santos(b),T.C.Rojas(a),D.Philippon(a),M.C.JimenezdeHaro(a), S.Lucas(b),A.Fernandez(a) (a)InstitutodeCienciadeMaterialesdeSevillaCSIC‐Uni.Sevilla,Av.AmericoVespucio49, 41092Sevilla,Spain (b)NAmurResearchInstituteforLIfeSciences(NARILIS),ResearchCenterinPhysicsofMatter andRadiation(PMR),UniversityofNamur(FUNDP),61RuedeBruxelles,5000Namur,Belgium Abstract Ti1−xAlxNthinfilmswithdifferentAlcontentweredepositedbymagnetronspuering.The combinationofelectronenergylossspectroscopy(EELS)andenergydispersivespectroscopy (EDS)wasusedtoevaluatethecompositionofthecoatings.TheeffectofAlcontentonthe morphologyandpropertiesofthecoatingswasinvestigated.Highresolutionelectron microscopyandrelatedtechniquesrevealedtheformationofapillaredmoth‐eye‐like nanostructurewithvariablesizeanddistributionofmeso‐ andnano‐columnsanddifferent degreeofopenporositythatdependsontheAlcontentonthecoating.ForlowAlcontent(x≤ 0.21)c‐(Ti,Al)Nhighlyporouscolumnsendinginasharppyramidalshapepresentlow reflectivityandhighhydrophobicity.Whiletheprecipitationofh‐AlNphaseatthecolumn boundariesforx=0.71suppressesthec‐(Ti,Al)Ncolumnargrowthandproducesasmother surface,withhigherreflectivityandlesshydrophobiccharacter. Keywords Ti1−xAlxNcoangs;Moth‐eye‐likestructure;EELS;HAADF/STEM;Antireflective;Hydrophobic 1.Introduction IntherecentyearsthefieldofapplicationofTi1−xAlxNcoangshasspreadfromthe traditionalmechanicalcuttingtools[1]toelectronicdevices[2],solarselectiveapplications[3], [4]and[5]andeventosatelliteindustry[6]and[7].TheopticalpropertiesoftheseTi1−xAlxN coatingscanbetunedinawiderangefrommetallictodielectriccharacterbycontrollingthe stoichiometryx.Inparticular,thesecoatingsarewidelyusedinselectiveinfrared(IR) applicationslikesolar‐controlwindowswhereahighcolor‐neutraltransmittanceinthevisible (VIS),goodreflectanceatthenear‐infraredandlowthermalemittanceareimportant properties[5];solarabsorbercollectorswithhighabsorptanceinthesolarspectrumandlow thermalemittanceintheIR[8];andsatellitetemperaturecontrolcoatingswherethebalance betweensolarabsorptanceandthermalemittanceiscrucialtocontrolthetemperature,
2 decreasingtheabsorptanceandincreasingtheemittanceofthecoatingsanequilibrium temperatureforthesatellitecanbeachieved[6]. ThetypicalmorphologiesofTiAlNcoatingsdepositedunderlowenergeticconditions, composedofcolumnsinthegrowingdirection,endinginapyramidalshape(inthe100nm range)havebeenreportedasagoodstrategyforobtainingantireflectivecoatings[9]and[10]. Thesestructuresmimictheeyesandwingsofcertainspeciesofmothwhicharecoveredby taperedpillarssmallerthanthewavelengthofincidentlightsuppressingthereflectionoflight atwavelengthsfromtheultraviolettomid‐infrared[11]. Inaddition,thenanoscalehierarchicalstructuresoftheTiAlNcoatingsproducedunder differentdepositionconditions,generatecoatingswithdifferentsurfacefreeenergyand wettingproperties[12]whosecontrolisofinterestfortheabovementionedapplications; namelyforsolar‐controlwindowsorsolarabsorbers.Thepartialpressureofnitrogenduring depositionisreportedtoaffectthemorphologyofTi1−xAlxNcoangs;anincreaseintheN2 fractionproducesadecreaseofsurfaceroughnessforsimilarpowersuppliedtothetargets[7] and[13]. Fromthemicrostructuralpointofview,theeffectofcontinuouslyaddingAltotheTiNlattice hasbeenwidelystudiedbymanyauthors[14],[15],[16]and[17].Thereisageneralconsent thatforlowAlcontent,asinglephasesolidsolutionc‐(Ti,Al)NispresentwhileforhigherAl contentadual‐phasestructureofc‐(Ti,Al)Nandh‐AlNcanbefound,afurtherincreaseinthe Alamountleadstoasinglephaseh‐AlN.Thelimitsoftheseregionsarenotalwaysthesame forthedifferentauthors[3]anditisalsopossiblethatthedepositionparametersplayamajor roleonthemicrostructureandthelimitsofeachphase[15]. Inthisworkwereporthowbychangingasetofdepositionconditions,coatingswithdifferent morphologiesandcompositionscanbepreparedbymagnetronsputtering.Inparticular,the effectsofAlcontentonthereflectancespectraandwettabilityofthecoatingsare investigated.ThemicrostructureofcoatingswithdifferentAlcontentisdeeplyinvestigated usingelectronmicroscopytechniquesandrelatedwiththedifferentpropertiespresentedby thecoatings.Previouspapers[5],[6],[7]and[9]investigateddifferentaspectsofthiscomplex systemfocusingmainlyoncompositionalchangesandreflectivitymeasurements.Theinterest ofthepresentpaperistofocus,notonlyonthecompositionvs.reflectivity(inawiderange fromVIStofarIR),buttodemonstratethatmorphological/microstructuralcharacteristicscan alsoberelatedtothemeasuredmacroscopicreflectivityandhydrophobicityproperties.The tailoredcombinationofthepropertiesandmicrostructureinawiderangeofcompositionsis ofhighinterestforpotentialapplicationsinsolar‐controlwindows[5]orsolar‐absorber collectors[8]. 2.Experimental Fig.1showsaschematicdiagramofthedepositionchamberusedandTable1summarizesthe synthesisconditionsfortheTi1−xAlxNcoangsstudiedinthiswork.PureTiandpureAltargets wereused(KurtLesker99.99%)underdcdischarge.Thedepositiontookplaceinseveralsteps, startingwithanargonatmosphereandincreasinggraduallythenitrogenpartialpressure, keepingthetotalpressureinthechamberconstant(seeTable1).InTable1thenitrogen percentagecorrespondstothenitrogenpartialpressureinthegasphaseforeachstepkeeping thetotalpressure(N2+Ar)at1.33Pa.Nosubstratebiasoradditionalheatingwasapplied duringdeposition(depositiontemperaturewas∼120°C).ThepurposeistoobtainTi1−xAlxN coatingswithdifferentxanddifferentstructure.
3 ThecoatingsweredepositedonAISI316steel,sodiumchlorideandsilicon(100)substrates. Scanningelectronmicroscopywithfieldemissiongun(SEM‐FEG)wasperformedtostudythe morphologyandthicknessofsamplesinahighresolutionmicroscope,HITACHIS4800.The sampleswerecleavedfromcoatingsgrownontosilicon,andwereobservedwithout metallizationincrosssectionalviewsat1–2kV. Thesurfaceroughness(rootmeansquareroughness—rmsroughness)wasevaluatedusinga Mahr“stylus”profilometer(conicstylusdimensionof2μmwithalateralresolutionof0.76 nm)alonga7.5mmprofile. ThestructureofthesecoatingswasinvestigatedbygrazingincidenceX‐raydiffractionand selectedareaelectrondiffraction.ThegrazingincidenceX‐raydiffraction(GIXRD)equipment usedwasaSiemensD5000(CuKαradiation).ForGIXRDanalysissamplesdepositedonsilicon substrateswereused. APhilipsCM200transmissionelectronmicroscope(TEM)equippedwithaparalleldetection electronenergylossspectroscopy(EELS)spectrometerfromGatan(766‐2kV)wasused.For TEManalysisthinfoilswerepreparedfromcross‐sectionedslicesofcoatingsdepositedon silicon(100),bymechanicalpolishingfollowedbyargonionmillingtoelectrontransparency. Alsoplanarviewsampleswerepreparedfromsamplesdepositedonsodiumchloride substrates,whichwerefloatedoffindistilledwaterandsupportedonaTEMcoppergrid. Electronenergy‐lossspectroscopyandX‐rayenergydispersivespectroscopy(EDS)wereused ontheTEMtoanalyzethecompositionofthefilmsinthecross‐section.Selectedareaelectron diffraction(SAED)wasusedtoinvestigatetheformationofcrystallinephasesinthenanoscale. High‐angleannulardarkfieldinascanningtransmissionelectronmicroscopy(HAADF‐STEM) andEELSspectraobtainedbythespectrum‐imagingmodewererecordedonaJEOLJEM‐2010F workingat200KeV.ThesignalintensityoftheHAADF‐STEMimagesisproportionaltothe atomicnumberandthethicknessofthesampleontheanalyzedarea.Thespectrumimaging mode,consistinginacquiringaseriesofEELSspectrawhilea0.5nmbeamscannedalongthe samplearea,wasused.TheHAADFsignalwassimultaneouslycollectedateachpointwithin thescannedarea.Thisapproachallowscorrelatingnanoanalyticalandstructuralinformation oftheregionunderstudy. SamplessurfacecompositionwasinvestigatedbyX‐rayphotoelectronspectroscopy(XPS).The XPSequipmentwasaLeyboldHerausSpectrometerworkingintheconstantanalyzerenergy modewithapass‐energyof50eV.ThesampleswereanalyzedwithMgKα radiation.The bindingenergyreferencewastakenasthemaincomponentoftheC1speakat284.6eVfora mixtureofadventitiousandpurecarbon. NearIRandIRreflectivitymeasurementsinthespectralrangeof0.8–1.1and1.3–200μ m respectively,havebeendoneusingIFS66v/s(BRUKER)infraredFourier‐transform spectrometerwithreflectionunitsatthenearnormalincidenceandattheobliqueanglesof incidenceuptothegrazingangles(reflection–absorptionspectroscopy(RAS))inpolarizedlight forsamplesgrownonsteelsubstrates.Thespectralresolutionwas4cm−1. Inthevisiblerange(0.4–0.8μm)thecoatingstotalreflectancewasmeasuredinaShimadzu UV‐2101PCspectrometerequippedwithanintegratingsphere.Bariumsulphatewasusedas reference.Themeasurementswereperformedoncoatedsteelsubstrates(AISI316).
4 Hydrophobicityanalysiswasperformedthroughstaticwatercontactanglemeasurementswith OCA20equipmentandSCA20software(Dataphysics).Bidistilledwatersessiledropsof1μl wereemployed.Thereportedvaluesareanaverageoftenmeasurementstakenforeach examinedsurfaceofcoatingsdepositedonsilicon.Inasimilarway,contactanglesof iodomethanewerealsomeasuredtoestimatethepolarandnon‐polarcontributionstothe surfacetensionbytheOwens–Wendt–Kaelbleapproximation[18]. 3.Resultsanddiscussion OnTable2thechemicalcompositionofthecoatingsgivenbyEELSandEDSispresented.EELS spectra,NandOK‐edgesandTiL2,3‐edge,wereregisteredoncoatingscross‐section,and thesedatatogetherwithEDSdata(ofTiandAl)wereusedtodeterminethechemical compositionofthesamples.Bycontrollingthedepositionparameterssampleswithdifferent stoichiometryxofTi1−xAlxNwereobtained.However,thecoatingsarecomposed,notonlyof Ti,AlandN,alsoanOcontamination(from5to22at%)wasfound.Duetothevacuum conditionsandthehighreactivityofTiandAltowardoxygenitisnotpossibletoexcludethe incorporationofoxygeninthefilmsduringdepositionunderourexperimentalconditions,in particularforlowerdepositionsrates(seedepositionratesinTable1)[19].Inadditionaswe willdiscussbelow,thedesiredcolumnarmicrostructureofthesefilmsforpromotinglow reflectivityandhighhydrophobicity,willalsoinfluencethepost‐depositionincorporationof oxygen. SimilarxvaluesforAlcontentwerefoundbyXPS.However,surfaceanalysistechniquesasXPS arenotappropriateforthecharacterizationofchemicalcompositionofcoatingscomposedof elementswithhighaffinitytooxygenduetoitsreadilyoxidationwhenexposedtoair. ThecrystallinephasesformedwereevaluatedbyGIXRD.Fig.2showsthedifferentdiffraction patternsofthesesamples.ThediffractionpeaksarebetweenthoseofcubicTiNandcubic TiAlN.AstheAlcontentincreasesthereisadisplacementinthediffractionpeakstohigher2θ. AddinggraduallyAltothecubicTiNphaseleadstoanincreaseofthecrystallinephaseanda decreaseoflatticeparameterscausedbythesubstitutionofTiatomsbyAlatomswithsmaller atomicradius;shiftingthe2θ peaksinTi1−xAlxNtohighervaluesascomparedtoTiN.The highestAlcontentleadstotheappearanceofasmallpeakat2θ=66°thatcouldbeassociated withthepresenceofh‐AlNphases. Nocrystallineoxidephaseswerefound.Thehighoxygencontaminationisprobablyleadingto theformationofamorphousoxideoroxynitridephasesofTiandAl. ThephasesformedwerealsoanalyzedbySAED.Fig.3presentstheresultsasafunctionofthe Alcontentinthefilms.ForAlcontentslowerthanx=0.59asinglecubic(Ti,Al)N(c‐(Ti,Al)N) phaseispresent.Atx=0.59anhexagonalAlN(h‐AlN)phaseco‐existswiththec‐(Ti,Al)N phase.TheseresultsareinagreementwithZhouetal.[15]thatdefinearegionbetweenx= 0.6andx=0.7whereitispossibletofindbothphases,whilebelowandabovetheselimits onlyc‐(Ti,Al)Nandh‐AlNphasesarerespectivelyfound. ThebondingstructureofthecoatingswasevaluatedbyEELSinsidetheTEM.Fig.4(a)presents indetailtheNK‐edge.ThespectraaresimilartothosepresentedbyMacKenzieetal.[20]for TiAlNcoating,andshowcleardifferencesintheNK‐edgewiththeincreaseofAlcontent.The introductionofAlinthec‐TiNphaseincreasesthesecondresonanceoftheNK‐edge,whilethe firstresonancedecreases.ThecontinuousincreaseinAlleadstotheprecipitationoftheh‐AlN phasethatproducesbroadeningofthesecondresonanceandthedisappearanceofthefirst resonance.
5 AccordingtoMacKenzieetal.[20]theEELSspectraofc‐(Ti,Al)Nshouldbealinearcombination ofc‐TiNandc‐AlNspectra.Inthissense,usingthereferencespectraofc‐TiN,c‐AlNandh‐AlN, theNK‐edgesoftheTi1−xAlxNcoangswerecalculatedtakingintoaccountthedifferent compositionsandtheSAEDresultsforthedifferentphasesformed.Fig.4(b)showstheresults. ThelinearcombinationswereperformedassumingthatallTiinthecoatingwasintheformof c‐(Ti,Al)N.WhenthereisanexcessofAl,thisexcessisintheformofh‐AlNgivingrisetothe coexistenceofthetwophasesasdemonstratedbyTEM‐SAED.Thecalculatedspectraarein goodagreementwiththeexperimentalresults,andthemethodispresentedhereasgoodway topredictexperimentalresults. However,notonlycompositionisinfluencingthepropertiesofthecoatings,thesurface morphologyofTiAlNcoatingsisveryimportantinwhatconcernstheirwettabilityandits applicationasantireflectivecoatings.Thenitrogenpartialpressureduringdepositionis reportedtoaffectthesurfaceroughnessofTi1−xAlxNcoangs[7]and[13],buttheeffectsof Alcontentonsurfaceroughnesshavebeenalsoreported[12].Inthisworkwefocusonthe studyoftheeffectofthedifferentAlcontentonthemorphologyandmicrostructureofthe coatingsdepositedundersimilarnitrogenpressure.Inparticular,Fig.5showsindetailthe differentmorphologiesobtainedbySEM‐FEGcross‐sectionalandplanarviewsfortheselected sampleswithx=0.07,0.21and0.71. AcolumnargrowthtypicalofsamplesgrownunderlowadatommobilityasdescribedbyZone 1ofThornton'sdiagramisobserved[21].Thedifferentpowerratiosappliedunderthesame nitrogenpressuregiverisetodifferentstructures.Ingeneralallsamplespresentacolumnar structureconstitutedby“meso‐columns”(understandingthatmesoisbetweenmicroand nano)whichareformedbyaggregatesofsmallercolumns(“nano‐columns”).Itispossibleto seethevoidsbetweenthemeso‐columns. ForverylowAlcontentthec‐(Ti,Al)Nmeso‐columns(150–200nmwide)areformedbynanocolumnswithopenspacesbetweenthem,whichgrowtiltedinrespecttothedirectionofthe meso‐columns.Acompetitivegrowthofthenano‐columnsinalldirectionsresultsinapeculiar “fractalishcauliflower”structurewithpyramidalending. Whenthesamepowerisappliedtobothtargets(x=0.21)thecompetitivegrowthofthe nano‐columsissuppressed.Thenano‐columnsgrowparalleltoeachotherinanorderedway. Themeso‐columnsof70–90nmendintriangularpyramidstypicalofc‐(Ti,Al)Ncoatings[5],[7] and[12]. Forx=0.71themeso‐columnsarenowformedbynano‐columnsinthegrowingdirectionwith globularending,resultinginacompactstructurewithacauliflowerlooksurface.Theincrease ofthepowersuppliedtotheAltargetsproducesanexcessofAl,thatprecipitatesinanh‐AlN phasewhichinterruptsthec‐(Ti,Al)Ncolumnargrowthoriginatinginasmoothersurface[7]. Highresolutiontransmissionelectronmicrograph(oncoating'scross‐section)intheinsetof Fig.6revealedthatthehexagonalAlNphase,characterizedbyaninterplanardistanceof2.7Å, precipitatesatthenano‐columnboundaries.Alsothecharacteristic2.4Åinterplanardistance correspondingtocubicTiAlNphaseisobserved.Asobservedfortheplanarobservationsthe SAEDpatterncorrespondstoamixtureofc‐(Ti,A)Nandh‐AlNphase. Asonecouldexpect,thehighoxygencontaminationobservedcanalsobeduetothecolumnar structureandtheopenvoidsbetweenandinsidethemeso‐columns.Inparticular,samples TiAlN(O)/200–200andTiAlN(O)200–400withx=0.21andx=0.59respectively,present
6 oxygencontentsof22and13at%.Themicrostructureofthesecoatingswasinvestigatedin moredetailbycross‐sectionalTEM(XTEM).InFig.7(a)and(b)thecolumnarstructureofthese samplescanbeobservedinmoredetailbothbySEMandTEM.SamplewithhigherAlcontent (x=0.59,Fig.7(a)presentsmeso‐columnsfrom100to150nmwideendinginpyramidsof∼20 nmheight.FromtheTEMmicrographitispossibletoseetheabruptcolumnswithintracolumnarvoids.Insidethecolumns,asuperlatticemicrostructurewithaperiodof8nmcanbe seen.Thiscorrespondstothenano‐columnsgrowingtiltedwithrespecttothegrowing direction.Thecolumnsareformedbyc‐(Ti,Al)Nwiththeprecipitationofh‐AlNatcolumns boundariesasindicatedbytheSAEDpattern,correspondingtotheco‐existenceofboth phases.Inthiscasetheamountofh‐AlN(estimatedataround18%fromtheEELSlinear combination)phaseisnotsufficienttointerruptthegrowthofthetypicalc‐(Ti,Al)Nstructure asobservedinthesamplewiththehigherAlcontent(withanh‐AlNcontributionof42% accordingtotheEELSresults). ForlowerAlcontent(x=0.21,Fig.7(b)thecolumnsarenarrower,butfromthecontrastsin theXTEMimage,asimilarsuperlatticestructureofthesmallnano‐columnstiltedinrespectto thefilmgrowingdirectionisobserved.IntheHAADF‐STEMimagethedarkerareas, correspondingtotheporesbetweenthetiltednano‐columns,arefoundtofollowthesame pattern.TheEELSanalysisinsideandoutsidethepores,ispresentedinFig.7(b).No appreciablechangesareobservedfortheNKandTiL2,3‐edges.However,insidethepores thereisasignificantincreaseoftheOK‐edge.Thehighlyporousstructure,withopenporesis thereforethereasonforthehighoxygencontentmeasuredbyEELSinthesecoatingscrosssection. TheopticalpropertiesofthedifferentstructuresinFig.5withsub‐wavelengthscalefeatures wereinvestigated.Moth‐eye‐likeantireflectingcoatingsplayanimportantroleinawide varietyofopticaltechnologiesbyreducingreflectivelossatinterfaces.Fig.8presentsthe reflectivityofthesecoatingsinthespectralrangefrom0.4to0.8,0.8–1.1and1.3–200μm. Thethreecoatingspresentverylowreflectivity(<0.3)upto1.1μm.Thesharpandregular moth‐eye‐likestructureofthecoatingwithx=0.21isresponsibleforthelowerreflectivity spectrum. Forwavelengthshigherthan1.3μmthereflectivityofthecoatingsincreases,beingsimilarto TiNinthecaseofx=0.21.Thehighthicknessofx=0.07causesinterferenceeffects,thatcan beusedtocontrolthesolarselectiveresponseofthecoatingsinparticulartoreducethe equilibriumtemperature. Thewettingpropertiesofthesecoatingswerealsotested.Table3presentstheresultsofstatic watercontactangleandsurfaceenergy.Ti1−xAlxNcoangsshowanotablehydrophobic character,withwatercontactanglesaround110°.However,itispossibletoobserveslight differencesbetweenthewettabilityofthesesamplesthatcouldbeattributedtodifferences onthedevelopedmicrostructureandthesurfacechemicalcomposition.Inthiscase,the smootherandmorecompactsamplewithx=0.71(afterthestructuralphasechangewith segregationofh‐AlN),presentsasmallerwatercontactangle.Meanwhile,anincreaseinthe surfaceroughness,throughthecolumnargrowthofthecubicphase,providesmore hydrophobicsurfaces,asisthecaseofthesampleswithx=0.07andx=0.21.Thistendencyis inagreementwiththeknowneffectoftheroughnessonthewatercontactangleproposedby theWenzelmodel[22].Itisworthtomentionthatnodirectcorrelationwasfoundbetween thesamplesthicknessandsurfaceroughness(seetheresultsonTable1).Besides,previous workshavepresentedsimilarwatercontactanglesonsimilarTi1−xAlxNmorphologies[12].
7 Surfacetensionmeasurementsconfirmthisresult:thesmoothercoating(x=0.71)hasthe lowersurfacetension. ThesurfacechemistryandbondingstateoftheelementswereinvestigatedbyXPS.According toFig.9ofthenitrogenN1sXPSspectra,thecontributionofdifferentbondedspeciescanbe foundontheTi1−xAlxNsurface.Thepeaksarebroadandslightlyshiedtohigherbinding energycomparedtotypicalvaluesreportedforTiAlNcoatings[23](indicatedinthefigureby themarkedarea).Theappearanceofhighbindingenergyshouldersisobservedwithahigher intensityforthesamplewiththehigherAlcontent(x=0.71).Thesebandsareattributedtothe contributionofNsinglebondOsurfacebondsexpectedat399.1eV[23].Itappearsthatthe samplewithahigherAlcontentalsoshowsahighersurfaceoxidation.Accordingtotheplanar viewimagesinFig.5thesamplewithx=0.71showsasmoothersurfaceformedbysmaller grainsseparatedbyaclearporositywhatmakesoxygenmoreaccessibletoproduceahigher surfaceoxidation.Therefore,theprecipitationofthenewh‐AlNphaseseemstogeneratea lesshydrophobiccharacteroftheTi1−xAlxNsurface,withasmoothertopographyandahigher contentofAlNandNOgroups. 4.Conclusions Ti1−xAlxNthinfilmswithdifferentcomposionandstructureweredepositedbymagnetron sputtering.TheinfluenceofAlcontentonthemicrostructureandphaseformationhasbeen investigatedbyXRD,SAED,TEMandSEM‐FEG.Theco‐existenceofh‐AlN,andc‐(Ti,Al)Nphase, wasdetectedbySAEDandconfirmedbyEELSlinearcombinationsforx≥0.59.Thehigh resolutionTEMstudyshowedthattheh‐AlNphaseprecipitatesatthecolumnboundaries. ThedetailedmicrostructuralanalysisinthenanoscalebyhighresolutionSEMandTEM revealedtheformationofapillaredmoth‐eye‐likenanostructurewithvariablesizeand distributionofmeso‐andnano‐columnsdependingonthedepositionconditions.Forthesame nitrogenpressuredifferentmorphologieswereobtainedinfunctionoftheAlcontent.For higherAlcontent,theprecipitationoftheh‐AlNphaseatthecolumnsboundariessuppresses thegrowthofthec‐(Ti,Al)Nphaseandasmothersurfaceisobtained. HAADFandSTEM/EELSanalysis,revealedthattheoxygencontentinthecoatingsismainlydue tothepresenceofahighlyporouscolumnarstructure.TheEELSmeasurementsinsideand outsidetheporesindicatehigheroxygencontentinsideporesprovingthattheporosityisopen contributingtopost‐depositionoxidationinthecoating. Thereflectancespectrashowthatthesecoatingspresentverylowreflectivityinparticularfor columnarstructureswithsharppyramidaledges.Thisstructurealsofavorsthehydrophobic characterpresentedbyallcoatings.Thecombinationofalltheresultsreportedinthispaper allowustoproposetheTiAlNcoatingsasveryinterestinghydrophobicmaterialsforinfrared selectiveapplications.Themethodofpreparationissimpleandmorphologyandchemical compositioncanbetunedeasilybyvaryingthedepositionconditions.Inthispaperweshowed howthepromotionofcolumnarstructures(forlowIRreflectivityandhydrophobicity applications)canbeeasilycombinedwithawiderangeofcompositionswhatallowalsoto tunetheelectricalpropertiesfrommetallic(TiN)todielectric(AlN)character. Acknowledgments
8 AuthorsthankthefinancialsupportfromEC(REGPOTAL‐NANOFUNC),CSIC(201060I41and 201060E102),SpanishMinistryMICINN(ConsoliderFUNCOAT)andJuntadeAndalucía.
9 References [1]K.D.Bouzakis,G.Skordaris,N.Michailidis,I.Mirisidis,G.Erkens,R.Cremer Surf.Coat.Technol.,202(2007),pp.826–830 [2]T.H.Cha,D.G.Park,T.K.Kim,S.A.Jang,I.S.Yeo,J.S.Roh,J.W.Park Appl.Phys.Lett.,81(2002),pp.4192–4194 [3]V.Godinho,D.Philippon,T.C.Rojas,N.N.Novikova,V.A.Yakovlev,E.A.Vinogradov,A. Fernandez SolarEnergy,84(2010),pp.1397–1401 [4]H.C.Barshilia,N.Selvakumar,K.S.Rajam,D.V.S.Rao,K.Muraleedharan,A.Biswas Appl.Phys.Lett.,89(2006) [5]A.Schuler,V.Thommen,P.Reimann,P.Oelhafen,G.Francz,T.Zehnder,M.Duggelin,D. Mathys,R.Guggenheim J.Vac.Sci.Technol.A‐Vac.Surf.Films,19(2001),pp.922–929 [6]M.Brogren,G.L.Harding,R.Karmhag,C.G.Ribbing,G.A.Niklasson,L.Stenmark ThinSolidFilms,370(2000),pp.268–277 [7]J.T.Chen,J.Wang,F.Zhang,G.A.Zhang,X.Y.Fan,Z.G.Wu,P.X.Yan J.AlloysCompd.,472(2009),pp.91–96 [8]K.Gelin,T.Boström,E.Wäcklegard SolarEnergy,77(2004),pp.115–119 [9]F.Burmeister,C.Kohn,R.Kuebler,G.Kleer,B.Blasi,A.Gombert Surf.Coat.Technol.,200(2005),pp.1555–1559 [10]F.Burmeister,E.Schaffer,G.Kleer,W.Doll,B.Blasi,A.Gombert Surf.Coat.Technol.,200(2005),pp.1088–1092 [11]K.S.Liu,L.Jiang NanoToday,6(2011),pp.155–175 [12]V.Chawla,R.Chandra,R.Jayaganthan J.AlloysCompd.,507(2010),pp.L47–L53 [13]K.Chakrabarti,J.J.Jeong,S.K.Hwang,Y.C.Yoo,C.M.Lee ThinSolidFilms,406(2002),pp.159–163 [14]J.C.Schuster,J.Bauer J.SolidStateChem.,53(1984),pp.260–265
16 Figure2
17 Figure3
18 Figure4
19 Figure5
20
21 Figure6
22 Figure7
23 Figure8
24 Figure9