ContentslistsavailableatScienceDirect
Applied Surface Science
j o u r n a l ho me p ag e :w w w . e l s e v i e r . c o m / l o c a t e / a p s u s c
Effect of the deposition conditions of NiO anode buffer layers in organic solar cells, on the properties of these cells
D.T. Nguyen
a, A. Ferrec
a, J. Keraudy
a,c, J.C. Bernède
b, N. Stephant
a, L. Cattin
a, P.-Y. Jouan
a,∗aInstitutdesMatériauxJeanRouxel(IMN),UMR6502,UniversitédeNantes,2ruedelaHoussinière,BP92208,Nantes44000,France
bMOLTECHAnjou,UMR6200,UniversitédeNantes,2ruedelaHoussinière,BP92208,Nantes44000,France
cIRTJulesVerne,CheminduChaffault,44340Bouguenais,France
a r t i c l e i n f o
Articlehistory:
Received3September2013 Receivedinrevisedform4April2014 Accepted5May2014
Availableonline27May2014
Keywords:
Organicphotovoltaiccells Anodebufferlayer NiOthinfilms DCreactivesputtering Formingprocess Annealing
a b s t r a c t
NiOthinfilmsdepositedbyDCreactivesputteringwereusedasanodebufferlayerinorganicphotovoltaic cells(OPVs)basedonCuPc/C60planarheterojunctions.FirstlyweshowthatthepropertiesoftheNiO filmsdependontheO2partialpressureduringdeposition.Thefilmsarefirstconductivebetween0and 2%partialoxygenpressure,thentheyaresemiconductorandp-typebetween2and6%partialoxygen pressure,between6and9%partialoxygenpressuretheconductionisverylowandthefilmsseemtoben- typeandfinally,forapartialoxygenpressurehigherthan9%,theconductionisp-type.Themorphologyof thesefilmsdependsalsoontheO2partialpressure.WhentheNiOfilmsisthickof4nm,itspeaktovalley roughnessis6nm,whenitissputteredwithagascontaining7.4%ofoxygen,whileitismorethandouble, 13.5nm,whenthepartialpressureofoxygenis16.67%.Thisroughnessimpliesthataformingprocess, i.e.adecreaseoftheleakagecurrent,isnecessaryfortheOPVs.Theformingprocessisnotnecessaryif theNiOABListhickof20nm.Inthatcaseitisshownthatoptimumconversionefficiencyisachieved withNiOABLannealed10minat400◦C.
©2014ElsevierB.V.Allrightsreserved.
1. Introduction
Whatever the organic photovoltaic cell (OPV) families, that basedonbulkheterojunctions(BHJ)orthatbasedonaplanarhet- erojunction(PHJ), thecarriercollectiondepends onthecontact betweentheorganicmaterialsandtheelectrodes,theheterojunc- tionbeingbasedonthecontactbetweenanelectrondonor(ED)and anelectronacceptor(EA).Oneofthemostimportantrequirements forhighpowerconversionefficiencyoftheorganicsolarcellsisthe carrierextractionfromtheorganicdiodeintotheelectrodes.The energybarrierforholeextractiondependsontheenergyleveldif- ferencebetweentheworkfunctionoftheelectrodeandthehighest occupiedmolecularorbital(HOMO)oftheorganicelectrondonor ED.GoodbandmatchingbetweentheanodeandtheEDisnecessary toachievegooddeviceperformance.Howeverthemostsuccessful anode,theindiumtinoxide(ITO)hasdifficultiestoextractholes fromtheED,becauseofitslowworkfunctionrelativetotheHOMO oftheED.Tosolvetheenergylevelmismatch,electrodesmodifica- tionisnecessary[1].Oneofthepossibilitiesistheintroductionof
∗Correspondingauthor.
E-mailaddress:pierre-yves.jouan@cnrs-imn.fr(P.-Y.Jouan).
ananodebufferlayer(ABL)betweentheITOanodeandtheorganic ED. One successful demonstration is using a PEDOT:PSS layer (poly(ethylenedioxythiophene):polystyrenesulphonate)[2].How- ever,PEDOT:PSShasbeenreportedtopresentsomedisadvantages suchasinconsistentfilmmorphology,inefficientelectron-blocking performance,poorstabilityunderambientconditionsanditiscor- rosivetoITO[3].ThereforeABLhavebeenshowntobesuccessful suchasanultrathingoldfilm[4]oratransitionmetaloxide[5–8].
Amongthetransitionmetaloxide,NiOhasbeenusedwithsuccess [9–13].IthasbeenreportedthatNiOthinfilmsisap-typesemicon- ductorwhichhaveitsvalencebandedgeat5.4eVandaconduction bandedgeat1.8eV[10],sothatabetterenergylevelalignmentis obtainedthanwithITOalone.Inthepresentworkweshowthat theeffectoftheNiOABLonthebehaviouroftheorganicsolarcells dependsstronglyonthedepositionconditionsofNiO.Asamaterof fact,theNiOfilmsbeingdepositedbysputtering,theirproperties dependstronglyonthedepositionconditions
2. Experimental 2.1. Anodepreparation
TheITOcoatedglasssubstratesusedinthisstudywerecom- merciallyobtainedfromtheSOLEMS.Itsthicknessis100nm,its http://dx.doi.org/10.1016/j.apsusc.2014.05.020
0169-4332/©2014ElsevierB.V.Allrightsreserved.
conductivityisaround25/sq,itsaveragedtransmittanceinthe visibleis93%,itsroomsquaremeanssurfaceis0.8nmwhileitssur- faceworkfunctionis4.7eV[14].Afterscrubbingwithsoap,these ITOsubstrateswererinsedinrunningdeionisedwater.Thenthe substratesweredriedwithanitrogenflowandthenloadedinto avacuumchamberwhereweproceedtothedepositionofNiO.
TheNiOfilmsweredepositedbyDCreactivemagnetronsputtering usingapurenickeltarget(99.99%)andanAr+O2gasmixture.We haveusedatarget2inches(5.08cm)indiameter.Substrateswere placedatapproximately3cmfromthetarget.Thegasesargonand oxygenwereintroducedinthereactorbyamassflowcontrollers;
thedischargecurrentduringdepositionwas110mAand80mA.
2.2. Solarcellsrealisation
Inthepresentstudy,theplanarsolarcellsusedwerebasedon theclassicaljunctioncopperphthalocyanine(CuPc)/fullerene(C60).
CuPcistheelectrondonorandC60theelectronacceptor.Itisknown thatabufferlayer,calledexcitonblockinglayer(EBL),introduced betweentheelectronacceptorandthecathodeincreasesstrongly thecellsperformances(Bathocuproine(BCP)oraluminiumtris(8- hydroxyquinoline).InthepresentworkaluminiumAlq3hasbeen usedasEBLbecauseitallowsthelifetimeoftheOPVs.CuPc,C60 andAlq3 have beendeposited inavacuumof10−4Pa. Thethin filmdepositionratesand thicknesswereestimatedinsitu with a quartz monitor.Thedeposition rateand final thicknesswere 0.05nm/sand35nminthecaseofCuPc,0.05nm/sand40nmin thecaseofC60and0.1and9nmforAlq3.Thesethicknesseshave beenchosenafteroptimisation.Afterorganicthinfilmdeposition, thealuminiumtopelectrodeswerethermallyevaporated,without breakingthevacuum,throughamaskwith1mm×8mmactive areas.TheABLbeingthesputteredNiOfilm,thecellsweretypically:
glass/ITO/NiO/CuPc/C60/BCP(orAlq3)/Al.
2.3. Characterisationtechniques
Thethin filmsstructureswere analysedby X raydiffraction (XRD)byaSiemensD5000diffractometerusingK␣radiationfrom Cu(K␣=0.15406nm).
Theopticalmeasurementswerecarriedoutatroomtemper- ature usinga Carryspectrometer.The film optical densitywas measuredatwavelengthsof1.2–0.30m.
XPSmeasurementswereperformedwithaKratosAxisUltra spectrometerusingasourcemonochromatorAlK␣(1486.6eV).The spectrawererecordedforNiOthinfilmsdepositedatdifferentoxy- genlevelsforsampleswiththicknessof20nm.Kratosneutralise thesystemloadwasusedforallanalyses.Basicunitpressureis 8.10–8Torr.Highresolutionspectrawereobtainedusing20eVpass energy,ananalysisareaof≈500×500m.
Themorphologyofthedifferentstructuresusedasanodewas observedthroughscanningelectronmicroscopy(SEM)withaJEOL
7600F at the“centrede microcaractérisation del’Universitéde Nantes”.Imagesinsecondary(SEM)andbackscattering(BEI)mode havebeendone. Thecompositionofthefilmsisdeterminedby electronprobemicroanalysis(EPMA),thescanningelectronmicro- scope beingequipped with a PGT X-raymicroanalysis system;
X-raysweredetectedbyagermaniumcrystal.
AtomicForceMicroscopy(AFM)imagesondifferentsitesofthe filmweretakenexsituatatmosphericpressureandroomtem- perature.Allmeasurementshavebeenperformedintappingmode (NanowizardIII,JPKInstruments).Classicalcantileverswereused (TypePPP-NCHR-50,Nanosensor).Theaverageforceconstantand resonancewereapproximately14N/mand320kHz,respectively.
Thecantileverwasexcitedatitsresonancefrequency.
Themajoritycarriertypehasbeencheckedbythehotprobe technique.Ann-typeconstantanwirewasusedasthereference sample.TheelectricalresistivityandHallmobilityofthefilmswere determinedbyHalleffectmeasurementsinavanderPauwconfig- uration.
Photovoltaic characterisation was performed with an auto- mated I–V tester, in the dark and under sun global AM 1.5 simulatedsolarillumination.Performances of photovoltaiccells weremeasuredusingacalibratedsolarsimulator (Oriel300W) at100mW/cm2lightintensityadjustedwithaPVreferencecell (0.5cm2 CIGSsolarcell,calibratedatNREL,USA).Measurements wereperformedatanambientatmosphere.Alldeviceswereillu- minatedthroughTCOelectrodes.
3. Experimentalresults
3.1. Effectoftheoxygenpartialpressureonthepropertiesofthe NiOsputteredthinfilms
Asdescribedinapreviouspaperwitha1.3inchestarget[15], therearethreebasicfeedbacksignalsforcontrollingthereactive processduringsputterdeposition:massspectrometry[16],optical emissionspectroscopy[17]andcathodevoltage(i.e.dischargevolt- age)[18],whichdependsstronglyonthepropertiesofthematerial whichisformedonthecathodesurfaceduringthesputteringpro- cess[19].Witha2inchestarget,wehavefoundthesamebehaviour ofthecathodevoltage(Fig.1)increasingtheoxygencontentwith fixedvaluesofcurrentandflowofinertgas(10sccm,standardcubic centimetreperminute,ofArgon).Thepartialoxygenpressurehas beendeterminedbytheEq.(1).
%O2= O2flux
O2flux+arflux×100 (1)
AsobservedbyPiersonandRousselotforsilveroxidefilms[20], eachregionexhibitsatypicalX-raydiffractionpatternfrom111 to200preferentialorientationthroughtypicalpowderlikeX-ray diffractionpatternforstoichiometricfilms[15]andalsoatypical electronicbehaviour.Between0and2%oxygenpartialpressure,the
a b
Fig.1.(a)Differentzonesasafunctionofoxygencontent,(b)voltagecurvesV=f(%O2)forI=80,110and150mA.
Fig.2.XPSdeterminationoftheFermilevelEFofNiOfilmsdepositedwithadischargecurrentof80mAandapartialoxygenpressureof:(a)of5.66%O2(point1inthe Fig.1);(b)of7.4%O2(point2intheFig.1);(c)of16.67%O2(point3intheFig.1).
filmsareconductive(metallicfilms).Between2and6%oxygenpar- tialpressure,thefilmsaresubstoichiometric,thereisO2−vacancies andtomaintaintheelectroneutrality,thereissomemetallicnickel.
Indeed,amixturebetweenmetallicnickelandNiOiseasilyformed atlowoxygenambients[21].Therefore,metallicnickeldefectsand nickelvacanciescoexistinaNiOfilmgrownatlowoxygenpartial pressureand,asweincreaseoxygencontentinthegasmixture, metallicnickeldefectsdecreaseandnickelvacanciesincrease[22].
Between6and9%partialoxygenpressurethefilmcomposition isclosetothestoichiometry,theconductionisverylow.
Finally,forpartialoxygenpressurehigherthan9%,thenickel vacanciesincreasesandNi2+issubstitutedbyNi3+inordertomain- taintheelectroneutrality.Typically,forNi0.95Otheconcentration ofnickelvacanciesisabout6.75×1020/cm3,whichwouldmakeit ahighlydopedp-typesemiconductor[22].
AsshownonFig.1,thesefourregimesdependonthedischarge current(i.e.:NisputteredspeciesorNi/Oratio).
Theseresultsarecomfortedbythemeasurementofthediffer- encebetweenthevalencebandmaxima(EV)andtheFermilevel (EF)byXPS.ThebindenergyequaltozerocorrespondstotheFermi energywhichisthereferenceforpotentialifweconsiderthatwe haveagoodthermodynamicalcontactbetweenthesampleandthe spectrometer.Actuallytheresultsofthisdifferenceforthezones1, 2and3arepresentedinFig.2.
Forthezone1(5.66%O2),EV−EF=1.1eV,thefilmsarep-type.
Inthezone2(7.4%O2),ononehand,thefilmshaveaverylow conductivity(oneortwoorderofmagnitudeslowerthanforp-type thinfilms),ontheotherhand,EV−EF=3.5eV.Thatisthereason whywethinkthatthefilmscouldben-typeandthebandgaphigher than4eV.Finally,inthezone 3(16.67%O2)EV−EF=0.4eV,the filmsareclearlyp-type.Theseresults,addedtoonespublishedina previouspaper[15],provethatallcharacteristicsofNiOthinfilms
suchastransmittance,conduction,bandgap...,canbeadjusted withplasmaparameters,especiallywiththeoxygencontentinthe films.
AfterthedeterminationofthemajoritycarriersoftheNiOfilms depositedunderdifferenttypicalpartialoxygenpressure,theseNiO filmswerestudiedbyAFMandSEMcharacterisation.Typicalexam- plesoffilmsobtainedinthesputteringconditionsofzones2and3 arevisualisedinFig.3byAFMandbySEMinFig.4.
AbouttheNiOthinfilmscharacterisationbyAFMandbySEM, itmustbehighlightedthatthemorphologyoftheNiOsputtered filmsdependsontheoxygenpartialpressure.Forinstance,when theNiOfilmsisthickof4nm,itspeaktovalleyroughnessis6nm, whenitissputteredwithagascontaining7.4%(zone2,n-type) ofoxygen,whileitismorethandouble,13.5nm,whenthepartial pressureofoxygenis16.67%(Fig.3).Thesehighpeaksdistributed alloverthesurfaceoftheanodestructure,canbecorrelatedtothe darkpointsvisible intheimagesobtainedinthebackscattering modeoftheSEM(Fig.4b).Actually,thedensityof“blackpoints”
visibleinthebackscatteringmode,issimilartothatofthepeaks visualisedintheAFMimages.Inordertoinvestigatetheoriginof theseblackpoints,wehaveproceededtomicroprobemappingof thevisualisedsurfaces.Unfortunately,Ni,InandOmapsarehomo- geneouswithoutanyfeaturessimilartothatvisualisedbytheSEM, maybetheyaretoosmalltobevisualisedbythistechniquewhich resolutionislimitedbytheeffectofthepearshapedinteraction volume.ThepeakvisualisedbytheAFMbeingverysharp,they mayinduceaneffectofblindnesswhentheyarestudiedbySEM inthebackscatteringmode.Indeed,backscatteredelectronsleave thesampleperpendicularlytoitssurface.Therefore,ifthepeaks areverysharp,theywillappear blackinthephotos,thedetec- torbeingunabletodetectelectronparalleltothesurfaceofthe sample.
Fig.3.AFMimagesofNiOfilmsthickof4nmdepositedinthefollowingconditions:(a)partialoxygenpressure7.4%O2;(b)partialoxygenpressure16.67%O2.
Fig.4. MicrophotographsofNiOfilmsthickof4nmdepositedinthefollowingconditions:(a)partialoxygenpressure7.4%O2(p-type,zone1);(b)partialoxygenpressure 16.67%O2(p-type,zone3).
Itisknownthatthesurfaceroughnessoftheelectrodecaninflu- encetheefficiencyoftheorganiccells.Asamatteroffact,arough surfacecaninducealargeinterfacebetweentheEDandtheEA, whichcanimprovetheefficiencyofchargeseparationandthere- foretheCurrentoftheOPVs.Ontheotherhandatoohighroughness ofthebottomelectrodemayinduceleakagecurrentinthedevice, whichdecreasestheshuntresistanceofthediodeandtherefore theperformanceoftheOPVsispoor[23].So,sincethemorphology oftheNiOfilmcanbemanagedthroughthedepositionconditions used,wehavesystematicallyusedNiOfilmsdepositedatdifferent O2partialpressuretoprobetheeffectofthemorphologyofthe bottomlayerontheOPVscharacteristics.Aswecanseeinthenext paragraph,wehavealsostudiedtheinfluenceofthethicknessand thepost-depositionannealingoftheNiOfilmsontheperformances oftheOPVs.
3.2. Organicsolarcellscharacterisation
FirstlywehavestudiedtheinfluenceoftheO2partialpressure ontheJ–VcharacteristicsoftheITO/NiO/CuPc/C60/Alq3/AlOPVs.
TypicalcharacteristicsareshowninFig.5(7%O2,NiOn-type)and Fig.6(16.7%O2,NiOp-type)theNiOfilmbeingthickof4nm.Itcan beseenthatthereisanimprovementoftheOPVsperformances withtimeduringthefirstminutesoflightexposureinthecaseofp- typeNiO.Thereissomethinglikea“formingprocess”foreachcycle,
Fig.5.TheJ–VcharacteristicsofthecellswithaNiOABLdepositedinthefollowing conditions:80mAofcurrent,16.67%O2and4nmofthickness.Thedifferentcycles weremeasuredafteracontinuousexposuretoAM1.5lightsources0min( ), 10min( )and1h30min( ).
Fig.6. TheJ–VcharacteristicsofthecellswithaNiOABLdepositedinthefollowing conditions:80mAofcurrent,7.4%O2and4nmofthickness.Thedifferentcycles weremeasuredafteracontinuousexposuretoAM1.5lightsources0min( ) and30min( ).
whichconsistsinadecreaseoftheleakagecurrentofthediode.In thedark,forthefirstcyclethecurrentisohmic.Underlightthereisa commutationformthisohmicbehaviourtoarectifyingeffect.Then, theshuntresistanceincreaseforeachcycleanditstabilisesafter10 ofexposuretolight(AM1.5),thePCEincreasesfrom1.48%to1.70%
(Fig.5,Table1).Itmustbenotedthatiftheformingprocess,i.e.the increaseoftheshuntresistance,isfasterunderlight,itisalsoactive inthedark.InFig.7onecanseethatthereisaslowtransitionfrom theohmictotherectifyingbehaviourcycleaftercycle,themeasure beingdoneinthedark.Inthesameway,whenthemeasuresare madeunderlight,itcanbeseeninFig.8thatwhenthesamplesare storedinthedarkbetweeneachmeasure,theformationoftheOPVs
Table1
FormingprocessunderlightofOPVswithNiOABLthinfilmsthickof4nm,theOPVs beingsubmittedtocontinuousexposuretolight.
t(min) Voc(V) Jsc(mA/cm2) FF(%) (%) Rs() Rsh() NiO,p-type,4nm(Fig.5)
0 0.460 6.21 55 1.48 3 185
10 0.470 6.16 58 1.70 3 620
300 0.490 5.84 58 1.67 4 755
NiO,n-type,4nm(Fig.6)
0 0.45 6.6 53 1.56 9 500
30 0.43 6.7 54 1.55 13 475
Fig.7.TheJ–VcharacteristicsofthecellswithaNiOABLdepositedinthefollowing conditions:80mAofcurrent,16.67%O2and4nmofthickness.Thesampleissub- mittedtoelectricalcyclesintheobscurity(11times)andthen,1hafter,thesample issubmittedtoanewcycleintheobscurityandunderAM1.5.
Fig.8.TheJ–VcharacteristicsofthecellswithaNiOABLdepositedinthefollowing conditions:80mAofcurrent,16.67%O2and4nmofthickness.Thedifferentcycles weremeasuredafter0min( ),10min( )and90minunderAM1.5.After eachmeasurement,theOPVswereplacedinclosedbox,intheobscurityatroom temperature.
ismoreprogressive,butthestabilisedefficiencyisslightlyhigher thanthatobtainedaftercontinuousillumination(Table2).Inthe caseofn-typeNiOthereisnoformingprocess(Fig.6,Table1),the PCEis1.55%.
Ifformingprocesseshavealreadybeendescribedinorganiclight emittingdiodes(OLEDs)[24]itisnotthecaseforOPVs.Asamater offact,in thecaseof OLEDs,beforethecurrentthreshold volt- age,i.e.thevoltagefromwhichthecurrentincreasessteeply,there arereversiblemicroswitcheffects[25].Thesenegativedifferential resistance(NDR)effectsinducecurrentchangebyseveralordersof magnitude.SuchNDRareusuallyattributedtosmallshortcircuit effectinducedbythesurfaceroughness.Itturnsoutthattheprob- abilityforappearanceofNDRdecreasesmarkedlywithincreasing Table2
Thedifferentcyclesweremeasuredafter0min,10minand1h30minunderAM1.5.
Aftereachmeasurement,thesampleswereplacedinclosedbox,intheobscurityat roomtemperature(Fig.8).
t(min) Voc(V) Jsc(mA/cm2) FF(%) (%) Rs() Rsh()
0 0.25 6.45 32 0.51 3.5 45
10 0.43 6.28 43 1.16 3.5 170
90 0.48 6.03 61 1.75 2.5 930
Fig.9.TheJ–VcharacteristicsofthecellswithaNiOABLdepositedinthefollowing conditions:150mAofcurrent,23.07%O2and20nmofthickness,thedifferentcycles weremeasuredafter0minand30min,6hand24h.
thickness[26].Such areasoningcanapplyhereforourcells.As amatteroffact,whenitisthickofonly4nm,theNiOfilmmay notcompletelycovertheITOfilm.So,thedifferenceofbehaviour ofthecellsusinga4nmthickNiOABLcanberelatedtothedif- ferentmorphologiesofthefilms. Whendepositedunder16.67%
O2 ofoxygen(p-typeNiO) thepeakheightismorethandouble thanthatobtainedwithapartialpressureof7.4ofoxygen(n-type NiO).Therefore,thevacuumdepositionoforganicfilmscouldbe dramaticallyaffectedby“shadoweffect”.Ifthetopoftheanode becomestoorough,throughthepeaksdiscussedabove,itwould beratherdifficulttocovercompletelyuniformlytheanodewith theorganiclayerand,duetodifferentdepositionangle,thecath- odecanbe,inverysmallplaces,indirectcontactwiththeanode, conductivepathscanform,leadingtocurrentleakage.Therefore, theohmiclikeconductionoftheOPVsisduetoultrathinlocalised metalpathsbridgingthetopandbottomelectrodes.Whensubmit- tedtoapotential,i.e.whenacurrentcirculatesintheseultrathin filamentstheyaredestroyedbyJouleeffect.Ofcoursethiseffect isfasterwhenthecurrentishigher,whichisthecasewhenthe OPVsaresubmittedtolight.Moreovertheformingprocessisactive whenNiOisp-type,thatistosaywhenitismoreconductive.Itwas shown,inthecaseofWO3,anothertransitionmetaloxide,that, bychangingtheadditionaloxygencontentoftheoxide,switching insulating-conductivestatesmaybemanaged.Thecommutation isattributedtoconductivepathformation[27].Whenitisn-type, NiOissmoothanditishighlyresistiveandthereforethereisnot formingprocess.
IthasalreadybeenshownthattheNiOABLimprovesstrongly thelifetimeoftheOPVs[12],whatmakespossibletheforming processandtheimprovementwithtime,duringthefirstminutes ofairexposure,ofthePCEoftheOPVs.
ItisknownthatbymodifyingthethicknessoftheNiOABLispos- sibletomodifytheperformancesoftheOPVs.Therefore,inorder toavoidanyformingprocesswehaveincreasedtheNiOthickness.
WhentheNiOanodebufferlayeristhickof20nmtheJ–Vchar- acteristicsarestableasitcanbeseeninFigs.9and10.Thereis nomoreanyformingprocesswhateverthetypeoftheNiO,p-type orn-type(Table3).Thecurvesexhibitsystematicallyaclassical shape,whatevertheO2partialpressureduringNiOdeposition.As amatteroffact,whentheNiOfilmis20nmthickitrecoverscom- pletelytheITOfilm,whichavoidsanydirectcontactbetweenthe anodeandthecathodeevenif“shadoweffects”arepresent,which
Fig.10.TheJ–VcharacteristicsofthecellswithaNiOABLdepositedinthefollowing conditions:80mAofcurrent,7.4%O2and20nmofthickness.
Table3
OPVsperformanceswithNiOABLdepositedinthefollowingconditions:150mAof current;23.07%O2and20nmofthickness,thedifferentcyclesweremeasuredafter 0min,30min,6hand24h.
t(min) Voc(V) Jsc(mA/cm2) FF(%) (%) Rs() Rsh() NiO,p-type,20nm(Fig.9)
0 0.46 4.19 59 1.15 2.5 1200
30 0.46 4.43 59 1.20 2.5 1200
360 0.47 4.17 59 1.16 3 1150
1840 0.46 3.81 55 0.97 3.5 900
NiO,n-type,20nm(Fig.10)
0 0.42 5.5 56 1.5 29 1600
explainsthataformingprocessisnotnecessarytoachieveclassical behaviour.
Therefore,inordertochecktheeffectofannealingoftheNiO ABLonthePCEoftheOPVs,NiOfilmsthickof20nmhavebeen used.Beforedepositionoftheorganiclayers,theITO/NiOstructures havebeensubmittedtoannealingunderoxygenatmospherefor 10mininatubularoven.ItcanbeseeninFig.11that,upto400◦C theannealingimprovesthePCEoftheOPVs,forhigherannealing temperatureitdecreasesprogressively(Table4).Thisimprovement isduetotheincreaseofthetransmittance[28].
Forhigherannealingtemperature,theITObottomlayercon- ductivitydecreases,duetothedecreaseofoxygenvacanciesand thereforetheseriesresistanceincreasestothedetrimentofthe PCE.
Fig.11.TheJ–VcharacteristicsofthecellswithaNiOABLdepositedinthefollowing conditions:80mAofcurrent,7.4%O2and20nmofthickness,withoutannealingand after10minofannealingat300◦C,400◦C,500◦Cand600◦C.
Table4
AnnealingeffectontheperformancesofOPVswithNiOABLdepositedinthefol- lowingconditions:80mAofcurrent,7.4%O2and20nmofthickness(Fig.11).
Annealing temperature(◦C)
Voc(V) Jsc(mA/cm2) FF(%) (%) Rs() Rsh()
– 0.47 5.48 56.3 1.45 3 1600
300 0.49 5.96 55.5 1.62 3.5 500
400 0.50 7.71 45.1 1.75 21 540
500 0.51 6.75 46 1.60 47.5 710
600 0.51 6.24 35 1.11 255 560
AboutthethicknessoftheNiOABL,ithasalreadybeenshown, inthecaseofP3HT:PCBMbulkheterojunctionthatwhentheNiO filmisthickof4–8nmtheOPVshavepoorFF,whichisattributed tolargeleakagecurrentanddeficientelectronblockingcapability [11].Thisresultcorroboratesthepresentstudy,whichshowsthat reproducibleandoptimumresultsareachievedwith20nmthick NiOABL.
4. Conclusions
ByvaryingthepartialoxygenpressureduringNiOsputtering depositionitispossibletomanagethepropertiesoftheNiOthin films.Firstmetallicfilmsareobtainedbetween0and2%partial oxygenpressure, then theNiO filmshave a p-type conduction between 2 and 6% partial oxygenpressure, between 6 and 9%
partialoxygenpressuretheirconductionisverylowandn-type.
Finally, for a partialoxygen pressure higher than 9%, the con- ductionoftheNiOfilmsisp-type.Whenthesefilmsareusedas ABLinITO/NiO/CuPc/C60/BCP/AlcellsthepropertiesoftheseOPVs dependontheO2partialpressureduringNiOdepositionandon thethicknessofthisABL.Whenthickof4nm,aformingprocess ispresentduringthefirstelectricalcycles.Thisformingprocess dependsontheNiOsputteringconditions.Whenthefilmsaren- type,O2partialpressure7%,theformingprocessismarginaldue tothefactthattheNiOfilmsareresistiveandsmooth.Whenthe NiOfilmsarep-type,O2partialpressure16.7%,theJ–Vcharacter- isticsvarystronglyduringthefirstcycles,whichisattributedto thedestructionofleakagefilamentpresentintheOPVs.Asamat- teroffact,whendepositedintheseexperimentalconditionsthere isahighdensityofthinpeaksatthesurfaceoftheNiOfilmsand thepresenceofthesepeaksmayinducesultrathinleakagefila- mentsthroughsomeshadowingeffects.Theconductivefilaments aredestroyedbyJouleeffectduringtheformingprocess.Inorderto avoidtheformingprocess,NiOthinfilmshavebeenusedasABL.In thatcaseoptimumefficiencyisachievedwhentheNiOisannealed at400◦C.
References
[1]A.Godoy,L.Cattin,L.Toumi,F.R.Diaz,M.A.delValle,G.M.Soto,B.Kouskoussa, M.Morsli,K.Benchouk,A.Khelil,J.C.Bernède,Sol.EnergyMater.Sol.Cells94 (2010)648.
[2]M.Cai,T.Xiao,E.Hellerich,Y.Chen,R.Shinar,J.Shinar,Adv.Mater.23(2011) 3590.
[3]S.-Y.Park,H.-R.Kim,H.-J.Kang,D.-H.Kim,J.-W.Kang,Sol.EnergyMater.Sol.
Cells94(2010)2332–2336.
[4]J.C.Bernède,L.Cattin,M.Morsli,Y.Berredjem,Sol.EnergyMater.Sol.Cells92 (2008)1508–1515.
[5]F.Zhang,F.Sun,Y.Shi,Z.Zhuo,L.Lu,D.Zhao,Z.Xu,Y.Wang,EnergyFuels24 (2010)3739.
[6]L.Cattin,F.Dahou,Y.Lare,M.Morsli,R.Tricot,S.Houari,A.Mokrani,K.Jondo, A.Khelil,K.Napo,J.C.Bernede,J.Appl.Phys.105(2009)034507.
[7]M.S.Ryu,J.Jang,Sol.EnergyMater.Sol.Cells95(2011)3015.
[8]I.Hancox,L.A.Rochefort,D.Clare,P.Sullivan,T.S.Jones,Appl.Phys.Lett.99 (2011)013304.
[9]S.-Y.Park,H.-R.Kim,Y.-J.Kang,D.-H.Kim,J.-W.Kang,Sol.EnergyMater.Sol.
Cells94(2010)2332.
[10]K.X.Steirer,J.P.Chesin,N.E.Widjonarko,J.J.Berry,A.Miedaner,D.S.Ginley,D.C.
Olson,Org.Electron11(2010)1414–1418.
[11]N.Sun,G.Fang,P.Qin,Q.Zheng,M.Wang,X.Fan,F.Cheng,J.Wan,X.Zhao,Sol.
Energ.Mater.Sol.Cells94(2010)2328.
[12]R.Betancur,M.Maymo,X.Elias,L.T.Vuong,J.Martorell,Sol.Energ.Mater.Sol.
Cells95(2011)735.
[13]M.D.Irwin,J.D.Servaites,D.B.Buchholz,B.J.Leever,J.Liu,J.D.Emery,M.Zhang, J.-H.Song,M.F.Durstock,A.J.Freeman,M.J.Bedzyk,M.C.Hersam,R.P.H.Chang, M.A.Ratner,T.J.Marks,Chem.Mater.23(2011)2218.
[14]J.C.Bernède,S.Houari,D.T.Nguyen,P.Y.Jouan,A.Khelil,A.Mokrani,L.Cattin, P.Predeep,Phys.Stat.Sol.209(7)(2012)1291–1297.
[15]A.Karpinski,A.Ferrec,M.Richard-Plouet,L.Cattin,M.A.Djouadi,L.Brohan, P.-Y.Jouan,ThinSolidFilms520(2012)3609–3613.
[16]V.Vancoppenolle,P.-Y.Jouan,M.Wautelet,J.-P.Dauchot,M.Hecq,J.Vac.Sci.
Technol.A17(1999)3317.
[17]V.Vancoppenolle,P.-Y.Jouan,A.Ricard,M.Wautelet,J.-P.Dauchot,M.Hecq, Appl.Surf.Sci.205(2003)249.
[18]W.D.Sproul,D.J.Christie,D.C.Carter,ThinSolidFilms491(2005)1.
[19]D.Depla,S.Mahieu,R.DeGryse,ThinSolidFilms517(2009)2825.
[20]J.F.Pierson,C.Rousselot,Surf.Coat.Technol.200(2005)276.
[21]S.Stanescu,C.Boeglin,A.Barbier,J.-P.Deville,Phys.Rev.B(2003),67,035419.
[22]S.Seo,M.J.Lee,D.H.Seo,E.J.Jeoung,D.-S.Suh,Y.S.Joung,I.K.Yoo,I.R.Hwang, S.H.Kim,I.S.Byun,J.-S.Kim,J.S.Choi,B.H.Park,Appl.Phys.Lett.85(23)(2004).
[23]F.Z.Dahou,L.Cattin,J.Garnier,J.Ouerfelli,M.Morsli,G.Louarn,A.Bouteville, A.Khellil,J.C.Bernède,ThinSolidFilms518(2010)6117.
[24]J.C.Bernède,V.Jousseaume,M.A.delValle,F.R.Diaz,Curr.TrendsPolym.Sci.6 (2001)135.
[25]F.Brovelli,F.R.Diaz,M.A.DelValle,J.C.Bernède,P.Molinié,Synth.Met.122 (2000)123.
[26]S.Berleb,W.Brütting,M.Schowoever,Synth.Met.102(1999)1034.
[27]B.U.Jang,A.I.Inamdar,J.Kim,W.Jung,H.Im,H.Kim,J.P.Hong,ThinSolidFilms 520(2012)5451.
[28]LeiAi,GuojiaFang,LongyanYuan,NishuangLiu,MingjunWang,ChunLi,Qilin Zhang,JunLi,XingzhongZhao,Appl.Surf.Sci.254(2008)2401.